DEV Community: John Builds

We A/B tested value-first onboarding (generate before subscribe) — here is what happened

John Builds — Tue, 12 May 2026 13:23:12 +0000

We've been running an A/B test on our onboarding flow and the early results are interesting enough to write up.

The old sequence

Voice setup → connect account → subscribe → generate posts. Classic SaaS — get commitment early, then show value.

The problem: users were dropping at the subscribe step having never seen a single piece of output. They were being asked to pay for something they'd only read about.

The value-first sequence

Pick topics → generate 5 posts in your voice → review them → THEN connect and subscribe.

By the time they hit the paywall, they've already seen their content calendar for next week. Drop-off on the subscribe step is lower. Makes sense — it's a lot easier to say yes to a tool when you're already holding the output.

Why it works

The subscribe step isn't a commitment to something abstract anymore. It's "do you want to keep the posts you just made?" That's a very different ask.

The implementation details

A few things worth noting if you're building something similar:

We run the sequence via a PostHog feature flag (onboarding-v2, variant value-first) — easy to roll back or adjust the split without a deploy
The full-tab onboarding (not a popup) was intentional — popups close if you click away, which kills any flow that involves a file upload or multi-step form
Preferences sync locally first (chrome.storage), then backend — so the extension is usable even if the API is temporarily unreachable

Full breakdown of both sequences and the design decisions behind each step: https://xreplyai.com/blog/xreplyai-onboarding-wizard?utm_source=devto&utm_medium=social&utm_campaign=blog-2026-05-12

Best AI Tools for Social Media in 2026 (Ranked and Reviewed)

John Builds — Mon, 11 May 2026 19:56:15 +0000

AI has changed how people manage social media — but not all tools are doing it the same way. Some generate generic templates. Some are X-only. Some charge $99/mo for features you can get elsewhere for $19. This is an honest breakdown of the best AI social media tools in 2026, ranked by use case.

What to look for in an AI social media tool

Before comparing tools, it helps to know what actually matters. Most AI social media tools do one or more of these things:

Draft content — generate posts from a topic or prompt
Schedule and publish — queue posts across platforms on a calendar
Generate replies — draft responses to comments and mentions
Voice matching — learn your tone so content sounds like you, not a template
Analytics — track what performs and why

No single tool does all of these equally well. The right choice depends on which combination matters most for your situation.

The best AI social media tools in 2026

#1 XreplyAI — Best for solo founders and creators who want AI that sounds like them

Platforms: X, LinkedIn, Instagram, Threads, YouTube, Pinterest, Bluesky, TikTok

Pricing: Pro $49.99/mo · BYOK from $19/mo · 7-day free trial

Pros:

Voice profile trained on your own posts — drafts sound like you wrote them
BYOK plan: pay your AI provider directly, no markup
AI reply generation + scheduling in one tool
8 platforms with per-platform formatting
MCP integration for power users

Cons:

Newer product — smaller community than legacy tools
No native analytics dashboard yet

Best for: Solo founders building in public, creators who post on multiple platforms, anyone tired of generic AI content

#2 Typefully — Best for X/Twitter and LinkedIn writers who want a clean writing experience

Platforms: X, LinkedIn

Pricing: Free plan available · Pro from $29/mo

Pros:

Best-in-class thread composer for X
Clean, distraction-free writing UI
AI writing assist built in
Good LinkedIn scheduling

Cons:

Only 2 platforms (X and LinkedIn)
No voice matching or training on your past posts
No reply generation

Best for: Writers and thought leaders focused on X and LinkedIn

#3 Hypefury — Best for X-focused creators who want automation and engagement tools

Platforms: X (primary), Instagram

Pricing: From $65/mo

Pros:

Auto-retweet, auto-plug, auto-DM features
Engagement automation built in
Strong X-native feature set
Good inspiration feed

Cons:

Expensive at $65/mo
Primarily X-focused — limited multi-platform support
AI content is template-based, not voice-trained

Best for: X power users who want engagement automation

#4 Tweet Hunter — Best for X creators who want analytics and a content library

Platforms: X

Pricing: From $49/mo

Pros:

Large library of high-performing tweet examples
Good tweet analytics
AI tweet generation with examples
Engagement features

Cons:

X-only
Expensive for what you get vs newer alternatives
No voice matching

Best for: X creators focused on growth with data-driven content decisions

#5 Buffer — Best for teams and agencies who need simple multi-platform scheduling

Platforms: X, LinkedIn, Instagram, Facebook, Pinterest, TikTok, YouTube, Mastodon

