<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Eioki</title>
    <description>The latest articles on DEV Community by Eioki (@eioki_dev).</description>
    <link>https://dev.to/eioki_dev</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4246181%2Fe57d373d-aab0-4a84-97bd-a16032ec7809.png</url>
      <title>DEV Community: Eioki</title>
      <link>https://dev.to/eioki_dev</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/eioki_dev"/>
    <language>en</language>
    <item>
      <title>How I Automated SEO Competitor Outlines by Combining Google SERP Data, Async HTML Crawling, and Gemini Structured Outputs</title>
      <dc:creator>Eioki</dc:creator>
      <pubDate>Sun, 15 Nov 2026 15:52:25 +0000</pubDate>
      <link>https://dev.to/eioki_dev/how-i-automated-seo-competitor-outlines-by-combining-google-serp-data-async-html-crawling-and-2dih</link>
      <guid>https://dev.to/eioki_dev/how-i-automated-seo-competitor-outlines-by-combining-google-serp-data-async-html-crawling-and-2dih</guid>
      <description>&lt;p&gt;Writing SEO content that actually ranks requires analyzing what already sits on Page 1—their heading structures, word counts, FAQ coverage, and schema types. Doing that manually across 5–10 tabs for every keyword is slow, and asking an LLM to "write an outline" without live SERP data produces generic fluff.&lt;/p&gt;

&lt;p&gt;To automate this research step, I built an &lt;strong&gt;Apify Actor (&lt;code&gt;SERP &amp;amp; Competitor Content-Gap Analyzer&lt;/code&gt;)&lt;/strong&gt; that fetches top Google results, crawls their live HTML concurrently, measures structural metrics, and uses &lt;strong&gt;Gemini Flash&lt;/strong&gt; to generate an outrank blueprint.&lt;/p&gt;

&lt;h3&gt;
  
  
  How the Pipeline Works
&lt;/h3&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Live SERP &amp;amp; People Also Ask Extraction&lt;/strong&gt;: Queries Google search results for your target keyword, country (&lt;code&gt;gl&lt;/code&gt;), and language (&lt;code&gt;hl&lt;/code&gt;), capturing organic rankings and &lt;code&gt;People Also Ask&lt;/code&gt; questions.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Concurrent Lightweight Page Analysis (&lt;code&gt;httpx&lt;/code&gt; + &lt;code&gt;BeautifulSoup&lt;/code&gt;)&lt;/strong&gt;: Crawls top competitor URLs asynchronously without headless browsers, extracting exact word counts, &lt;code&gt;H1&lt;/code&gt;/&lt;code&gt;H2&lt;/code&gt; heading hierarchies, JSON-LD &lt;code&gt;Schema.org&lt;/code&gt; types, and FAQ presence.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Structured Content-Gap Blueprint (Pydantic + Gemini Flash)&lt;/strong&gt;: Outputs &lt;code&gt;searchIntent&lt;/code&gt;, &lt;code&gt;keywordDifficultyEstimate&lt;/code&gt;, &lt;code&gt;recommendedWordCount&lt;/code&gt;, &lt;code&gt;competitorBlindSpots&lt;/code&gt;, and a 6–10 heading &lt;code&gt;outrankH2Outline&lt;/code&gt;.&lt;/li&gt;
&lt;/ol&gt;

&lt;h3&gt;
  
  
  Try It Out
&lt;/h3&gt;

&lt;p&gt;👉👉 &lt;a href="https://console.apify.com/actors/5A385sfmBRkAqsido/info/readme?build=latest" rel="noopener noreferrer"&gt;&lt;strong&gt;SERP &amp;amp; Competitor Content-Gap Analyzer on Apify Store&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;

