Last week I talked about the dashboard tax — the friction of logging into six platforms before lunch. But there’s a deeper problem than the logins.
Even if every dashboard loaded instantly, you’d still be looking at fragments.
Google Search Console shows impressions and clicks. Analytics shows what people did after they arrived. Your ranking tool shows positions. Your email shows what the client said last Tuesday. Your website’s code shows what actually changed on the page. Each one gives you a piece. None of them talk to each other.
The diagnosis lives in the gap between your tools, not inside any one of them.
Fragments Don’t Diagnose
Here’s what a typical SEO investigation looks like. Rankings have dropped on a key page. You log into your ranking tool — positions are down, confirmed. You switch to Search Console — impressions dropped three weeks ago. You switch to Analytics — bounce rate went up around the same time. You check your email — the client mentioned updating some copy five weeks ago.
Four tabs. Four logins. Four sets of data. And you’re the one stitching them together in your head, trying to figure out whether the copy change caused the impressions drop, or whether Google’s algorithm shifted, or whether a competitor published something better, or whether it’s something else entirely.
That’s not diagnosis. That’s detective work with half the evidence scattered across different rooms.
The problem isn’t that the data doesn’t exist. It’s that it lives in silos that were never designed to connect.
What One Flow Looks Like
When I investigate a ranking shift for a client, I don’t open four tabs. I pull the data into one place — and the thing that makes that possible is Claude Code.
Claude Code is a development tool from Anthropic. It runs in the terminal — no browser, no dashboard, no login screen. What makes it different from a chatbot is that it can reach into other systems. It can pull Search Console data through Google’s API. It can check keyword rankings through DataForSEO. It can read the version history of a client’s website and show me exactly what changed on any page, when, and why. It can search through client emails to find the conversation where a change was discussed.
One tool. One session. Every data source the diagnosis needs.
I don’t reconstruct the timeline from memory or from four different dashboards. I describe what I’m looking for, and the system assembles the picture from the data that’s already there. Search Console shows impressions dropped on the 4th. The site history shows a heading was changed on the 2nd. The email trail shows the client requested that change three weeks earlier. The ranking data confirms the positions shifted a week after the edit.
What this actually changes
The connection between cause and effect becomes visible — not because I’m smarter than someone using SEMrush, but because the data is in the same room for the first time. A ranking tool can tell you positions moved. Claude Code can tell you positions moved, cross-reference that against what changed on the page, check whether the content that was supposed to be deployed actually was, and surface the email where the change was first discussed. That’s the difference between monitoring and diagnosis.
This didn’t require a team of developers to build. It required using a tool properly — connecting it to the right APIs, giving it access to the right data, and building the context around it so that when I ask “what happened to this client’s rankings?” the answer draws from everywhere it needs to.
The SEMrush Problem
I think the money spent on SEO tools is usually paying for monitoring and calling it diagnosis.
SEMrush, Ahrefs, Moz — they’re good at what they do. They track positions, they estimate traffic, they flag technical issues. But they can only see what Google sees from the outside. They can’t see what changed on your website. They can’t see the email where you asked your designer to “just tweak the homepage a bit.” They can’t see that a page rewrite was approved in January and still hasn’t been deployed.
When rankings drop, these tools tell you that it happened. They don’t tell you why. The why lives in the combination of ranking data, search console data, site changes, and the conversation trail between you and whoever manages your site.
If your SEO provider’s investigation starts and ends inside a ranking tool, they’re monitoring, not diagnosing.
Why This Matters Now
A year ago, connecting all of these data sources would have meant building custom software or hiring a team. The APIs existed — Google’s Search Console API, ranking data providers, version control systems — but the glue between them was expensive. You’d need a developer to write the integrations, a dashboard to display the results, and someone to maintain all of it.
Claude Code changed the economics of that completely. One tool that can talk to any API, read any data source, and synthesise the results in plain English. The connective tissue that was missing between your ranking tool and your search console and your site history — it exists now, and it doesn’t cost six figures to implement.
I’ve been in this industry for twenty-six years. For most of that time, the answer to “what changed?” was a shrug. Not because people didn’t care, but because connecting the data genuinely was hard. It isn’t hard anymore. The tool exists. The question is whether your provider is using it.
The businesses that recover fastest from ranking drops aren’t the ones with the best tools. They’re the ones where the data connects — where someone can see the search performance, the site changes, and the conversation trail in one sitting and find the thread.
The Thread
This is the third week running on the same theme. Week 10: knowing what to work on first. Week 11: removing the friction between you and your tools. This week: connecting the tools so the picture is whole.
Three weeks. One argument. A business that can see itself makes better decisions than one that can’t. See what needs doing. See the tools without friction. See the connections between them.
The expensive problems I’ve seen over twenty-six years almost always come down to the same root cause: someone couldn’t see what was happening in their own business. The data existed. The tools existed. The connection between them didn’t.
Tony Cooper
We Build Stores
tony.cooper@webuildstores.co.uk
07963 242210
P.S. If your current SEO setup involves logging into one tool to check rankings, another to check search performance, and then trying to remember what changed on the site last month — that gap is where the answers live. Reply to this email and I’ll show you what it looks like when all of that is in one place.
