[{"data":1,"prerenderedAt":205},["ShallowReactive",2],{"writing-introducing-rankrecon":3},{"id":4,"title":5,"author":6,"authorImage":7,"body":8,"dateModified":185,"datePublished":186,"description":177,"editorsPick":187,"extension":188,"intro":189,"mainImageAltText":190,"mainImageCaption":191,"mainImageCredit":191,"mainImageId":192,"meta":193,"navigation":194,"ogImage":195,"path":196,"readingTimeMinutes":197,"schemaImages":198,"seo":199,"series":191,"shortIntro":200,"sitemap":201,"slug":202,"stem":203,"twitterImage":195,"__hash__":204},"writing/writing/introducing-rankrecon.md","Introducing RankRecon: A SERP analysis tool that runs locally in your browser","Daniel Betts","/images/writing/author-daniel-betts.webp",{"type":9,"value":10,"toc":176},"minimark",[11,16,35,48,65,69,72,75,78,81,84,94,97,107,111,114,117,120,123,133,136,139,142,146,154,158],[12,13,15],"h2",{"id":14},"what-is-it","What is it?",[17,18,19,24,25,29,30,34],"p",{},[20,21,23],"a",{"href":22},"https://www.missionsystems.co.uk/rankrecon/","RankRecon"," is a free browser extension for ",[20,26,28],{"href":27},"https://chromewebstore.google.com/detail/rankrecon/chfndljngecdigdblnfbajjbfmfcdbde","Chrome"," and ",[20,31,33],{"href":32},"https://addons.mozilla.org/en-GB/firefox/addon/rankrecon-serp-analysis-tool/","Firefox"," that you can use on any Google SERP for a few things:",[36,37,38,42,45],"ol",{},[39,40,41],"li",{},"Convert the SERP into a dataset that you can save to a file: all the details from the SERP, including URLs from the AI Overview and the Local Pack, plus a host of metrics taken directly from each site listed in the SERP.",[39,43,44],{},"Give you a programmatic read on the SERP - RankRecon uses signals from the data to determine the intent of each site, and the overall intent of the SERP. It looks for patterns in the SERP to give you a quick read on what is succeeding and why.",[39,46,47],{},"Convert the SERP into a text document you can paste straight into your AI model for review. If you want to use AI to break down a SERP for you, this tool gives it all the data it needs to do that.",[49,50,53,62],"figure",{"className":51},[52],"post-figure",[54,55],"img",{"src":56,"srcSet":57,"sizes":58,"loading":59,"decoding":60,"alt":61},"https://api.missionsystems.co.uk/api/images/11?w=800","https://api.missionsystems.co.uk/api/images/11?w=800 800w, https://api.missionsystems.co.uk/api/images/11?w=1200 1200w, https://api.missionsystems.co.uk/api/images/11?w=1600 1600w","(min-width: 768px) 720px, 100vw","lazy","async","RankRecon showing a completed analysis",[63,64,61],"figcaption",{},[12,66,68],{"id":67},"why-did-i-build-it","Why did I build it?",[17,70,71],{},"There are a few reasons, and they all come out of how I interact with the SERP when working for a client.",[17,73,74],{},"When a client approaches me, I first want to find out how they're ranking. Yes, I could use Ahrefs or equivalent, but for most of the smaller, local clients I work with, they just don't show up there in the Rank Tracker. So I need to go to the actual SERP and see them there. They're unlikely to be on page 1, though (that's why they came to me), so cue lots of manual clicking through SERP pages and scanning each one until I find them. I really don't like wasting my time on that stuff, so I thought: why not pull in as much of the SERP as I want to look at, so it's all sitting in Excel.",[17,76,77],{},"That dataset is also my historical record. Saved automatically with a timestamp, it's a snapshot of where they were, and where their competitors were, when we started out. As we work through the project, I can add further records and track their progress in the real SERP over time.",[17,79,80],{},"But what about the rest of that SERP? When a client initially approaches and I've found where they are (page 5), the job is to figure out how to get them to page 1. To do that, I need to get into who is on the first page and why. Again, for smaller, local SERPs, often that data is missing from Ahrefs. But also, where better to look than the actual SERP? So RankRecon is a tool that saves me manually clicking into 10 different sites and looking at page titles, H1s etc and holding it all in my head to piece together how sites rank in this SERP. Instead, it's a one-shot run and then I've got all the data in a spreadsheet.",[17,82,83],{},"Is there more to be gained by going and looking at the actual sites? Absolutely. But the tool was built on the 80/20 rule - it's 20% of the effort to get me 80% of the way there in understanding the SERP. The rest I can fill in after, if it warrants further investigation.",[49,85,87,92],{"className":86},[52],[54,88],{"src":89,"srcSet":90,"sizes":58,"loading":59,"decoding":60,"alt":91},"https://api.missionsystems.co.uk/api/images/9?w=800","https://api.missionsystems.co.uk/api/images/9?w=800 800w, https://api.missionsystems.co.uk/api/images/9?w=1200 1200w, https://api.missionsystems.co.uk/api/images/9?w=1600 1600w","The top of the Page 1 Analysis Report, showing the data summary and actions available",[63,93,91],{},[17,95,96],{},"As I went, I added other things I felt were useful: a map of the main blocks Google is showing on the SERP; an extraction of the AI Overview to see who is cited there and compare to the organic results; an extraction of the Local Pack data. And I've tried to build a multi-use tool that could fit many different workflows and hopefully is genuinely useful to anyone working with the Google SERPs. Have I