Pricing: Free plan · Essentials from $6/mo per channel

Pros:

Free plan available
Excellent multi-platform scheduling
Clean team collaboration tools
AI assistant for post drafting

Cons:

AI is generic — no voice training
No reply generation
Per-channel pricing adds up for many platforms

Best for: Teams and agencies who need reliable scheduling across many platforms

#6 Taplio — Best for LinkedIn-only creators who want AI content and CRM features

Platforms: LinkedIn only

Pricing: From $49/mo

Pros:

LinkedIn-native AI content generation
Built-in LinkedIn CRM and lead tracking
Good content inspiration from viral posts
Carousel and document post support

Cons:

LinkedIn only
Expensive for a single-platform tool
No voice matching

Best for: LinkedIn-focused B2B founders and consultants

The BYOK difference

One thing worth understanding before you pick a tool: most AI social media tools are charging you a markup on AI. You pay $49–99/mo, and part of that covers their Gemini or ChatGPT API costs. The tool bundles the AI in so you don't have to think about it — but you also can't see what you're actually paying for AI vs the software.

XreplyAI is the only major tool with a BYOK (Bring Your Own Key) plan. You connect your own Gemini or ChatGPT API key, pay the AI provider directly, and the actual AI cost for most solo founders is $1–5/mo. The tool costs $19/mo. Total: roughly $20–24/mo vs $49–99/mo elsewhere.

Whether that matters depends on how much you use AI generation. If you're generating a lot of content daily, the savings compound quickly. If you use the AI occasionally, the difference is smaller.

How to choose

If you want...	Use
AI that sounds like you across 8 platforms	XreplyAI
Best X/Twitter thread writing experience	Typefully
X engagement automation (auto-RT, auto-DM)	Hypefury
Data-driven X growth with content library	Tweet Hunter
Team scheduling across many platforms	Buffer
LinkedIn-only content + CRM	Taplio

FAQ

What is the best AI tool for social media in 2026?

It depends on your use case. For solo founders and creators who want voice-matched content across multiple platforms, XreplyAI is the strongest option. For pure scheduling, Buffer. For LinkedIn, Taplio. For X-focused growth, Hypefury or Tweet Hunter.

What is BYOK and why does it matter?

BYOK (Bring Your Own Key) means you connect your own AI API key instead of paying the tool's markup. With BYOK, you pay the AI provider directly — often $1–5/mo in actual usage — instead of $49–99/mo to the tool. XreplyAI is the only major social media tool with a BYOK plan.

Can AI social media tools write in my voice?

Most generate generic content from a prompt. XreplyAI trains on your own past posts to learn your tone, vocabulary, and sentence structure so drafts sound like you, not a template.

Which AI social media tools work for multiple platforms?

XreplyAI supports 8 platforms. Buffer supports most major platforms. Typefully covers X and LinkedIn. Hypefury is X-focused. Taplio is LinkedIn-only.

Are there free AI social media tools?

Most offer free trials rather than permanent free tiers. XreplyAI has a 7-day free trial and a BYOK plan from $19/mo. Buffer has a free plan with limited scheduling. Most others are paid-only with trials.

Originally published at xreplyai.com.

Auto-DM new Twitter/X followers: how we built it, and the API wall blocking Instagram and LinkedIn

John Builds — Fri, 08 May 2026 14:53:26 +0000

I shipped Auto DM replies in XreplyAI this week. Here's how it works under the hood, and why bringing it to Instagram and LinkedIn is a much harder problem.

The X implementation

The core loop is straightforward: poll for new followers, diff against a snapshot, send DMs to net-new ones. The details that matter:

First run snapshots up to 1,000 existing followers so you don't blast your whole audience on day one
Each subsequent cycle polls the 100 most recent followers and compares against the snapshot
Rate-limited to 15 DMs per cycle — Twitter's DM rate limits are easy to hit if you're not careful
Template supports a {name} placeholder; everything else is plain text (500 char limit)
All sends logged with status: pending / sent / failed

The tricky edge: if a user has 50,000 followers, the snapshot only captures 1,000. Anyone outside that window who was already following before first run could theoretically get a DM. We flag this in the UI and let the user decide.

Why Instagram and LinkedIn are a different problem

Instagram DMs require instagram_manage_messages — a permission that goes through manual Meta app review. The review checks your automation use case against their policy. Outcome isn't guaranteed, timeline is weeks, and the reviewer feedback loop is slow.

