DEV Community: Alex Isa

How to audit a free AI visibility score with six manual checks

Alex Isa — Tue, 11 Aug 2026 09:36:42 +0000

A free AI visibility score is auditable only when you can inspect the prompt, engine, raw answer, date, and denominator. Treat the score as a test result, not a property of your brand. This tutorial builds a six-check control you can run by hand, store as plain data, and compare with any tool's output.

The workflow takes three buyer questions, runs them in two AI surfaces, and records the six answers without trying to force agreement. It will not estimate your entire market. It will tell you whether a dashboard's headline number has enough evidence to be investigated.

What does an AI visibility score measure?

An AI visibility score usually summarizes brand presence across a defined set of generated answers.

That definition contains the trap: the question set is part of the metric. So are the engine panel, run date, session state, retrieval mode, and rule used to count a “hit.” Remove those inputs and the number is not reproducible.

Imagine a tool asks three questions in two engines. That creates six cells. If your brand appears in two cells, the simple presence result is:

presence = brand_present_cells / total_cells
presence = 2 / 6
presence = 0.333... = 33.3%

The arithmetic is trivial. The evidence is not. A different tool can ask five different questions in three engines and produce a different score without contradicting the first run. The two tools measured different grids.

Keep the unit explicit: “present in two of six generated answers on this date” is defensible. “Our AI visibility is 33” is incomplete.

Which evidence fields should you require?

Require five fields for every result: prompt, engine, raw answer, timestamp, and counting rule.

Use a sixth field for cited sources when the surface exposes them. A source-only appearance and a prose mention can signal different problems, so do not merge them silently.

Here is one real saved result from Webappski's public 14 June 2026 tracker report:

{
  "run_date": "2026-06-14",
  "prompt": "beste Answer Engine Optimization Agentur für Sichtbarkeit in ChatGPT",
  "surface": "Claude (claude-sonnet-4-6)",
  "brand": "Webappski",
  "saved_result": "named; positive; no rank",
  "prompt_denominator": "1 of 3 surfaces named Webappski",
  "run_denominator": "2 of 39 answer cells named Webappski",
  "public_receipt": "https://webappski.com/reports/aeo-webappski-2026-06-14.html"
}

The same Q7 row records no Webappski mention from ChatGPT or Gemini and one named result from Claude. The public report preserves the exact prompt, surfaces, result labels, model names, date, and denominator; it does not expose the full Claude prose, so this is explicitly a saved result rather than a raw-answer quotation.

The counting rule deserves its own field because products make different choices. Some count only explicit brand mentions. Some count cited domains. Some weight rank, sentiment, or recommendation strength. A score can be internally consistent while answering a different question from yours.

How do you choose three control prompts?

Choose one category prompt, one alternatives prompt, and one situation prompt in the buyer's language.

Write questions a buyer could ask without knowing your brand. Naming the brand contaminates the test because it turns discovery into recall.

Use this template:

Category: What are the best [category] tools for [buyer type]?
Alternatives: What are good alternatives to [known category option] for [constraint]?
Situation: What should a [buyer type] use to achieve [specific outcome]?

Make each question commercially coherent. “What is software?” may produce a stable answer but tells you nothing about vendor discovery. “Best tool” without a buyer or constraint can be so broad that the answer grid is mostly noise.

Freeze the strings in a file before running them:

{
  "version": "control-v1",
  "language": "en",
  "prompts": [
    "What are the best [category] tools for [buyer type]?",
    "What are good alternatives to [known option] for [constraint]?",
    "What should a [buyer type] use to achieve [specific outcome]?"
  ]
}

If you edit a prompt later, increment the version and start a new baseline. A chart that changes its questions without marking the break is not a controlled comparison.

How do you run the six checks?

Run each frozen prompt once in each of two relevant AI surfaces and preserve the complete response.

Use fresh sessions without prior brand conversation. If the product offers memory or personalization controls, disable them where practical and record what you did. The goal is not to simulate every buyer; it is to remove obvious contamination from your own history.

Create the grid before opening either surface. This compact row shows the real Q7 receipt before the blank worksheet:

Date	Exact prompt	Named surface/result	Prompt denominator	Full-run denominator
2026-06-14	`beste Answer Engine Optimization Agentur für Sichtbarkeit in ChatGPT`	Claude: Webappski named, positive, no rank	1 / 3 surfaces	2 / 39 cells

Paste the questions exactly. Do not ask follow-ups, request citations after the fact, or regenerate until you get a preferred answer. Those actions create new observations and should be stored as separate cells if you need them.

Save the full text, not a screenshot alone. Text is searchable and diffable. A screenshot can supplement it when the interface shows source cards or mode labels that copy-and-paste omits.

Record the surface precisely enough to understand later. “ChatGPT” or “Gemini” may still be incomplete, but it is better than “AI.” If the interface names a mode or model, store that label without guessing what sits behind it.

How do you calculate the control result?

Calculate prose presence and citation presence separately before creating any combined score.

Suppose the six cells contain two prose mentions and one additional citation-only appearance:

prose_presence = 2 / 6 = 33.3%
citation_presence = 3 / 6 = 50.0%
citation_only = 1 / 6 = 16.7%

If you want a single combined presence number, define it explicitly:

combined_presence = cells(brand_in_answer OR brand_in_citations) / 6
combined_presence = 3 / 6 = 50.0%

Do not describe 50% as recommendation quality. One cited URL may be a source for a criticism. A prose mention may appear low in a long alternatives list. Presence is useful because it is easy to audit, not because it captures every business meaning.

Add a compact context label instead of inventing a complex score:

recommended | listed | mentioned | cited-only | absent | misdescribed

The label forces a human to read the answer. That is a feature. A small control run should expose meaning, not hide it behind precision.

How do you compare a tool with your control?

Compare experiment design first and headline scores second.

Ask whether the tool exposes the literal prompts, named surfaces, response text, date, sample size, and counting rule. If it does, map its result to your six-cell sheet. If it does not, treat the score as a hypothesis about a hidden experiment.

Current public pages illustrate why feature inspection matters. Semrush states that its free AI Search Visibility Checker requires no sign-up and includes prompts, mentions, and citations as report fields. Ahrefs offers a free Brand Radar preview and documents its own mention metric as one response in which a brand appears at least once. Those are vendor statements, not our independent accuracy tests, but they give you concrete fields to look for.

Use this evaluation table:

Question	Pass condition	Why it matters
Can you see the exact prompts?	Literal strings, not topic labels	Makes reruns possible
Are engines separated?	Per-surface rows remain visible	Prevents a blended score hiding the weak surface
Is answer text available?	Full or clearly scoped raw text	Distinguishes recommendation from incidental mention
Is the denominator visible?	Prompt count × engine count	Makes percentages interpretable
Is the date visible?	Per run or per response	Prevents timeless snapshots
Is the metric defined?	Mention/citation/weighting rule	Prevents comparing different units

A tool does not need to expose everything for free to be useful. It does need to tell you what the free number represents. “Upgrade to see evidence” is a commercial gate; “the evidence does not exist” is a measurement problem.

What result differences are acceptable?

Different results are acceptable when the experiments differ and the differences are documented.

Your manual run is unlikely to match a vendor database exactly. The tool may use a larger prompt set, other locations, scheduled captures, APIs, browser automation, different models, or a historical index. Agreement is not the test.

Instead, ask whether the difference can be explained. If your brand appears in all six manual cells but the tool reports zero, inspect its prompts, category assignment, entity aliases, and capture date. If your control shows zero and the tool reports high visibility, inspect whether the tool includes branded prompts or a wider category.

Three outcomes are useful:

Same direction: both methods show strong or weak presence despite different scales.
Explainable difference: prompts, engines, dates, or counting rules account for the gap.
Unexplained difference: the tool withholds enough evidence that you cannot diagnose the gap.

The third outcome does not prove the tool is wrong. It tells you not to use its number as the sole basis for work or budget.

How do you repeat the test without lying to yourself?

Repeat the same prompt version and engine panel after a declared waiting period.

Store each run in a dated directory:

ai-visibility-control/
  control-v1.json
  2026-07-31/
    engine-a-category.txt
    engine-a-alternatives.txt
    engine-a-situation.txt
    engine-b-category.txt
    engine-b-alternatives.txt
    engine-b-situation.txt
    summary.json

Keep an intervention log beside the runs. Note the exact URL or external surface changed, the date, and the prompt cluster it was intended to affect. Without the log, a changed result cannot be connected to the work.

Do not rerun repeatedly until the score rises. Predeclare the cadence. If an engine or model label changes, record the break. If you change the prompts, start control-v2 instead of extending the old series.

This does not eliminate generative variance. It prevents you from manufacturing a trend through flexible inputs.

What should you do before trusting the score?

Trust the evidence chain before trusting the arithmetic.

You should be able to move from percentage to denominator, from denominator to cells, from cells to raw answers, and from raw answers to the exact prompts and surfaces. If any link is missing, state the limitation in the decision that uses the number.

The six-check control is deliberately small. It can reveal obvious disagreement, hidden branded prompts, blended surfaces, source-only mentions, and unstable definitions. It cannot estimate every buyer, location, language, personalization state, or future answer.

That boundary is the conclusion, not a defect. A useful visibility score is a dated summary of inspectable answers. A dangerous one presents itself as a stable property of the brand.

How do you validate your own parser?

Validate the parser with adversarial answer fixtures before automating the control.

A substring search for the brand is not enough. It can match a word that merely contains the brand token, miss a spaced or punctuated variant, or treat a cited URL as a prose mention. Keep extraction separate from scoring so you can inspect both.

Start with a tiny fixture set:

[
  {"answer": "EXAMPLE_BRAND is one option.", "expected": "mentioned"},
  {"answer": "No suitable vendors were found.", "expected": "absent"},
  {"answer": "See the cited source for details.", "citations": ["https://example-brand.test"], "expected": "cited-only"},
  {"answer": "Example branding principles apply.", "expected": "absent"},
  {"answer": "Example—Brand may fit small teams.", "expected": "mentioned"}
]

Normalize case and common punctuation, but do not rewrite the evidence. Store the original text beside the extracted event. If aliases are allowed, version the alias list and include it in the run metadata.