succeeded? I don't know that yet! I could really use your help to try it out (see below).",[49,98,100,105],{"className":99},[52],[54,101],{"src":102,"srcSet":103,"sizes":58,"loading":59,"decoding":60,"alt":104},"https://api.missionsystems.co.uk/api/images/10?w=800","https://api.missionsystems.co.uk/api/images/10?w=800 800w, https://api.missionsystems.co.uk/api/images/10?w=1200 1200w, https://api.missionsystems.co.uk/api/images/10?w=1600 1600w","The Page 1 Analysis Report showing the detected patterns in the SERP and the \"Shape of the SERP\" map",[63,106,104],{},[12,108,110],{"id":109},"how-does-it-work","How does it work?",[17,112,113],{},"It's running in your browser, so a fair question before you install it is: what's it actually up to?",[17,115,116],{},"Let me confirm the important part straight away: it runs entirely in your browser. There's no server and no account, and nothing you search for ever leaves your machine.",[17,118,119],{},"When you run an analysis, RankRecon first reads the SERP itself. It pulls out the organic results and maps the structure around them: ads, videos, People Also Ask, the AI Overview (including who's cited in it) and the Local Pack (including each business's rating, category and website). You can configure a run to walk from 1 to 10 pages of results and it will stitch everything together into a single dataset.",[17,121,122],{},"Then it visits the sites. Each result is loaded in a background tab and measured: word count, headings (the actual H2 and H3 text, not just counts), internal and external links, images, schema markup, and a couple of dozen other structural facts. It filters out hidden elements, pop-ups and cookie banners, because a consent dialog with 200 headings in it will otherwise wreck your numbers (that's a real example, and it's why the filter exists).",[49,124,126,131],{"className":125},[52],[54,127],{"src":128,"srcSet":129,"sizes":58,"loading":59,"decoding":60,"alt":130},"https://api.missionsystems.co.uk/api/images/12?w=800","https://api.missionsystems.co.uk/api/images/12?w=800 800w, https://api.missionsystems.co.uk/api/images/12?w=1200 1200w, https://api.missionsystems.co.uk/api/images/12?w=1600 1600w","RankRecon running an analysis",[63,132,130],{},[17,134,135],{},"The report is computed from those measurements. Each page gets an intent classification (informational, commercial, transactional, navigational or local) from a weighted read of its schema, title, URL and structure, and the same signals are used to make a call on the SERP's overall intent. Pages are then clustered by shape - length, heading density, link density - into the content patterns you see in the report, with the pattern holding the top positions flagged as what's currently winning. Finally, it cross-references: which AI Overview citations also rank organically, and which Local Pack businesses show up in the organic results.",[17,137,138],{},"Under the hood, it's been built \"properly\", with over 90% test coverage. The analysis is pinned by a set of \"golden masters\": real SERPs I've captured and replayed through the full pipeline, with every conclusion locked in. If a code change alters any call the tool makes - an intent classification, a pattern grouping, a winner - it shows up as a visible diff before it ships. And because Google likes to quietly rearrange its markup, the extractors carry drift alarms: when Google rotated the AI Overview markup this month, the tool flagged it, and the fix took one saved SERP and an afternoon.",[17,140,141],{},"It's not perfect - it can't be. There are an infinite number of SERPs, and it's a procedural algorithm running in a browser extension. The aim is to give you a quick read and a feel for the SERP that would otherwise take much longer to get.",[12,143,145],{"id":144},"help-me-shape-it","Help me shape it",[17,147,148,149,153],{},"At the moment it's a tool I've built for myself. I'm hoping there's a useful core in it that fits other people's workflows, but I need help to shape it. If you have time to try it out, and to ",[20,150,152],{"href":151},"mailto:rankrecon@missionsystems.co.uk","give me feedback",", it would be greatly appreciated.",[12,155,157],{"id":156},"try-it","Try it",[159,160,161,166,171],"ul",{},[39,162,163],{},[20,164,165],{"href":22},"Home page for RankRecon",[39,167,168],{},[20,169,170],{"href":27},"Page on the Chrome Web Store",[39,172,173],{},[20,174,175],{"href":32},"Page on Firefox Add-ons",{"title":177,"searchDepth":178,"depth":178,"links":179},"",2,[180,181,182,183,184],{"id":14,"depth":178,"text":15},{"id":67,"depth":178,"text":68},{"id":109,"depth":178,"text":110},{"id":144,"depth":178,"text":145},{"id":156,"depth":178,"text":157},"2026-07-24T14:57:29.4768700Z","2026-07-23T15:00:14.5902069Z",false,"md","I kept doing the same manual job at the start of every client project: clicking through SERP pages to find where they actually rank, then clicking into the sites above them to work out why. RankRecon does that job in one run, entirely inside your browser.","RankRecon - Turn the SERP into data",null,8,{},true,"/api/images/8","/writing/introducing-rankrecon",5,[195],{"title":5,"description":177},"A browser extension that turns any Google SERP into a dataset, a structural read of who is winning and why, and an AI-ready brief. No server, no account, nothing leaves your machine.",{"loc":196},"introducing-rankrecon","writing/introducing-rankrecon","1BPg0HLgQSKyrGqmG-DbsiJrGTlwRNnMdJdd0eYFKh4",1784905143054]