LinkedIn is the real wall. Automated DMs to followers or post engagers requires the Community Management API. It's invite-only, no public application, and requires a LinkedIn partnership contact to even get into the queue. I've been trying to find the right entry point for weeks.

If you've navigated either of these — especially LinkedIn's partnership process — I'd genuinely love to compare notes in the comments.

X Auto DM is live now on Pro and BYOK plans: https://xreplyai.com?utm_source=devto&utm_medium=social&utm_campaign=feature-2026-05-08

Why AI content sounds generic (and the fix isn't better prompts)

John Builds — Wed, 06 May 2026 18:49:00 +0000

Most AI content sounds the same. Here's why — and the fix that actually works.

I spent weeks editing AI drafts for 20 minutes each just to make them sound like me. Then I realized the problem wasn't the output. It was that the tool had never read anything I'd written.

Generic AI tools don't know:

Whether you write long and dense or short and punchy
Whether you use humor or stay completely dry
Whether you lead with data or lead with a story
What your actual opinions are on anything in your niche

So they produce the median. Technically correct, completely forgettable, impossible to attribute to any specific person.

The fix isn't better prompts

Better prompts help at the margins. But you're still fighting the same underlying problem: the model is generating from a blank slate every time.

The thing that changed my output quality: training on my own archive first.

Feed it 200 of your own posts. Let it build a model of how YOU actually write — your sentence length, your humor level, your tendency to lead with data vs. story, how abstract or concrete you tend to be.

Now when it generates, the starting point is calibrated to you. Not to the average of everything it was trained on.

What this looks like in practice

Before: AI gives me a draft that's 60% there. I spend 15–20 minutes editing to remove the generic phrasing, add my actual opinions, restructure the sentences to match my rhythm.

After: AI gives me a draft that's 85–90% there. I change a word or two. Done in 30 seconds.

The difference isn't the model quality. It's whether the model knows how you write before it starts.

How I built this into XreplyAI

This is the core of the voice profile feature in XreplyAI. You upload your Twitter/X archive (or LinkedIn/Instagram export), it analyzes your posts across six writing dimensions — formality, expression, density, humor, assertiveness, abstraction — and every post or reply it generates starts from that model.

The editing experience is fundamentally different when the tool already knows your voice.

If you've been fighting AI drafts to make them usable, the issue is almost certainly that the tool doesn't know how you write yet.

How to Format Text in LinkedIn Posts (Bold, Italic, Strikethrough)

John Builds — Fri, 01 May 2026 15:41:31 +0000

LinkedIn does not have a native formatting toolbar for feed posts. You cannot press Cmd+B to bold a word in a standard LinkedIn post — but you can get bold, italic, strikethrough, and more using Unicode characters that LinkedIn renders just fine. Here is exactly how to do it.

Why LinkedIn formatting works this way

LinkedIn's post composer is a plain text field. It does not interpret HTML or markdown — asterisks stay as asterisks, and underscores stay as underscores. The formatting workaround uses Unicode Mathematical characters: a separate block of characters in the Unicode standard that look identical to bold or italic versions of the Latin alphabet but are technically different characters.

When you paste 𝗯𝗼𝗹𝗱 text into LinkedIn, LinkedIn is rendering standard Unicode — the same standard it uses for emoji and other special characters. No tricks, no hacks, no risk of your post being penalised. It is how most high-performing LinkedIn posts get their formatting.

How to bold text in a LinkedIn post

Go to the LinkedIn Text Formatter tool.
Type or paste the text you want to format in the input box.
Click Copy next to the Bold style row.
Paste the result directly into your LinkedIn post composer. It renders as bold text on desktop and mobile.

The same steps work for italic, bold-italic, and strikethrough — just select the corresponding style row before copying.

Available formatting styles

Style	Example	Best used for
Bold	𝗛𝗲𝗮𝗱𝗹𝗶𝗻𝗲 𝘁𝗲𝘅𝘁	Post openers, key phrases
Italic	𝘌𝘮𝘱𝘩𝘢𝘴𝘪𝘴 𝘵𝘦𝘹𝘵	Quotes, emphasis within copy
Bold Italic	𝙎𝙩𝙧𝙤𝙣𝙜 𝙚𝙢𝙥𝙝𝙖𝙨𝙞𝙨	CTAs, key takeaways
Strikethrough	o̶l̶d̶ ̶t̶h̶i̶n̶k̶i̶n̶g̶	Myth-busting posts, before/after

Where LinkedIn formatting works

Feed posts — the most common use case. Bold openers and section headers increase scroll-stopping ability.
LinkedIn bio / About section — bold your job title, key skills, or a standout line. This is underused and makes bios scannable.
Comments — bold a key point in a comment to draw attention to your response in a long thread.
Captions on images and documents — the same Unicode trick works here.
Headline field — your LinkedIn headline is plain text, but Unicode bold makes key words stand out in search results and connection requests.