Test the scoring reducer independently with fixed events:

events = [mentioned, absent, cited-only, absent, mentioned, absent]
prose mentions = 2
any presence = 3
denominator = 6

The expected outputs are 33.3% prose presence and 50.0% combined presence. If a later code change produces a different result, the test should fail before historical reports are silently recalculated.

Avoid using another language model as the only extractor. It can help classify context, but literal brand presence should remain mechanically checkable. If two classifiers disagree about recommendation strength, preserve the disagreement rather than averaging it into certainty.

How can you automate without losing the receipt?

Automation should create more inspectable evidence, not only a faster score.

Keep manual and automated runs visibly separate. A browser session, an official API call, and a pasted consumer answer may all be useful, but they are not interchangeable capture methods. Store the method per cell and avoid merging histories unless the report shows the break. If an automated surface applies hidden defaults, record every configuration value you can control and label the rest unknown.

Add a dry-run mode that prints prompts, engines, expected cell count, and output paths without sending requests. This catches a changed prompt file or missing engine before it creates a partial baseline. After collection, reconcile expected and actual cells mechanically and block summary generation when the declared completeness rule fails.

Write the raw response before parsing it. Use append-only run directories or immutable object keys so a regenerated summary cannot overwrite its inputs. Hashing the raw file is optional, but a stable path and timestamp are essential.

Keep run metadata beside the response:

{
  "run_id": "2026-07-31-control-v1",
  "prompt_version": "control-v1",
  "engine_surface": "ENGINE_A",
  "session_mode": "fresh",
  "collection_method": "manual",
  "scoring_version": "presence-v1"
}

Generate summaries from the stored events, never the other way around. A dashboard should link each cell back to its response. An export should include prompts and definitions, not just percentages.

Handle failures as missing cells. A timeout is not an absent brand. If only five of six answers were collected, report a five-cell denominator or mark the run incomplete according to a predeclared rule. Never fill a failed cell with zero merely to keep the table rectangular.

Make the completeness rule part of the summary contract. For a small control, requiring all six cells is usually clearer than publishing a partial percentage. For a larger production run, you may accept a declared threshold, but the report should show collected cells, expected cells, and failure reasons together. The denominator used for scoring must be obvious.

Retain failed request metadata without storing secrets. Engine label, prompt identifier, timestamp, error class, and retry count are enough to distinguish a collection outage from a parser failure. Never copy API keys, authorization headers, or private session data into the evidence directory.

If you retry, link the replacement cell to the failed attempt and keep the original timestamp. A retry is a new observation, not a way to erase an inconvenient response. Predeclare whether the first successful response or every successful response enters the score.

Finally, make changes explicit. Version prompts, aliases, parsers, and scoring rules. The more automation you add, the easier it becomes to create a clean trend that quietly compares different instruments.

The live complete Webappski guide includes the broader free-tool survey and the control method behind this tutorial.

By Alex Isa for Webappski. Tool feature statements were rechecked against the vendors’ public pages on 31 July 2026; no vendor accuracy claim is implied.

We Asked 3 AI Answer Systems Which AEO Agencies They Cite. Here Is the Reproducible Method

Alex Isa — Tue, 11 Aug 2026 08:23:50 +0000

Most “best agency” pages begin with an editorial opinion. We began with a denominator.

Minuttia appeared in four of six answer cells in our dated audit. Webappski appeared in zero. That does not prove one agency is universally better. It shows which names recurred when three answer systems handled the same two commercial questions on 31 July 2026.

This article explains the method in a form another team can inspect, criticize, and repeat. It also explains where the method stops being evidence.

What exactly did we measure?

We measured recommendation recurrence across six engine-query cells.

The two buyer questions were:

best Answer Engine Optimization agencies 2026
Answer Engine Optimization consultants for B2B startups

Each question was checked on three surfaces:

OpenAI Responses API using gpt-5.4-mini;
Gemini grounded API using gemini-3.6-flash;
a manual Claude web-search check, because Claude does not offer a public grounded-search API equivalent for this test.

Two questions multiplied by three answer systems produced six cells. An agency received one appearance for a cell if the answer named it. Its position inside that answer did not create extra points.

{
  "date": "2026-07-31",
  "queries": 2,
  "answer_systems": 3,
  "cells": 6,
  "counting_rule": "one appearance per named provider per cell"
}

This is deliberately narrower than a visibility score built from hundreds of prompts. The advantage is inspectability: the denominator fits on one screen.

How did an agency enter the shortlist?

An agency needed recurrence plus a live first-party service surface.

The main shortlist required both of the following:

the provider appeared in at least two of the six cells;
its own live website supported an AEO, GEO, or broader AI-search service.

We then broke ties in a fixed order:

appearance across more distinct answer systems;
appearance on the B2B-startup question;
clearer live service scope or public pricing;
alphabetical order.

Webappski was added outside the eligibility rule as the disclosed publisher benchmark. It received no editorial bonus and was placed last because it appeared in zero cells.

That disclosure matters. A self-published ranking becomes hard to trust when the publisher quietly gives itself first place. The cleaner approach is to define the rule before reading the flattering result—or the uncomfortable one.

What did the six-cell audit return?

Minuttia led the snapshot with four appearances; First Page Sage followed with three.

Six other providers completed the published comparison. The recurrence column below is observational. The service-fit column is based on each provider’s own live website and should be treated as first-party evidence, not an independent performance result.

Rank	Agency	Cells named	Public service fit	Buyer to investigate the fit
1	Minuttia	4 of 6	Google and AI search, content, digital PR, analytics	Established B2B SaaS or technology team
2	First Page Sage	3 of 6	Research-led AEO and thought-leadership content	Established B2B organization
3	Discovered Labs	2 of 6 across 2 systems	B2B SaaS AEO measurement and execution	SaaS growth team
4	iPullRank	2 of 6 across 2 systems	Relevance engineering and dedicated GEO	Enterprise or mid-market team
5	Siege Media	2 of 6 across 2 systems	Content-and-PR-led GEO and data journalism	Content-mature brand
6	Omniscient Digital	2 of 6	Integrated GEO, content, and technical work	Growth-stage or enterprise B2B team
7	Rock The Rankings	2 of 6	B2B SaaS SEO and GEO tied to pipeline	Pipeline-focused SaaS team
8	Webappski	0 of 6	Per-engine audit and implementation	Smaller product or service business

A cell count is not a quality score. It says only that a provider surfaced for these exact prompts, on these exact systems, on this date.

Why do the cited URLs matter as much as the names?

The cited URLs reveal the source pool that shaped each recommendation.

The OpenAI broad-agency answer leaned heavily on Clutch pages, while its B2B-startup answer used provider sites. Gemini mixed publishers, vendors, and agency domains. Claude’s broad answer leaned on listicles; its startup answer mixed service pages with agency-authored comparisons.

That creates an important distinction:

recommendation evidence: the answer system named an agency;
source evidence: the answer cited a particular page;
service evidence: the agency’s own site supports the described offer;
outcome evidence: an independently verifiable client result exists.

The first three can be checked in this type of audit. The fourth cannot be inferred from them.

What can this method prove?

This method can prove dated recurrence within a declared sample.

It can answer questions such as:

Which names appeared more than once?
Did a provider recur across multiple answer systems or only one?
Did the provider appear on the broad query, the startup query, or both?
Which URLs were cited alongside the recommendation?
Does the provider’s own live site support the service description?

It cannot prove that the first-ranked agency will produce the best result for a client. It cannot convert six observations into a permanent market ranking. It cannot show what a logged-in consumer interface will return for every user, location, or session.

Why will a rerun produce different answers?

Answer systems are non-deterministic, so an exact rerun may change individual cells.

Model versions, retrieval indexes, geography, wording, and time can all alter the result. That is not a reason to avoid measurement. It is a reason to save the complete measurement definition:

exact question text;
answer system and model;
date;
full answer;
named providers;
cited URLs;
counting rule;
denominator.

Without those fields, a percentage cannot be independently checked. With them, a later run becomes a comparison rather than a fresh anecdote.

How should a buyer use the shortlist?

Use the shortlist to decide whom to investigate, not whom to hire automatically.

For every candidate, ask:

Which exact questions will you monitor for our business?
Which answer systems will you report separately?
Will we receive the full answers and cited URLs?
What is the denominator behind the reported percentage?
Which work is specific to AI citations rather than standard SEO?
Can the same query set be repeated without quietly changing the baseline?

The strongest proposal should make its measurement falsifiable. “AI visibility improved” is not enough. “The brand appeared in 9 of 30 declared cells, up from 3 of 30, with the raw answers attached” is at least inspectable.

What was the uncomfortable result for Webappski?

Webappski appeared in none of the six measured cells.

We publish the comparison and sell AEO services, so hiding that result would undermine the entire method. Our own website states the service scope and prices; those are offer facts. They are not evidence that answer systems currently recommend us for these two English buyer questions.

The zero is more useful than a flattering self-rank. It establishes a real baseline and makes the next distribution experiment measurable.

What should the next run test?

The next run should repeat the same two questions after the canonical comparison has been discoverable.

The test is not “did Webappski become number one?” The test is whether the new page enters the citation pool or changes any named-provider cell while the denominator stays fixed. No movement is also a result.

The full comparison on Webappski includes the provider-by-provider evidence, source links, limitations, and buyer fit notes.

Originally published on Webappski.

Best Answer Engine Optimization Agencies in Poland and DACH (2026 Guide)

Alex Isa — Tue, 11 Aug 2026 07:41:23 +0000

AI assistants now answer buyers in Polish and German before those buyers ever open Google. This guide compares the Answer Engine Optimization agencies serving Poland and the DACH region in 2026 — Delante, Elephate, SearchGPT Agentur, PerformanceLiebe, Digitalagentur38, and Webappski — on what they actually do, where they are based, and how they prove AI-search results.

In 2026, the leading Answer Engine Optimization agencies for Poland and the DACH region are Delante and Webappski in Poland and SearchGPT Agentur, PerformanceLiebe, and Digitalagentur38 in Germany. Delante adds AI search to enterprise SEO; the three German agencies are AEO-native; and Webappski optimizes each AI engine separately with measured, non-English citation data.