</description>
      <category>seo</category>
      <category>ai</category>
      <category>automation</category>
      <category>python</category>
    </item>
    <item>
      <title>How I Built a Self-Qualifying B2B Lead Engine with Google Places API, Lightweight Tag Scanning, and Gemini Flash</title>
      <dc:creator>Eioki</dc:creator>
      <pubDate>Sun, 15 Nov 2026 15:52:15 +0000</pubDate>
      <link>https://dev.to/eioki_dev/how-i-built-a-self-qualifying-b2b-lead-engine-with-google-places-api-lightweight-tag-scanning-and-214i</link>
      <guid>https://dev.to/eioki_dev/how-i-built-a-self-qualifying-b2b-lead-engine-with-google-places-api-lightweight-tag-scanning-and-214i</guid>
      <description>&lt;p&gt;If you have ever built a local B2B lead-generation workflow in &lt;strong&gt;n8n&lt;/strong&gt;, &lt;strong&gt;Make&lt;/strong&gt;, or Python, you probably know the exact pain of "scraper chaining":&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Run a Google Maps scraper to get business names and websites.&lt;/li&gt;
&lt;li&gt;Pipe those URLs into a second tool to extract emails and social links.&lt;/li&gt;
&lt;li&gt;Run a third check to see if the business actually needs your service (e.g., missing GA4, no Meta Pixel, no online booking widget).&lt;/li&gt;
&lt;li&gt;Send everything to an LLM to write a non-generic cold email opener.&lt;/li&gt;
&lt;li&gt;Realize you paid three times for duplicate franchise locations across neighboring zip codes.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;To eliminate that friction, I designed a single-pass, API-first pipeline—packaged as an &lt;strong&gt;Apify Actor (&lt;code&gt;Local Business Lead &amp;amp; Tech-Gap Analyzer&lt;/code&gt;)&lt;/strong&gt;—that handles extraction, deduplication, tech-stack auditing, and structured AI scoring in one step.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Official Data First (No Brittle Headless Browsers)
&lt;/h3&gt;

&lt;p&gt;Instead of spinning up heavy Playwright/Puppeteer instances that break every time a DOM class changes, Step 1 queries the &lt;strong&gt;Google Places API (New)&lt;/strong&gt; (&lt;code&gt;places:searchText&lt;/code&gt;) using strict &lt;code&gt;X-Goog-FieldMask&lt;/code&gt; headers so only required fields are fetched.&lt;/p&gt;

&lt;p&gt;Before doing any downstream processing or billing, records pass through a two-key deduplication gate (&lt;code&gt;placeId&lt;/code&gt; + normalized root domain). If &lt;code&gt;"Austin, TX"&lt;/code&gt; and &lt;code&gt;"Round Rock, TX"&lt;/code&gt; return the same clinic, the duplicate is dropped immediately.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Fast, Async Website Tech-Gap Scanning (&lt;code&gt;httpx&lt;/code&gt; + &lt;code&gt;BeautifulSoup&lt;/code&gt;)
&lt;/h3&gt;

&lt;p&gt;Using &lt;code&gt;httpx.AsyncClient&lt;/code&gt; with a strict 6-second timeout and 350 KB HTML cap, a lightweight scanner inspects each business homepage for concrete conversion and analytics signals:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Analytics &amp;amp; Ad Pixels&lt;/strong&gt;: Checks for &lt;code&gt;gtag(&lt;/code&gt; / &lt;code&gt;G-&lt;/code&gt; (GA4), &lt;code&gt;googletagmanager.com/gtm.js&lt;/code&gt; (GTM), and &lt;code&gt;fbq(&lt;/code&gt; (Meta Pixel).&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Online Booking Friction&lt;/strong&gt;: Detects Calendly, Acuity, Square Appointments, Mindbody, OpenTable, Resy, Zocdoc, and direct &lt;code&gt;/book&lt;/code&gt; CTAs.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Contact &amp;amp; Social Surface&lt;/strong&gt;: Extracts clean public email addresses and social profiles.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  3. Structured Bottleneck Scoring with Gemini Flash
&lt;/h3&gt;

&lt;p&gt;The combined business profile, recent reviews, and website tech-scan JSON are passed to &lt;strong&gt;Gemini Flash&lt;/strong&gt; using Pydantic-enforced &lt;code&gt;response_schema&lt;/code&gt; JSON output to generate a 1–5 bottleneck score, concrete weaknesses, and a tailored 2-sentence cold outreach hook.&lt;/p&gt;

&lt;h3&gt;
  
  
  Try It Out
&lt;/h3&gt;

&lt;p&gt;👉 &lt;a href="https://console.apify.com/actors/5A385sfmBRkAqsido/info/readme?build=latest" rel="noopener noreferrer"&gt;&lt;strong&gt;Local Business Lead &amp;amp; Tech-Gap Analyzer on Apify Store&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>automation</category>
      <category>n8n</category>
      <category>python</category>
    </item>
  </channel>
</rss>