The one place it does not apply: LinkedIn Articles have a native rich text editor with real bold, italic, and heading support. Use those native controls for articles instead of Unicode.

Tips for using LinkedIn formatting effectively

Bold your first line. The LinkedIn feed shows approximately 2–3 lines before the "see more" truncation. A bold opener signals that the post is worth expanding.
Use formatting sparingly. A post where every other line is bold has no contrast. Reserve bold for the 2–3 most important phrases per post.
Numbers and symbols stay unformatted. Unicode mathematical blocks only cover A–Z and a–z. Numbers and punctuation render in your original style regardless of the format you select.
Test on mobile before posting. Unicode renders correctly on all platforms, but paste your test post into the LinkedIn app on your phone to confirm before publishing.

The quick way to format LinkedIn posts

The LinkedIn Text Formatter handles all of this in one step. Paste your full post, copy each section in the style you want, and reassemble in LinkedIn. It supports bold, italic, bold-italic, strikethrough, and more — free, no login required.

Frequently Asked Questions

How do you bold text in a LinkedIn post?

LinkedIn does not have a native bold button for posts. Use Unicode Mathematical Bold characters: type your text into a LinkedIn text formatter, select the Bold style, copy the result, and paste into your LinkedIn post or bio.

Can you italicise text in a LinkedIn post?

Yes, using the same Unicode trick. Unicode Mathematical Italic characters paste directly into LinkedIn posts, bios, and comments and render as italic text.

Does LinkedIn support markdown formatting?

No. LinkedIn posts do not render markdown. Asterisks do not create bold and underscores do not create italic. The only way to get formatted text in LinkedIn feed posts is to use Unicode characters.

Will Unicode bold text work in LinkedIn articles?

LinkedIn Articles have a native rich text editor with real bold, italic, and heading support — use those native controls instead. Unicode formatting is only needed for feed posts, bios, comments, and captions.

Does bold text in LinkedIn posts affect the algorithm?

There is no confirmed algorithmic penalty for Unicode formatted text. Bold characters are widely used in high-performing LinkedIn posts and can improve dwell time by making posts more scannable.

Originally published at xreplyai.com.

How we built auto-DM for new X followers without making it feel like spam

John Builds — Thu, 30 Apr 2026 15:37:45 +0000

Auto-DM tools on X have a bad reputation. Most of them work like this: connect your account, write one message, blast it to everyone who ever follows you. The result is predictable — generic "thanks for the follow, check my link!" messages that feel like a bot sneezed on your inbox.

When I built this into XreplyAI, I tried to follow a tighter pattern:

First run snapshots up to 1,000 existing followers and sends nothing — no retroactive blasts
Subsequent runs poll your 100 most recent followers, diff against the snapshot, DM only the net-new ones
500-char customizable template with a {name} placeholder
Rate-limited to 15 DMs per cycle to stay inside X's API limits
Every send is logged (pending/sent/failed) so there's a full audit trail

What we're actually using it for: thank the person for following, share our brand values and what kind of posts they can expect from us, and mention something we're currently excited about. Closer to a welcome email than a promo blast.

Nobody has unfollowed after receiving one yet — which was the metric I was most nervous about.

It's live on BYOK ($19/mo — bring your own Gemini or OpenAI key) and Pro ($49.99/mo), with a 7-day free trial on both.

https://xreplyai.com?utm_source=devto&utm_medium=social&utm_campaign=feature-2026-04-30

What is BYOK (Bring Your Own Key)? — Test

John Builds — Tue, 28 Apr 2026 03:02:25 +0000

What is BYOK (Bring Your Own Key)?

BYOK stands for Bring Your Own Key. Instead of paying a SaaS tool for AI credits, you connect your own API key from OpenAI, Gemini, or Claude — and the tool uses your key directly.

This means you pay the AI provider's wholesale rate, not a marked-up subscription.

Why it matters for solo founders

Most AI tools charge $30–50/month on top of what you'd pay the AI provider anyway. With BYOK, you cut out the middleman.

FAQ

Is BYOK safe?

Yes — your key is encrypted at rest and never logged.

Which providers does XreplyAI support?

OpenAI (GPT-4), Google Gemini, and Anthropic Claude.

Originally published at xreplyai.com.