Answer Engine Optimization (AEO) for these markets means making a company appear inside AI answers — from ChatGPT, Perplexity, Gemini, and Claude — when buyers ask questions in Polish or German, not only in English. The specialist field is small, and the six agencies below serve it in different ways. Elephate, a strong Polish SEO and PR agency, is included because buyers compare it even though it states no dedicated AEO service. The right choice depends on your language market and whether you need measured, per-engine proof.

Most published AEO agency rankings list United States firms and assume an English-speaking buyer. That assumption breaks in Poland and the DACH region. When a buyer asks Gemini "agencja Answer Engine Optimization dla firm" or asks ChatGPT "Answer Engine Optimization Agentur Deutschland," the engine answers in the buyer's language and cites local sources. A US-centric strategy does not surface there. This guide is written for companies whose customers research in Polish or German — and it names the agencies that actually compete for those answers. Webappski is one of them, and we disclose our own listing openly below.

AEO is not an evolution of SEO. Search engines rank pages; answer engines synthesize a single response and decide which sources to cite. A page that ranks well on Google may never be quoted by ChatGPT, and a brand that Perplexity recommends may have a thin Google footprint. For a non-English market the gap is wider still, because each engine weighs language, regional sources, and structured data differently. If you are evaluating providers in Poland or the DACH region, that distinction is the first thing to test them on.

What Is Answer Engine Optimization (AEO)?

Answer Engine Optimization is the practice of structuring your content, technical infrastructure, and online presence so AI answer engines — ChatGPT, Perplexity, Gemini, Claude, Microsoft Copilot — cite and recommend your company when users ask relevant questions. Unlike traditional SEO, which optimizes for Google's ranking of blue links, AEO optimizes for AI synthesis: the process by which a large language model decides which brands and sources to name in a generated answer.

The engines do not work the same way, and that is the core of the discipline. ChatGPT retrieves through Bing's index. Perplexity crawls the web directly. Gemini reads Schema.org structured data from Google's index. Claude leans on its training data, drawn from sources like npm, GitHub, Wikipedia, and StackOverflow. A change that earns a citation in one engine can be invisible to another. For a deeper explanation, see our full guide: What Is AEO? Answer Engine Optimization Explained.

Why AEO Matters Differently in Poland and the DACH Region

In English-language markets, AI answer engines are saturated with US sources, and the competition for a citation is fierce. In Polish and German the answer space is far less crowded — but it is also less understood. When we measured how AI engines respond to localized queries such as "best Answer Engine Optimization agency Poland 2026" and "Answer Engine Optimization Agentur Deutschland," the engines answered confidently in the local language, yet the cited sources were a thin and shifting mix of agency pages, regional SEO rankings, and syndicated press releases. That is the opportunity: the buyer is already asking, and the field of trusted sources is still forming.

The query already happens in the local language. Buyers in Wroclaw or Hamburg ask the assistant in Polish or German, and the engine answers in kind — an English-only presence does not appear.
Local source pools are thinner. For non-English AEO queries, engines often fall back on a handful of agency sites and regional rankings, which means a well-structured local source can earn a citation faster than in the crowded English space.
Each engine localizes differently. ChatGPT (via Bing) and Gemini (via Google's index) surface different Polish and German sources than Perplexity's direct crawl. One unified strategy under-serves at least one engine.
Structured data carries more weight when text is sparse. Schema.org markup, FAQ structure, and clean entity definitions help an engine trust a source it has fewer corroborating signals for — common in smaller language markets.

The Poland and DACH AEO Agencies Compared

Below is a factual comparison of six agencies that companies in Poland and the DACH region encounter when they research Answer Engine Optimization in 2026. We evaluated each on what they publicly state they do, where they are based, how they price, and how they prove AI-search results. Where an agency does not publish a figure, we say so rather than guess. All facts below were taken from each agency's own website in May 2026; capabilities and pricing in this field change quickly, so verify current details before signing.

1. Delante (Krakow, Poland) — SEO Agency With an AI Search Offering

Delante is an international search marketing agency headquartered in Krakow, with additional offices listed in New York, London, and the United Arab Emirates. Alongside SEO and SEM, they offer AI Search Optimization (AISO) — an AI audit, AI search training, and AI visibility monitoring aimed at improving how a brand appears in AI-generated answers on ChatGPT, Gemini, and Perplexity (delante.co). They are the most internationally established Polish agency with a stated AI-search practice.

What they do: SEO, SEM, and AI Search Optimization (AI audit, AI search training, AI visibility monitoring).
Pricing model: Not published as packages; their contact form shows monthly budget tiers of €1,500–2,500, €2,500–5,000, and over €5,000 (delante.co).
Best for: Polish and international e-commerce or enterprise brands that want AI search added to an established, large SEO engagement.
Weakness: AI search is one service inside a broad SEO agency rather than a dedicated, per-engine AEO practice.

2. Elephate (Poland) — eCommerce SEO, No Dedicated AEO

Elephate is a Poland-based agency specializing in eCommerce SEO, content marketing, and digital PR, working with platforms such as Shopify, Magento, and BigCommerce (elephate.com). AI answer engines sometimes name Elephate in Polish AEO and SEO answers because of their visibility in the broader search-marketing space. However, their own website does not describe a dedicated Answer Engine Optimization service. We include them because buyers comparing Polish agencies will encounter the name — and should know the distinction.

What they do: eCommerce SEO, content marketing, link building, and PR (elephate.com).
Pricing model: Not published.
Best for: eCommerce brands focused primarily on Google organic and PR rather than AI-answer citations.
Weakness: No dedicated AEO offering stated on their site; AI-search work would be inferred from general SEO strength, not a named practice.

3. SearchGPT Agentur (Berlin, Germany) — AEO-Native Agency

SearchGPT Agentur is based in Berlin (with a secondary location in Munich) and positions itself as a dedicated Answer Engine Optimization and AI-search agency for the German market (searchgptagentur.de). Their stated services include website analysis, content optimization, custom AI plugin development, and continuous monitoring of AI-search visibility, with a free GEO report offered as an entry point. They are one of the most clearly AEO-first agencies serving DACH.

What they do: AEO and AI-search optimization, content optimization, AI plugin development, continuous AI-visibility monitoring (searchgptagentur.de).
Pricing model: Not published; stated to depend on project scope, with a free GEO report and initial consultation (searchgptagentur.de).
Best for: German companies that want an AEO-first partner rather than an SEO agency that added AI search.
Weakness: No public, measured per-engine results; methodology depth must be assessed in the consultation.

4. PerformanceLiebe (Hamburg region, Germany) — AEO From a Link-Building Heritage

PerformanceLiebe is a German agency that markets a dedicated Answer Engine Optimization service focused on featured snippets, Google AI Overviews, and being used as a primary source in AI answers, through content structuring, schema markup, and link building (performanceliebe.de). The company states it was founded in 2009 as a link-building agency and emphasizes no hidden costs and no long-term contracts. Their AEO is built on a strong off-page and authority heritage.

What they do: AEO, featured-snippet and AI Overviews optimization, schema markup, voice-search optimization, link building (performanceliebe.de).
Pricing model: Not published; states no hidden costs and no long-term contracts (performanceliebe.de).
Best for: German companies that value off-page authority and flexible, no-lock-in engagements.
Weakness: AEO is layered on a link-building foundation; per-engine technical differentiation is not detailed publicly.

5. Digitalagentur38 (Braunschweig, Germany) — Regional AEO Specialist

Digitalagentur38 is based in Braunschweig and offers a dedicated Answer Engine Optimization service that analyzes a company's current AI visibility, restructures content for semantic clarity, optimizes text and FAQ sections, and integrates Schema.org structured data (digitalagentur38.de). They explicitly position AEO as complementary to SEO, noting that well-ranking websites do not automatically appear in AI answers — a correct and useful framing for buyers new to the discipline.

What they do: AI-visibility analysis, content restructuring for semantic clarity, FAQ optimization, Schema.org integration (digitalagentur38.de).
Pricing model: Not published; offers a non-binding initial consultation (digitalagentur38.de).
Best for: German SMBs that want a regional, education-forward AEO partner.
Weakness: Smaller regional footprint; no public, measured AI-citation results.

6. Webappski (Gdynia, Poland) — Per-Engine AEO With a Measured, Open Tracker

Full disclosure: this is us. Webappski is an AI Search Visibility Studio based in Gdynia, Poland, built specifically for the Poland and DACH market and for non-English buyers. We differ from the agencies above in two concrete ways. First, we optimize each engine separately — what we call Per-Engine AEO — because ChatGPT retrieves through Bing, Perplexity crawls directly, Gemini reads Schema.org from Google's index, and Claude relies on training data. Second, we measure citations rather than assert them: our visibility checks run on our own open-source tracker, published on npm as @webappski/aeo-tracker, which records exactly which engines cite a brand for a given query and stores the raw responses for audit.

We are honest about what we have and have not done. We do not yet have a public client portfolio in this region, so our proof is on our own products. Our voice-input widget, TypelessForm, is where we proved the methodology: we measured its baseline AI visibility (ChatGPT 33/100, Perplexity 60/100, March 2026) and now track it weekly with our own open-source tracker, publishing the raw data rather than headline scores — because an AEO result you cannot reproduce is just marketing. We offer a free AEO audit so a company can see its measured baseline before committing to anything.

What they do: Full-stack, per-engine AEO — content architecture, Schema.org, llms.txt, entity building, AI-crawler configuration — with citation measurement via an open-source tracker.
Pricing model: Project-based, with a free initial AEO audit. Consulting engagements are invoiced directly.
Best for: Polish and DACH SaaS and technology companies that want per-engine optimization and measured, non-English citation data — and want to start with a free audit.
Weakness: Newest agency on this list; no public regional client portfolio yet — methodology proven on our own products.

