rtcstats.com is a new initiative I started recently with Olivier and Philipp. It is genuinely useful and powerful as hell. Here’s the thing though… it sits in a tiny little niche of a niche: observability inside WebRTC inside VoIP inside communications. And in there, how do you even know it exists, besides maybe reading the stuff I write here on my blog?
I set out figuring out what needs to be done about this problem of mine, and along the way jumped head first into the rabbit hole of AEO, GEO, AI, agentic search and how developers really work in 2026.
TL;DR - The WebRTC tools guide now lives on my website.
A useful tool nobody could find
I am not new to getting sites the visibility they need. This website has high DR (Domain Rating) - I’ve worked and put effort to get there and now to maintain it. You need good writing, getting others to backlink to your site, and do SEO work. There’s an ongoing grind here, where bit by bit your scoring improves.
So I started doing the same for rtcStats. No big deal. We were at DR=7 when we started this initiative 2 months ago or so, now at DR=20, with lots of room to grow (and the actions to take next).
Here’s the thing. I made use of my own understanding and experience to do it, coupled with my trusted Claude Code environment, which is highly skilled at the things I need it to.
That Claude Code piece changed quite a bit how I now look at things…
I replaced half my own stack without noticing
When I started really putting the effort into using Claude, it was in creating a system that works for me and gives me more capabilities on an ongoing basis. Not coding skills - analysis skills that are useful for how I operate and how I assist clients in consulting.
What I noticed, is that I started replacing my own internal technology stack to fit what Claude needed:
- If something didn’t have APIs? It went out the window without a second thought
- Had an API but not for the things I needed it to do? I showed it the door
- Is the API there and working? Got it embedded even further in how I work
That meant saying goodbye to the trusted Feedly service I used to track my news all over the world, migrating to Inoreader without a second thought. It took me about a week to fully migrate, but it made my news reading habits work better with the creation of WebRTC Weekly and WebRTC Insights. It also enabled me to capture ideas and concepts a lot faster than ever before.
Notion which I happily worked with for over a year now felt slow and useless compared to using local md files in an Obsidian Vault. So we parted ways. I now run a simple task management, note taking and poor-man’s-crm inside that Vault.
It didn’t happen intentionally at first. There was no grand plan to get there. It just quietly evolved to where it is now.
Then I noticed something interesting. I now use around 10 more tools than I used before. Tools I never heard of. With some, the only visit I had to their website was to create an account and grab an API key to use (Claude did the rest). Sometimes I even shared my credit card with these services.
I didn't find any of it the old way
If I haven’t heard of these tools before, but now I am using them and happy about it within a span of a few days or weeks, how did that happen?
I asked Claude. That’s what happened.
It went about searching for whatever it is my Claude COO agent does when I ask him such questions, and came back with answers and a suggestion. That suggestion was usually a list of a few tools, or even a single tool.
Sometimes, I asked him to research APIs and interfaces, like the case of switching from Feedly to Inoreader.
Replacing my WordPress theme builder with pure Gutenberg blocks was a long session with actual experiments done to figure out what will work.
Things like adopting Vercel and Resend were just Claude telling me what it is going to do next and me approving.
The whole loop of discovery, comparison and decision changed drastically from how I did things just a year ago.
If it works that way for me, how does that work for our developer customers at rtcstats.com?
So we changed what "get seen" means
That meant that the important SEO work we’re doing at rtcStats wasn’t enough. It needed a fine tuning of AI friendliness.
We had to start looking also at visibility that doesn’t mean ranking higher on Google, but being legible to agents looking at what we do and how we do it. Enabling and empowering them to tinker around and use the service on behalf of potential human users - most of them developers or at least literate in engineering.
It nagged me for the past few months, when we started off with our SEO initiative. Over a month ago, we set out to improve things around for AEO and GEO. And from there towards looking at the next step - the usage of our service by agentic AI and not just visibility.
I’ve learned a lot about how this works. Synthesized it based on blogs read online about the topic, what my trusted Claude environment researched, surfaced and suggested. And the patterns and behavior changes I started noticing in my own daily work.
This was uncharted work for me, which is always the fun part. It also led me to the following realization 👇
Then I put my BlogGeek.me hat on
The analyst in me got curious. If I am doing it for rtcstats.com, I wanted to understand where others in our industry are positioned. What is happening to the rest of the WebRTC developer ecosystem?
I already had a nice infographic in place, but it was static and uninspiring, showing its age.
I asked myself:
What does discovery look like and how does it work in the WebRTC ecosystem?
Not in rtcStats, for one tool. But for us as an industry.
One thing led to another, and I decided to revamp and recreate the WebRTC developer tools ecosystem, just not as an infographic, but as a WebRTC tools hub instead.
The WebRTC tools hub
I listed all the tools in the ecosystem, starting off from the last update of the infographic. Added a few. Removed others. The usual market shifts that occurred.
Then split them into categories. Slightly different than the ones in the original infographic, since I needed an analysis to run inside each category in a meaningful way.
Then I decided that besides stating the categories and the tools, I am not going to do much else.
From there on, my work was to guide Claude and ask him what he sees (my COO agent that I interact with in Claude is named Adam), and then create a well documented methodology in place and have him run the show with searches, queries and analysis.
This isn’t a leaderboard. It isn’t my view of things (I added that in places as tiny thought bubbles). It is how AI sees our market from a bird’s eye view. Which tools are discoverable, which ones will it raise to answer questions, what will he find searching a specific tool when asked and will it be able to figure things out on its own.
Go look at the map
The WebRTC tools guide now lives on my website.
Go check it out.
If you build or maintain one of these tools and your row is factually off, tell me and I will fix it.
Now that I have a methodology and a system in place, I am likely to update this on a monthly or quarterly cadence.
As always - feedback and suggestions are welcome!