What I built this week: moving publish actions into the editor to fix a silent drop-off problem

John Builds — Wed, 22 Apr 2026 14:08:20 +0000

Sometimes the bug isn't in the code it's in the flow.

We have a thread editor in XreplyAI (Rails API + Next.js). It handles per tweet character limits, thread ordering, and multi-platform context. Users write threads for X, LinkedIn, Threads, all from one place.

The problem: after writing, they had to navigate to a separate screen to actually schedule or publish. One extra step. Invisible in analytics until I looked at completion rates by flow stage. Threads were being written but not posted.

The fix was straightforward once I diagnosed it: surface post now, schedule, and save draft directly in the editor. Conditional on having content in the first tweet. Button label updates based on whether a schedule time is already set.

The engineering challenge was keeping the component focused. The editor was already managing a lot of state. The solution was treating the publish actions as a read-only consumer of
existing editor state rather than adding new state to the component — the actions just observe what's already there and offer the right affordance.

Small change. Should have shipped it earlier.

If you're building content tools: audit where users are in the moment they need to act, and make sure the action is right there with them.

Why AI generated content sounds generic and the fix that actually worked for me

John Builds — Tue, 21 Apr 2026 17:59:32 +0000

Most advice about AI content says "write better prompts." That's not wrong, but it's missing the bigger issue.

The problem is that AI models are trained on massive, averaged datasets. When you prompt with "write a post for a software engineer who shares their journey," the output is statistically central it looks like the middle of all posts ever written for that audience. Competent. Forgettable.

What actually shifted the output quality for me: I stopped prompting and started training.

I fed the model 150+ of my own posts before asking it to generate anything new. Old tweets, LinkedIn updates, forum replies. Once it had enough examples of how I actually write the cadence, the opinions I repeat, the phrases I overuse the drafts stopped feeling like someone else's work.

Three things improved immediately:

The posts had actual positions
The sentence rhythm matched how I talk
I stopped rewriting every draft from scratch

I built XreplyAI around this exact idea it builds a voice profile from your own archive so content sounds like you wrote it, not like a template.

If you're a developer building in public and AI content keeps feeling off, this is worth trying before you conclude the tools aren't good enough. The tools are fine. The input is the problem.

DEV Community: John Builds

We A/B tested value-first onboarding (generate before subscribe) — here is what happened

The old sequence

The value-first sequence

Why it works

The implementation details

Best AI Tools for Social Media in 2026 (Ranked and Reviewed)

What to look for in an AI social media tool

The best AI social media tools in 2026

#1 XreplyAI — Best for solo founders and creators who want AI that sounds like them

#2 Typefully — Best for X/Twitter and LinkedIn writers who want a clean writing experience

#3 Hypefury — Best for X-focused creators who want automation and engagement tools

#4 Tweet Hunter — Best for X creators who want analytics and a content library

#5 Buffer — Best for teams and agencies who need simple multi-platform scheduling

#6 Taplio — Best for LinkedIn-only creators who want AI content and CRM features

The BYOK difference

How to choose

FAQ

What is the best AI tool for social media in 2026?

What is BYOK and why does it matter?

Can AI social media tools write in my voice?

Which AI social media tools work for multiple platforms?

Are there free AI social media tools?

Auto-DM new Twitter/X followers: how we built it, and the API wall blocking Instagram and LinkedIn

The X implementation

Why Instagram and LinkedIn are a different problem

Why AI content sounds generic (and the fix isn't better prompts)

The fix isn't better prompts

What this looks like in practice

How I built this into XreplyAI

How to Format Text in LinkedIn Posts (Bold, Italic, Strikethrough)

Why LinkedIn formatting works this way

How to bold text in a LinkedIn post

Available formatting styles

Where LinkedIn formatting works

Tips for using LinkedIn formatting effectively

The quick way to format LinkedIn posts

Frequently Asked Questions

How do you bold text in a LinkedIn post?

Can you italicise text in a LinkedIn post?

Does LinkedIn support markdown formatting?

Will Unicode bold text work in LinkedIn articles?

Does bold text in LinkedIn posts affect the algorithm?

How we built auto-DM for new X followers without making it feel like spam

What is BYOK (Bring Your Own Key)? — Test

What is BYOK (Bring Your Own Key)?

Why it matters for solo founders

FAQ

Is BYOK safe?

Which providers does XreplyAI support?

What I built this week: moving publish actions into the editor to fix a silent drop-off problem

Why AI generated content sounds generic and the fix that actually worked for me

ai #socialmedia #twitter