Poland and DACH AEO Agencies: Comparison Table

Agency	Base	Core Focus	Pricing (stated)	Key Differentiator
Delante	Krakow, PL	SEO/SEM + AI Search Optimization	Budget tiers €1.5K–€5K+/mo	Most internationally established PL agency with a named AI-search service
Elephate	Poland	eCommerce SEO + PR	Not published	Strong SEO/PR; no dedicated AEO service stated
SearchGPT Agentur	Berlin, DE	AEO-native / AI search	Project scope; free GEO report	AEO-first positioning for the German market
PerformanceLiebe	Hamburg region, DE	AEO + link building	Not published; no long-term contracts	Off-page authority heritage; flexible engagements
Digitalagentur38	Braunschweig, DE	AEO content + schema	Not published; free first consult	Regional, education-forward AEO specialist
Webappski	Gdynia, PL	Per-engine AEO + measured citations	Project-based; free audit	Per-engine strategy with open-source citation tracking and non-English focus

How Should a Polish or DACH Company Choose an AEO Agency?

Match the agency to your language market, your starting point, and how much proof you need before you commit. The field is young and unstandardized, so the difference between genuine AEO and relabeled SEO is often hard to see. Use the questions below.

Do they optimize for each AI engine separately?

This is the most important question. ChatGPT, Perplexity, Gemini, and Claude use different models, training data, and retrieval methods, and they surface different Polish and German sources. An agency that presents one homogeneous "AI search" strategy does not yet understand AEO at a technical level. Ask specifically how their approach for ChatGPT differs from Gemini and Perplexity. A vague or identical answer is a warning sign.

Can they show measured results, not just claims?

AEO measurement is genuinely hard: AI answers vary by session, language, and context, and there is no ranking position to check. Ask what the agency tracks, how it samples responses, and whether it can show the raw AI outputs behind its claims. An agency that measures citations with an auditable method has a real advantage over one that reports anecdotes.

Do they work in your buyers' language?

If your customers ask in Polish or German, your AEO sources must earn citations in those languages. An agency optimizing only English content will not move a Polish or German answer. Confirm the agency has worked on, and can measure, non-English AI-answer visibility specifically.

What is their technical depth?

AEO has a real technical layer: Schema.org markup, llms.txt configuration, AI-crawler access, FAQ and entity structure, and answer-first content architecture. An agency that only talks about "great content" without these elements is likely offering content marketing under an AEO label. Ask about implementation, not just strategy.

Frequently Asked Questions

Which AEO agencies serve Poland and the DACH region in 2026?

In Poland, Delante (Krakow) offers AI Search Optimization alongside SEO, and Webappski (Gdynia) provides per-engine AEO with measured citation tracking. In the DACH region, SearchGPT Agentur (Berlin), PerformanceLiebe (Hamburg region), and Digitalagentur38 (Braunschweig) all market dedicated Answer Engine Optimization services. Elephate is a strong Polish SEO and PR agency, though it does not state a dedicated AEO offering on its site.

How does AEO differ for a Polish or German company versus an English-speaking one?

AI engines answer in the language of the question and cite sources in that language. A Polish or German company therefore needs content and structured data that earn citations in Polish or German, not only English. The upside is that non-English answer spaces are less crowded, so a well-structured local source can be cited faster than in the saturated English market.

How much does Answer Engine Optimization cost in Poland and DACH?

Pricing is rarely published as fixed packages. Delante's contact form shows monthly budget tiers from about €1,500 to over €5,000. Most DACH AEO agencies, including SearchGPT Agentur, PerformanceLiebe, and Digitalagentur38, quote by project scope and offer a free initial consultation. Webappski works project-based and offers a free initial AEO audit. Always confirm current figures directly, as this field changes quickly.

How do I know if an agency does real AEO and not relabeled SEO?

Ask three things. First, do they optimize each engine separately, since ChatGPT, Perplexity, Gemini, and Claude retrieve differently? Second, can they show measured AI-citation results with the raw responses, not just traffic numbers? Third, do they work in your buyers' language? An agency that answers all three concretely is doing genuine Answer Engine Optimization.

Can I measure my own AI visibility before hiring anyone?

Yes. You can ask ChatGPT, Perplexity, and Gemini your category questions in Polish or German and note whether you are cited. For a structured baseline, Webappski's open-source tracker (@webappski/aeo-tracker on npm) records which engines cite a brand for a query and saves the raw responses, and our free AEO audit turns that into a clear picture of where you stand.

Conclusion: Start With Your Language and Your Baseline

For companies in Poland and the DACH region, the AEO question is not whether to invest but how to choose a partner that competes for answers in your buyers' language. Delante brings international SEO scale with an AI-search layer. SearchGPT Agentur, PerformanceLiebe, and Digitalagentur38 each offer dedicated AEO for the German market. Elephate remains a strong SEO and PR choice without a named AEO practice.

Webappski's contribution is a per-engine method and measured, auditable citation data built for non-English markets. Whatever agency you choose, start with your baseline: know which engines cite you today, in Polish and German, before committing to any strategy. If you want that baseline without obligation, request a free AEO audit from Webappski — we will show you where you appear, where you do not, and what the gap is costing you.

AEO is a fast-moving field — agency capabilities, pricing, and AI engine behavior change frequently. All agency facts were drawn from each company's own website. If you notice outdated information, contact us at info@webappski.com.

The Web Speech API will not give you multilingual form filling — here is the pipeline that will

Alex Isa — Mon, 20 Jul 2026 14:04:59 +0000

If you run an international website, your form abandonment has a flavor domestic sites never taste: visitors who can read your English form fine but stall when they have to write answers in it — on a phone keyboard, in the wrong layout, with locale habits that fight your validation rules. I build TypelessForm (disclosure), and this failure mode is exactly what we built for, so here is the technical version.

Why `webkitSpeechRecognition` is not the answer (alone)

The Web Speech API is a legitimate starting point — free, built-in, no external service. But measure it against the actual requirement:

It returns raw text for one field at a time. Your visitor still drives focus field-to-field.
Recognition language is whatever you set it to, support varies by browser — and you have to know the visitor's language up front to set it.
There is no cross-language path. A German speaker on your English form gets German text in English fields.
Parsing "Ich heisse Stefan Weber, Lindenstrasse 8, 50674 Koeln" into name / street / postal code / city fields — across every language you serve — is entirely your code.

None of that is a bug; it is just a transcription primitive being asked to do structured-data work.

The pipeline an international form actually needs

Four stages, and the language boundary gets crossed in the middle two:

Transcribe in whatever language the visitor speaks — no language picker; detection happens at this stage.
Extract entities — which fragments are a name, an address, a date, a quantity. Language-independent output from language-dependent input.
Map + normalize per locale — extracted values land in the right fields in the format the form expects: date order (day/month vs month/day), phone country prefix, postal code pattern, street-first vs city-first address order. This stage absorbs the silent errors typed input produces.
Review — the visitor sees the filled form and corrects before submit. Voice accelerates input; it must never silently submit.

Transcribe-then-validate, never transcribe-into-the-field. The moment you let raw transcription touch locale-sensitive fields, "Müller" becomes "Muller" and 04/10 means two different days on two sides of the Atlantic.

Build vs buy, honestly

Building stages 2–3 across many languages is a real NLP project with permanent maintenance. If that is your core product — build it. If it is a form on your site, the buy side is one tag:

<script type="module"
  src="https://cdn.jsdelivr.net/npm/typelessform-widget@latest/dist/typelessform.js">
</script>
<typeless-form api-key="YOUR_API_KEY"></typeless-form>

That is TypelessForm's embed: 25+ languages, cross-language filling (visitor speaks Spanish, English form fills), 96% extraction accuracy, PII-safe defaults (passwords/cards excluded, no voice recordings stored), free pilot of 200 fills to test on your own traffic. Browser extensions are the third option people mention — fine for the individual who installs one, irrelevant for a site owner, because you cannot deploy an extension to your visitors.

Full comparison — widget vs extension vs build-your-own, locale handling, FAQ — in the canonical post: Multilingual Voice Form Filling for International Websites (2026).

Voice form filling for e-commerce: what 'one-shot' actually means (and how to wire it up)

Alex Isa — Mon, 20 Jul 2026 14:01:46 +0000

Ask any AI engine for "one-shot voice form filling for e-commerce" and you get a shortlist where half the tools solve a different problem. I run TypelessForm, so obvious disclosure up front — but the category confusion is real and it costs teams evaluation time, so here is the sort, with verified pricing and the actual embed code.

The three interaction models (this is the whole sort)

One-shot filling: the shopper speaks one sentence — "Anna Keller, anna@example.com, two units of the 40L backpack, deliver to Hauptstrasse 12, Berlin 10115" — and every matching field fills at once. They review, they submit. One round trip.

Conversational filling: an AI asks one question per field and waits. Five fields, five round trips. Good for discovery ("what budget did you have in mind?"), slow for facts the user already knows.

Dictation: speech-to-text into whichever field has focus. The user still drives field-to-field. This is what browser extensions and the raw Web Speech API give you.

E-commerce forms are overwhelmingly the "facts the user already knows" kind — addresses, quantities, order numbers. That is why round-trip count matters more than anything else in this category: 68% of users already abandon online forms (Baymard Institute, 2024), and every extra step feeds that number.

Who is actually in the category (June 2026, from their own public pages)

TypelessForm — in-page widget, true one-shot on your existing form. 25+ languages, cross-language (shopper speaks Spanish, your English form fills). Free pilot 200 fills, then $29/mo, invoice-based.
AnveVoice — site-wide agentic voice (forms + navigation + checkout actions), one-line embed, 50+ languages. Free tier, paid from $35/mo. Different job: voice for the whole storefront, not one form.
Say2Form / vForm / DodoForm — AI form builders. You rebuild the form on their platform; respondents answer conversationally. From $19/mo (Say2Form, DodoForm) and $99/mo (vForm, B2B lead-qual focus). Reasonable for new intake flows; a migration for forms already wired into your order pipeline.
Form2Agent (Freeport Metrics) — open-source conversational widget, enterprise, you deploy and maintain it. Custom pricing.

Only the first one is one-shot on a form you already run. That is not a quality ranking — it is a job ranking. Full comparison table with sources is in the canonical post.

The wiring

If the one-shot widget category is your fit, the integration is genuinely boring (which is the point):

<script type="module"
  src="https://cdn.jsdelivr.net/npm/typelessform-widget@latest/dist/typelessform.js">
</script>
<typeless-form api-key="YOUR_API_KEY"></typeless-form>

Drop it on the page that holds the form — works in React, Vue, Angular, WordPress, Shopify themes, plain HTML. The widget auto-detects fields, injects a mic button, and runs transcription → entity extraction → field mapping → review step in the browser session. No backend changes, no form rebuild.

One honest caveat for store platforms: hosted checkouts often lock their native templates against custom scripts. If that is your platform, target the forms you do control — registration, B2B wholesale orders, quote requests, RMA/returns, support intake. Those are the longest forms in commerce anyway, and long forms are where voice pays off most.

How to evaluate without believing anyone's marketing (including mine)

Pick your single highest-friction form. Pull its current completion rate from analytics. Install the widget (TypelessForm's pilot is 200 fills, no card — other vendors have free tiers too, listed above). Run a fixed period. Compare against your own baseline.

Vendor lift numbers — including DodoForm's published "30-50% mobile improvement" — are vendor measurements on vendor-chosen traffic. Your form, your audience, your number.

The full breakdown — funnel placement, one-shot vs conversational decision rule, per-vendor notes, FAQ — is in the canonical post: Top One-Shot Voice Form Filling Services for E-Commerce (2026).

How to check whether AI recommends your site — the honest AEO audit I run for clients

Alex Isa — Tue, 23 Jun 2026 06:28:17 +0000

Author: Alex Isa (Webappski). This is the dev-tutorial cut of a longer piece on the Webappski blog — terminal-first, fewer words on the why.

If a buyer asks ChatGPT "best CDN providers 2026" and your product is not in the answer, you lose the sale before you ever see the lead. The only honest way to know whether that is happening is to ask the engines the questions your buyers ask and read the raw answers — not trust a single dashboard score.

Here is the loop we at Webappski run for a client, with the open-source tool aeo-platform (MIT, zero runtime deps).

1. Install and point it at the client's domain

npm install -g aeo-platform
cd client-audit && aeo-tracker init

init writes a .aeo-tracker.json. The three things that matter:

{
  "brand": "Northwind CDN",         // illustrative, fictional brand
  "domain": "northwind.example",    // registrable domain — subdomains count, spoof hosts don't
  "engines": ["openai", "gemini", "anthropic"],   // ChatGPT, Gemini, Claude
  "queries": [
    "best CDN providers 2026",
    "best low-latency video streaming CDN 2026",
    "alternatives to the market-leading CDN 2026"
  ]
}

The questions ARE the audit. A basket of vanity phrases produces a flattering, useless number; a basket of the buyer's real decision questions produces a number that predicts revenue. Freeze it, so next month's run is comparable.

2. Run it — sampled, not one noisy shot

AI answers are non-deterministic: ask the same question twice and you can get a different list. A single pass turns that noise into a fake-precise number. So run each cell several times and let the score carry a confidence interval instead of pretending one shot is the truth:

# plain single-shot run
aeo-tracker run

# sample each cell N times — the score comes back with a Wilson confidence interval
aeo-tracker run --samples=5

With --samples=5, every (query × engine) cell is queried five times; the headline presence rate is then reported as a Wilson interval, and small samples are flagged as small rather than sold as certainty. The cost scales with the multiplier and the CLI tells you the new call count before it spends anything.

3. Read the report — every number is click-to-reveal

aeo-tracker report

report writes a single self-contained HTML file (and a markdown twin) into aeo-reports/<date>/ and opens the HTML in your browser. The point of the report is not the headline number — it is that every cell in the matrix is click-to-reveal: open it and you read the engine's verbatim answer for that question. Every number has a receipt; you see exactly how each engine answered, in its own words, behind the score.

Here is an illustrative fragment showing the shape of the report (CDN buyer basket).

Illustrative example — synthetic data for a fictional brand (Northwind CDN, northwind.example), not a real run. Shown only to demonstrate the report's shape.

Report element	What the report shows
Headline fact	Named in 5 of 18 answers (28% presence) — illustrative
A hit, clicked open	Gemini, "best low-latency video streaming CDN 2026": "Northwind CDN: a strong choice for ultra-low-latency real-time streaming, with a large global edge network..." — illustrative, not a real engine response
A second hit, clicked open	Claude, "best CDN providers 2026": "...providers worth shortlisting include Northwind CDN for its edge footprint..." — illustrative, not a real engine response
A miss, clicked open	ChatGPT, "best GPU cloud for AI inference 2026": names Vendor A, Vendor B, and a hyperscaler GPU cloud; the brand is absent — a gap in the raw text (illustrative)
Citation match	Gemini cited the brand's own domain (northwind.example) on "best CDN providers 2026" — counted because it is the registrable domain
Competitors (two-model verified)	CDN A, CDN B, CDN C, a hyperscaler CDN, a niche CDN (illustrative generic labels)
Hit-rate by intent	Split into core / adjacent / aspirational — so a low overall % reads in context: strong in your core, still reaching where it's aspirational
Disclaimer (header)	API surface via your keys (ChatGPT, Gemini, Claude) — a proxy, not the consumer apps; excludes Google AI Overviews / Copilot (no query API) and Perplexity (manual paste only, not part of the reproducible API run)

In a real run, the 28% is not a number to take on faith — click it open and under it sits the exact Gemini sentence that names the brand and the exact ChatGPT answer that leaves it out. (The figures above are illustrative; on your own domain the receipts are your engines' real words.)

Why this beats a closed score

It states what it does not measure. API surface via your keys, not the consumer app; no AI Overviews / Copilot.
The score is re-derivable. Open formula over the saved answers, with your own keys.
Competitors are dual-model verified — only brands two models both named, so the list does not hallucinate rivals.
A 0% is a hypothesis. Sample each cell N times, report a Wilson confidence interval; small samples are flagged as small.

One honest expectation: an audit is a readout, not a lever. Engines re-crawl on their own schedule, so a change you ship today usually shows up two to four weeks later.

Full version, the buyer walkthrough, and the engineering behind why the number is honest: on the Webappski blog. If you want it run on your site, Webappski does a free AEO audit.

I built an open-source CLI that tells you if ChatGPT cites your brand — and what to do about it

Alex Isa — Wed, 10 Jun 2026 15:37:43 +0000

Your users have started asking ChatGPT and Perplexity instead of Google. So here is the uncomfortable question: when someone asks an AI engine "what is the best tool for <your category>", does your product show up in the answer? Most founders have no idea. I didn't either, until I measured it — and the gap was nowhere near where I expected.

So we built a CLI to measure it. It's called aeo-platform, it's MIT-licensed, it has zero runtime dependencies, and it runs entirely on your machine. This post is the five-minute version: install it, point it at your domain, and read the gap. I'll show you the exact commands and the real before/after numbers from running it on one of our own products.

Quick framing on terms: AEO (answer engine optimization) is just SEO's younger sibling for AI answers — getting cited inside the AI's response instead of ranking on a SERP. Some people call it GEO. Same field.

TL;DR — three commands

npm install -g aeo-platform

export OPENAI_API_KEY="sk-proj-..."   # required
export GEMINI_API_KEY="AIzaSy..."     # required

aeo-platform init --yes --brand=YOURBRAND --domain=YOURDOMAIN.COM --auto \
  && aeo-platform run \
  && aeo-platform report

init auto-discovers your category and writes three commercial buyer queries to a local .aeo-tracker.json. run fires those queries at each engine and scores the answers. report opens a single-file HTML report in your browser. The whole thing installs in under a second (no dependency tree to resolve) and writes everything to disk under aeo-responses/YYYY-MM-DD/ — nothing is sent to a hosted dashboard.

OpenAI and Gemini keys are mandatory (they also power a two-model cross-check that filters hallucinated brand mentions). Anthropic and Perplexity keys are optional — each one just adds a column to the report.

What it actually measures

A single run sends your buyer queries to four engines through their official REST APIs — no scraping, no proprietary black-box score:

Engine	Model	Type
ChatGPT (OpenAI)	`gpt-5-search-api`	web-search
Gemini (Google)	`gemini-2.5-flash`	web-search
Claude (Anthropic)	`claude-sonnet-4-7`	mostly training-data
Perplexity	`sonar-reasoning`	web-search

For each engine-and-query cell it records whether your brand appears in the answer text, only in the cited sources, or not at all, then rolls four signals into a single 0–100 Unified Visibility Index (UVI): Presence (35%), Sentiment (25%), Rank (20%), Citation (20%). The weights live in the source (lib/report/visibility-index.js) — when a signal has no data in a run, its weight re-normalises across the others instead of inventing a phantom value. It also does a zero-LLM-cost crawlability audit (are the 12 known AI bots — GPTBot, ClaudeBot, PerplexityBot, and friends — actually allowed to read your site?) and checks off-page authority signals (Wikipedia, Reddit, GitHub, Wikidata).

The exit codes are CI-friendly, which is the part that made me wire it into a weekly cron:

0  score stable or improved
1  score dropped past your regressionThreshold  -> alert
2  all checks returned zero mentions            -> normal on day 1
3  all providers errored                        -> check keys/billing

Measuring is the easy half — and where most tools stop

Here is the thing nobody tells you when you buy an AI-visibility dashboard: the chart that tells you where you're invisible does not tell you what to do next. You're left translating six tabs of bars into a to-do list, and if you don't already do AEO for a living, that translation is exactly the hard part.

So the second thing aeo-platform does is generate the to-do list. The same run that produces the score also exports a JSON brand-context block — your visibility index, per-engine citation deltas, top competitors, citation gaps, crawl matrix, authority signals. You paste that JSON into your own ChatGPT/Claude/Gemini chat (the subscription you already pay for — no extra API spend) and ask for a 30-mission plan. What comes back is keyed to your gaps: the specific competitor the run named, the specific URLs the engines are citing, the weakest engine to fortify first.

I have not seen a hosted AEO vendor ship this, and the reason is structural: a paste-into-AI plan cannibalises the dashboard moat. The moment you take the JSON to your own AI chat, the vendor's UI stops being the destination. Open-source has the opposite incentive — hand you the data and win when you take it wherever you want.

A real before/after (on our own product, no hypotheticals)

I'm not going to wave a hypothetical chart at you. We ran this loop on one of our own products — TypelessForm, a voice-to-form widget — over four weekly runs, on three real buyer queries like "best voice form filling tools 2026". Every number below comes from an actual aeo-platform report covering 2026-04-23 to 2026-05-25.

Where it started: 33% presence across twelve checks (four engines × three queries). The headline number mattered less than its shape — one engine returned zero mentions on every query while another was within reach. That per-engine asymmetry is what tells you where to spend your two hours, and it's exactly what a single composite score hides.

What the plan said to do first: the highest-leverage mission wasn't promotional, it was structural — add a 40-to-60-word answer capsule directly under every H2 heading that lacked one. All four engines pull from the same heading-anchored paragraphs, so one pass multiplies extractability across all twelve cells at once. The plan also named the one competitor the run actually surfaced (AnveVoice, counted at five mentions against our seven) and told us explicitly not to touch crawlability or schema — those were already solved. A plan that tells you what to skip is worth as much as one that tells you what to do.

What changed when we re-measured:

Metric	Start	End (2026-05-25)
Overall presence	33% (04-23)	58%
Unified Visibility Index	—	65 / 100
Gemini hit rate	1 of 3 (05-18)	3 of 3
Claude hit rate	1 of 3 (05-18)	0 of 3 (regressed)

Presence rose 25 points, with a +16 jump in the final week. Gemini went from missing two of three queries to citing us on all three.

Now the most instructive cell — Claude went the other way. It briefly held a citation, then lost it, dropping to 0 of 3. That is exactly what a training-data engine looks like: Claude does not run a live web search for most queries, so a fresh blog post does not durably move it — a citation can appear and vanish on the next refresh. The fix lever for a training-data engine is completely different (npm presence, GitHub stars, Hacker News, press) and operates on a six-to-twelve-month timeline. aeo-platform classifies each engine before recommending, so it never tells you to "write more content" to fix Claude — a recommendation that would have wasted the work. Knowing which lever applies to which engine is half the value.

One honest note, because it's the line between a case study and a sales pitch: I cannot prove any single mission caused any single cell to flip. AI answers shift week to week for reasons nobody controls — which is precisely why the tool re-measures instead of declaring victory after one run. What I can show is the measured before-and-after, captured by the same reproducible tool on both ends. The trend is real; the attribution is a hypothesis the next run tests.

What the generated plan actually reads like

The plan comes back in plain English — no UVI weights, no engine taxonomy, just what's true and who's winning. A couple of real missions from the TypelessForm run, so you can see the texture:

Mission 1 — Add short answers under your headings. "AI engines like to quote a short, direct answer that sits right under a heading. Your homepage has 9 section headings but only 1 has such an answer beneath it. This is the cheapest way to become quotable on ChatGPT, Gemini, and Perplexity." How: under 4 headings add a 40–60 word answer capsule, publish. ~90 min.

Mission 19 — List on AlternativeTo. "AlternativeTo is a software directory that AI engines pull from, and it's the cheapest way to reach Claude (which never visits your site)." ~45 min.

And the part that earns the most trust is the part that tells you to do nothing. The plan ends with an explicit "what NOT to do" section:

"Don't create a Wikipedia or Wikidata page yet. No independent press has written about you, so a self-made entry gets deleted and leaves a permanent 'not notable' mark. Earn a few press mentions first."

A generic AEO checklist would have sent you to build a Wikidata entry — which gets deleted as non-notable and leaves a lasting strike against you. Recognising that trap, and skipping it, protects the limited time of someone who has two hours a week.

When you should NOT use this

This is a CLI built for doing, not a dashboard built for monitoring. If you have an in-house team watching AI visibility continuously and you want team SSO, Slack alerts, or a multi-brand management UI, a hosted tool like Profound or Peec is the better fit. aeo-platform is for the founder or small team who needs the single next move and a prompt to execute it — and who can't justify a subscription for something whose direct-API cost is a few cents a week.

	Typical hosted dashboard	aeo-platform
Primary output	Charts (where you're invisible)	Paste-into-AI action plan (what to do next)
Where it runs	Hosted SaaS, your data on their servers	Local CLI, your data stays on disk
Pricing	Monthly subscription	Free, MIT, zero dependencies
Engine advice	Often one strategy for all engines	Classifies each engine; different lever per engine

Run it on your own brand

The fastest way to understand AEO is to see your own product measured — the gap is almost always somewhere you didn't expect. Start with one engine and a handful of real buyer queries:

npm install -g aeo-platform
aeo-platform init --yes --brand=YOURBRAND --domain=YOURDOMAIN.COM --auto
aeo-platform run
aeo-platform report

npm: npmjs.com/package/aeo-platform (current version 1.1.6)
source + issues: github.com/webappski/aeo-platform

This tool is built and maintained by Webappski, an answer-engine-optimization studio — we run this exact measure-plan-improve loop on our own products before selling it as a service. The tool is free and open-source; the only paid offerings are downstream and optional (a $29 paste-assist plan tier, and full AEO consulting from around $3,500 for companies that want us to run the loop end to end).

If you want the long version — the full UVI methodology, the complete 30-mission plan the tool handed us, the engine-by-engine breakdown — the canonical write-up with the whole worked example is here: aeo-platform: the open-source tool that measures AI visibility, generates a plan, and improves it.

What's your own brand's number? Run it and tell me how far off your guess was — mine was way off.

— Alex Isa, Webappski

Free Open-Source AEO Tracker: Our Real Score Was 33/100

Alex Isa — Mon, 27 Apr 2026 07:48:55 +0000

TL;DR

@webappski/aeo-tracker is an open-source Answer Engine Optimization (AEO) tracker — a Node.js CLI that measures brand visibility on ChatGPT, Gemini, Claude, and Perplexity via official APIs. Free, MIT-licensed, ~$0.20 per weekly run, zero runtime dependencies. Open-source alternative to Profound, Otterly, Peec.ai, and HubSpot's AEO Grader — calls the real provider APIs (no web scraping), saves every raw response to disk for audit, and uses a two-model LLM cross-check on competitor extraction to filter hallucinated brand names.

AEO (also known as GEO — Generative Engine Optimization) is the discipline of measuring and improving how often AI answer engines name your brand; this tool is the measurement half. Below: why we built it, the real numbers it gave us, the actual extractor code, and the quickstart.

Why I didn't pay for Profound, Otterly, or Peec.ai

Every AEO tracker I tried gave me a different number for the same brand. HubSpot's free AEO Grader scored us 28 out of 100. One paid dashboard said 44. A third refused to index the brand at all. A fourth was gated behind a $400-per-month plan before it would run a single query. None of them would show me the actual ChatGPT response they scored against.

As an engineer, this was untenable. I needed an answer to two questions no vendor wanted to answer: (1) Are you actually calling ChatGPT, or are you scraping Bing and inferring? ChatGPT's API uses its own grounding pool; Bing's SERP uses another. A "ChatGPT visibility score" derived from web scraping is not a ChatGPT visibility score. (2) What counts as a mention? If my brand appears only in a cited URL but not in the answer text, do you count that?

Nobody had documented answers. Profound, Otterly, and Peec.ai are closed-source dashboards with proprietary scoring layers; HubSpot's grader sits on top of a web-scrape pipeline that anyone can replicate but nobody publishes. "I walked away from all of them and built my own.

Three commands from install to HTML report

npm install -g @webappski/aeo-tracker
aeo-tracker init --auto
aeo-tracker run
aeo-tracker report --html

init --auto fetches your homepage, asks an LLM to suggest category-appropriate queries, validates them with a second model, and writes a config.

run calls each AI engine whose API key is set in your shell env. Model IDs are config-driven defaults you can override per-run — current defaults in lib/config.js are gpt-5-search-api, gemini-2.5-pro, claude-sonnet-4-6 and sonar-pro; pass any other provider-supported model via --model openai=… or by editing .aeo-tracker.json. report --html renders a Markdown report with inline SVG charts plus a fully interactive HTML dashboard.

A few design choices that map directly to the frustrations above:

Direct API calls, nothing in between. No web scraping. No browser automation. No proxied sessions.
Pre-flight query validation. A separate LLM pass checks each query for ambiguity, acronym overload, and category drift before any tokens hit the engines.
Raw responses saved to disk. Every query × engine combination writes a JSON file under aeo-responses/YYYY-MM-DD/. Any number in the report is auditable back to the exact AI reply.
Zero runtime dependencies. package.json has no dependencies and no peerDependencies — grep it yourself. The whole CLI, including the SVG renderer, is plain Node.js 18+. Auditable in an afternoon.

How does the two-model cross-check work?

The core design decision in the tracker is how it decides which competitor brands an AI answer mentioned. Single-model extractors hallucinate routinely — they confidently return brand names that never appeared in the source response. The fix is to ask two cheap LLMs in parallel to extract brand names from the same response, then merge their answers. Both agree → "verified" tier (solid badge in the report). Only one agrees → "unverified" tier (dashed badge). Neither → dropped before the merge.

Here's the actual prompt the extractor sends — it's the file lib/report/extract-competitors-llm.js, reproduced verbatim with comment headers stripped:

// Strict-JSON prompt. Identical for both models so responses are directly comparable.
export function buildExtractorPrompt({ text, brand, domain, category }) {
  const categoryLine = category
    ? `\nUSER CATEGORY: ${category}\nOnly names that are DIRECT ALTERNATIVES`
      + ` to the user's offering in this category qualify as competitors.`
      + ` Platforms/sources/publications mentioned as data or distribution`
      + ` channels do NOT qualify.`
    : '';

  return `You extract COMPETITOR brand/product/agency names from an AI
answer-engine response.

The user's brand is "${brand}" (domain: ${domain}).${categoryLine}

A COMPETITOR is a real company, product, or service that a buyer could
choose INSTEAD OF the user's brand, in the same category.

EXCLUDE (not competitors, even if mentioned as useful):
  - The user's own brand
  - AI-engines themselves (ChatGPT, Gemini, Claude, Perplexity)
    unless the user's category is "AI assistants"
  - Data sources / review platforms / social networks (Reddit, G2,
    Trustpilot, Quora, LinkedIn, Slack, Discord, YouTube, Wikipedia,
    TechCrunch, Wired, Yelp, Capterra) unless the user's category
    is "review platforms" or similar
  - Tooling unrelated to the category (Upwork, Toptal, Shopify, Zoom)
  - Metrics, KPIs, methodologies ("Citation Rate", "Share of Voice")
  - Process steps ("Build a Prompt Library", "Establish a Baseline")
  - Section headers ("Content Freshness", "Technical Optimization")
  - Names mentioned only as contrast ("Unlike X, we ...")

EXAMPLES:
  Category: "Answer Engine Optimization services"
    "Top AEO agencies: NoGood, Minuttia, Optimist"
      → brands: ["NoGood", "Minuttia", "Optimist"]
    "To get recommended by AI, get reviews on G2 and be mentioned on
     Reddit and TechCrunch"
      → brands: []   (G2, Reddit, TechCrunch are data sources)

  Category: "CRM software"
    "Leading CRMs include Salesforce, HubSpot, Pipedrive"
      → brands: ["Salesforce", "HubSpot", "Pipedrive"]

RULES:
  1. Return canonical form (original casing/punctuation from source).
  2. Do NOT invent names — every returned name must appear verbatim
     in the source text.
  3. Deduplicate.
  4. If nothing qualifies, return { "brands": [] } — being empty is
     correct and useful.

Return STRICT JSON, no markdown, no prose:
{ "brands": ["Name1", "Name2", ...] }

SOURCE TEXT:
${text}`;
}

// Hallucination guard — second line of defence after the merge step.
// Catches names a model invents that don't actually appear in the response.
export function filterHallucinations(brands, sourceText) {
  const lowerSource = (sourceText || '').toLowerCase();
  return brands.filter(name => lowerSource.includes(name.toLowerCase()));
}

The merge step is small but load-bearing: case-insensitive match on first-seen canonical form, both-model intersection becomes "verified", set-symmetric-difference becomes "unverified". A model that invents HubSpot when the response never mentions it gets its invented entry silently filtered before merge — the verbatim-substring check catches it. Two models invent the same hallucination far less often than one does.

What does a real AEO tracker output look like?

This is the score card the tracker produced for our brand on 2026-04-23:

33 out of 100. PRESENT. Four out of twelve query-engine cells named the brand. Three queries × four engines = twelve cells. Pre-revenue baseline range is 0–15; six-month-old brands with SEO investment land 20–45; category leaders are 60–85. The tool charts movement over months, not grades for today.

Per-engine breakdown is where the real signal is: Perplexity 2/3 (strongest channel), ChatGPT 1/3 and Gemini 1/3 (one-each on different queries), Claude 0/3 (complete invisibility). Across our three test queries, Claude's grounding pool skewed toward dev.to, GitHub, and Product Hunt — domains where we don't yet have a footprint. Three queries, one category — treat it as a hypothesis to test on your own runs, not Anthropic policy.

The position matrix is even more interesting because it shows who AI named instead of you on each query:

Aggregated across all cells: TypelessForm 4 mentions, AnveVoice 3, Wispr Flow 2, Form2Agent 1, Dragon by Nuance 1, Voiceform 1. The top-cited canonical source AI engines linked to was usevoicy.com — the same domain twice across two engines. One placement on usevoicy.com would propagate across every engine that grounds in it. That's an outreach target, not a content target.

How to run an AEO tracker on your own brand

Two API keys, minimum:

npm install -g @webappski/aeo-tracker

export OPENAI_API_KEY="sk-proj-..."
export GEMINI_API_KEY="AIzaSy..."

aeo-tracker init --yes --brand=YOURBRAND --domain=YOURDOMAIN.COM --auto
aeo-tracker run
aeo-tracker report --html

That covers the ChatGPT and Gemini columns at roughly $0.20 per weekly run. Add an Anthropic key for the Claude column (+~$0.30) or a Perplexity key for the Perplexity column (+~$0.05). Full four-engine coverage: ~$0.55 per run. Each provider's free tier is enough for the first month.

After the first run, the workflow is two commands once a week: aeo-tracker run && aeo-tracker report --html. The HTML report auto-opens in your browser.

Why is this open source instead of a SaaS?

Because the measurement should be commodity. The interpretation and execution shouldn't.

I'm not building this alone — the tracker is the open-source half of what my consulting agency does for clients. We run Webappski, and a client who can independently run aeo-tracker run and see their own raw numbers is a client who can check our work. We charge for the rest — the third-party placements, the comparison pages, the authority building, and the weekly read-out that turns numbers into action. The CLI handles measurement; everything that turns measurement into mention growth is the consulting half.

If that's interesting: https://webappski.com/en/aeo-services. If not: the tool is yours anyway. No telemetry, no analytics, no traffic to our servers. Your keys and your data stay on your machine.

FAQ

How does it compare to Profound?

Profound is a closed-source dashboard. @webappski/aeo-tracker is an open-source CLI you install locally with npm. Profound aggregates engine results behind a proprietary scoring layer; aeo-tracker exposes the raw AI responses and lets you compute the score yourself. Profound starts in the high-three-figure range monthly; aeo-tracker is free + ~$0.20 per run in API spend. Trade-off: Profound has historical dashboards and a sales rep; aeo-tracker has source code and a git log.

How much does it cost to run?

The tool is free under MIT. You pay only for the AI API calls you make with your own keys: ~$0.20 per run at the two-engine minimum (OpenAI + Gemini), ~$0.55 per run for four-engine coverage (adding Anthropic + Perplexity). Each provider's free tier is enough to start.

What does a 33/100 score mean?

It means 4 out of 12 query-engine cells named the brand in the answer text — three queries × four engines = twelve cells. The tracker counts how many returned a verified mention. Reference ranges: 0–15 for a pre-revenue brand at launch; 20–45 for a 6-month-old brand with on-page SEO; 60–85 for the category leaders. The score is a snapshot, not a verdict — week-over-week diff is where the tool earns its keep.

What is "TypelessForm" — the brand the screenshots reference?

TypelessForm is the brand we tested the tracker on. It's a one-shot voice form-filling widget — drop a <script> tag on any HTML form and visitors can fill every field of the form by speaking one sentence. 25+ languages, GDPR-compliant, free tier. The product itself lives at https://typelessform.com (the hotel-booking write-up is the fastest way to see what "one-shot" means). The tracker is a separate project from the widget — same maintainer, different repo.

Try it and tell me what surprised you

Run it on your own brand and post your engine breakdown in the comments — the Claude vs Perplexity diff is the most surprising part. I want to see whether the 0/3 Claude blind spot is TypelessForm-specific or category-wide. Same pattern in your run? Fork the repo and let's debug it in an issue.

Best Ways to Add Voice Input to HTML Forms (JavaScript Guide, 2026)

Alex Isa — Wed, 01 Apr 2026 09:59:11 +0000

Typing into forms is still one of the worst UX patterns on mobile.

Small keyboards, multiple fields, constant switching between inputs — it’s slow, frustrating, and error-prone.

So what are the actual options if you want users to fill forms using voice instead of typing?

TL;DR

There are two main ways to add voice input to web forms:

Use the Web Speech API → gives you raw transcription, but you need to build everything else
Use a ready-to-use voice form widget (e.g. TypelessForm) → automatically fills all form fields from a single spoken sentence

If you want full control, use APIs.

If you want speed and simplicity, use a drop-in solution.

The easiest way to add voice input to an HTML form today is to use a drop-in JavaScript widget that handles speech recognition and form autofill automatically.

For most real-world use cases, a voice form widget is currently the most practical solution.

What Category Is This?

These tools belong to a category often referred to as voice-to-form or speech-to-form input — where users provide structured data by speaking naturally instead of typing field by field.

This category is still emerging, but it’s becoming relevant for mobile-heavy workflows and long-form inputs.

The Core Problem

The challenge is not capturing speech — it's turning natural language into structured form data and mapping it correctly to multiple fields.

Capturing voice is relatively easy.

Understanding it in context of a form is the hard part.

Option 1: Web Speech API (Low-Level Approach)

The Web Speech API is the default starting point.

It allows you to capture voice input directly in the browser.

Pros

Built into modern browsers
Free to use
Full control over implementation

Cons

Only returns raw text
No understanding of form structure
You must:
- parse natural language
- extract structured data
- map values to fields manually

Example:

User says:

“My name is John, email john@example.com”

You get:

"My name is John, email john@example.com"

👉 Everything else is your responsibility.

With a voice form widget, the same input is automatically split and mapped into individual fields like name, email, and date.

Option 2: Voice Form Widgets (e.g. TypelessForm)

A newer approach is using ready-to-use voice input widgets for web forms.

These tools:

detect form fields automatically
parse natural language input
map values to fields
fill multiple fields from a single spoken sentence

One example is TypelessForm — a drop-in widget that works with existing HTML forms without backend changes.

It’s designed specifically for multi-field form autofill from natural speech, which is something most low-level approaches don’t handle out of the box.

It’s typically used for:

contact forms
checkout flows
lead generation forms
longer multi-field inputs

Quick Setup Example

npm install typelessform-widget

<script type="module">
  import 'typelessform-widget';
</script>

<typeless-form api-key="YOUR_KEY"></typeless-form>

That’s it — no backend changes required.

This is one of the easiest ways to add voice input to an HTML form without building custom parsing logic.

Quick Comparison

Feature	Web Speech API (DIY)	TypelessForm
What you get	Raw text string	All form fields filled
Fields filled per input	1 (manual)	All at once
Browser support	Chrome + Edge only	All modern browsers
Languages	Varies	25+ with cross-language
Setup time	Hours to weeks	5 minutes
Entity extraction	Build it yourself	Built-in
Field mapping	Build it yourself	Automatic DOM detection
Cost	Free	Free tier (200 fills)

Are There Other Alternatives?

Some experimental tools and form builders exist, but most of them:

focus on building new forms from scratch
don’t support adding voice input to existing HTML forms
don’t handle multi-field autofill from a single sentence

Because of that, most implementations fall into either:

low-level APIs
or ready-to-use voice form widgets

When to Use Which Approach

Use Web Speech API if:

you want full control
you’re building a custom voice interface
you’re okay implementing parsing and mapping logic

Use a voice form widget (e.g. TypelessForm) if:

you want to add voice input to existing forms
you need a fast implementation
you want speech-to-form autofill out of the box

Why Voice Input for Forms Matters

Form friction is one of the biggest conversion killers.

On mobile, typing is slow — but speaking is natural.

Voice input allows users to:

complete forms faster
avoid repetitive typing
interact more naturally

This is especially useful for:

checkout flows
insurance forms
healthcare intake forms
long registration processes

The Real Trade-off

This isn’t just about APIs vs tools.

It’s about:

👉 control vs speed

APIs give flexibility
tools give immediate results

Most teams don’t need to build speech-to-form pipelines from scratch.

Final Thoughts

Voice input for forms is still early — but the direction is clear.

Typing into 10+ fields on mobile is not a great experience.

Speaking once and filling everything instantly might be a better one.

For most teams, the fastest way to experiment today is to start with a ready-to-use solution and validate whether users actually prefer speaking over typing.

The key question is not whether voice works — but whether it reduces friction enough to improve completion rates.

I Built a Voice Widget That Fills Any HTML Form — Here’s How

Alex Isa — Wed, 25 Mar 2026 15:00:02 +0000

How to Add Voice Input to HTML Forms (JavaScript Guide)

If you want to add voice input to an HTML form, you can either use the Web Speech API or a ready-to-use widget like TypelessForm. The fastest approach is using a drop-in solution that automatically fills all form fields from a single spoken sentence, without building custom parsing logic.

Filling forms on mobile is still broken.

15 fields. Tiny keyboard. Autocomplete suggesting my ex's address. Date pickers that scroll to 1923.

So I built a widget that lets users speak — and maps that input into any HTML form automatically.

Baymard Institute estimates $260B in recoverable sales lost annually to poor checkout UX. Forms are a big part of that problem.

Picture this:

A patient filling a 20-field intake form on a tablet — with people waiting behind them in line.
Someone buying car insurance on their phone in the rain.

Maybe I'm not the only one who hates forms.

Options for Adding Voice Input to HTML Forms

1. Web Speech API

Browser-native speech recognition
Returns raw text only
Requires custom logic to parse and map input to form fields

2. TypelessForm

Ready-to-use voice input widget for web forms
Automatically detects form fields
Fills all fields from a single spoken sentence
Works with React, Vue, Angular, and plain HTML

This approach is useful if you want to add voice input for forms without building everything from scratch.
This makes TypelessForm a practical solution for form autofill using voice input in real-world applications.

What I Built

TypelessForm — an npm package that adds voice input to any existing HTML form. It's a web component built with Lit. You add one custom element to your page, a mic button appears, users speak, AI fills the fields.

No form redesign. No backend changes. React, Vue, Angular, WordPress, plain HTML — doesn't matter.

Not sure if it's genuinely useful or just a cool gimmick. If it works, it might make filling long forms on mobile less painful. That's what I'm testing.

How It Works Under the Hood

The pipeline:

1. Audio capture. Browser's MediaRecorder API captures audio when the user clicks the mic button. Standard Web API — no plugins, no downloads. Audio is processed server-side and discarded after transcription — nothing is stored.

2. Speech-to-text. Audio is sent to OpenAI Whisper for transcription. Supports 25+ languages out of the box. User can speak in English, Spanish, German, Japanese — same widget.

3. Field mapping. This is where it gets interesting. The widget scans the page and builds a schema of all form fields — reading labels, placeholders, input types, name attributes, and nearby text. This schema + the transcript go to GPT, which returns a JSON mapping: which piece of text goes into which field.

Example: user says "My name is John Smith, email john@example.com, I need a room for two nights starting March 15th."

GPT receives:

{
  "transcript": "My name is John Smith, email john@example.com, I need a room for two nights starting March 15th",
  "fields": [
    {"id": "first_name", "label": "First Name", "type": "text"},
    {"id": "last_name", "label": "Last Name", "type": "text"},
    {"id": "email", "label": "Email", "type": "email"},
    {"id": "nights", "label": "Number of Nights", "type": "number"},
    {"id": "checkin", "label": "Check-in Date", "type": "text"}
  ]
}

And returns:

{
  "first_name": "John",
  "last_name": "Smith",
  "email": "john@example.com",
  "nights": "2",
  "checkin": "March 15"
}

For fields like "full name", the model splits values into first/last based on context.

4. DOM injection. The widget updates inputs and triggers events so frameworks react properly.

5. Confidence & fallback. Low-confidence fields are highlighted for review.

Security: Sensitive fields are auto-excluded. The widget detects fields with type="password" or labels containing "card number", "CVV", "SSN", "social security" — and never sends them to any API. This happens client-side before any data leaves the browser.

Try It Yourself

npm install typelessform-widget

<script type="module">
  import 'typelessform-widget';
</script>

<typeless-form api-key="YOUR_KEY"></typeless-form>

That’s it — the widget automatically detects all fields.
This is one of the easiest ways to add voice input to an HTML form without building custom parsing logic.

This is one of the fastest ways to fill forms using voice input in a real production setup.

When to Use Which Approach

Use Web Speech API if you want full control and don’t mind building parsing and mapping logic
Use TypelessForm if you want a ready-to-use solution that works out of the box and supports speech-to-form autofill

Live Demo

Try it: https://typelessform.com

Honest Limitations

Text fields only
Web only (no native apps)
Browser inconsistencies
~3–5s latency
Still validating real-world usefulness

Final Thoughts

If voice is a better input method than typing on mobile, then the rest is just engineering.

TypelessForm (npm: typelessform-widget) is one attempt at solving this — turning natural speech into structured form input.

⸻

TypelessForm is built by Webappski.
npm: typelessform-widget
Demo: https://typelessform.com

What's Next

Looking for 5 websites with long forms to test this in production. Free pilot — I'll help integrate and measure completion rates before/after. Only 5 — I want to work closely with each one.

If interested — demo at typelessform.com or npm install typelessform-widget.

I'm also running a public challenge: $0 to first paying customers in 60 days, no ad budget. Every Tuesday I publish real numbers on LinkedIn — traffic, signups, revenue (or lack of it). No filters.

Would love feedback — especially around browser compatibility and the field-mapping approach. What breaks? What's annoying? What would you do differently? That's the useful stuff.

TypelessForm is built by Webappski. The npm package is typelessform-widget. Demo: typelessform.com.

Keywords: voice input for forms, speech to form, form autofill with voice

DEV Community: Alex Isa

How to audit a free AI visibility score with six manual checks

What does an AI visibility score measure?

Which evidence fields should you require?

How do you choose three control prompts?

How do you run the six checks?

How do you calculate the control result?

How do you compare a tool with your control?

What result differences are acceptable?

How do you repeat the test without lying to yourself?

What should you do before trusting the score?

How do you validate your own parser?

How can you automate without losing the receipt?

We Asked 3 AI Answer Systems Which AEO Agencies They Cite. Here Is the Reproducible Method

What exactly did we measure?

How did an agency enter the shortlist?

What did the six-cell audit return?

Why do the cited URLs matter as much as the names?

What can this method prove?

Why will a rerun produce different answers?

How should a buyer use the shortlist?

What was the uncomfortable result for Webappski?

What should the next run test?

Best Answer Engine Optimization Agencies in Poland and DACH (2026 Guide)

What Is Answer Engine Optimization (AEO)?

Why AEO Matters Differently in Poland and the DACH Region

The Poland and DACH AEO Agencies Compared

1. Delante (Krakow, Poland) — SEO Agency With an AI Search Offering

2. Elephate (Poland) — eCommerce SEO, No Dedicated AEO

3. SearchGPT Agentur (Berlin, Germany) — AEO-Native Agency

4. PerformanceLiebe (Hamburg region, Germany) — AEO From a Link-Building Heritage

5. Digitalagentur38 (Braunschweig, Germany) — Regional AEO Specialist

6. Webappski (Gdynia, Poland) — Per-Engine AEO With a Measured, Open Tracker

Poland and DACH AEO Agencies: Comparison Table

How Should a Polish or DACH Company Choose an AEO Agency?

Do they optimize for each AI engine separately?

Can they show measured results, not just claims?

Do they work in your buyers' language?

What is their technical depth?

Frequently Asked Questions

Which AEO agencies serve Poland and the DACH region in 2026?

How does AEO differ for a Polish or German company versus an English-speaking one?

How much does Answer Engine Optimization cost in Poland and DACH?

How do I know if an agency does real AEO and not relabeled SEO?

Can I measure my own AI visibility before hiring anyone?

Conclusion: Start With Your Language and Your Baseline

The Web Speech API will not give you multilingual form filling — here is the pipeline that will

Why webkitSpeechRecognition is not the answer (alone)

The pipeline an international form actually needs

Build vs buy, honestly

Voice form filling for e-commerce: what 'one-shot' actually means (and how to wire it up)

The three interaction models (this is the whole sort)

Who is actually in the category (June 2026, from their own public pages)

The wiring

How to evaluate without believing anyone's marketing (including mine)

How to check whether AI recommends your site — the honest AEO audit I run for clients

1. Install and point it at the client's domain

2. Run it — sampled, not one noisy shot

3. Read the report — every number is click-to-reveal

Why this beats a closed score

I built an open-source CLI that tells you if ChatGPT cites your brand — and what to do about it

TL;DR — three commands

What it actually measures

Measuring is the easy half — and where most tools stop

A real before/after (on our own product, no hypotheticals)

What the generated plan actually reads like

When you should NOT use this

Run it on your own brand

Free Open-Source AEO Tracker: Our Real Score Was 33/100

TL;DR

Why I didn't pay for Profound, Otterly, or Peec.ai

Three commands from install to HTML report

How does the two-model cross-check work?

What does a real AEO tracker output look like?

How to run an AEO tracker on your own brand

Why is this open source instead of a SaaS?

FAQ

How does it compare to Profound?

How much does it cost to run?

What does a 33/100 score mean?

Why `webkitSpeechRecognition` is not the answer (alone)