The growth arm of the toolchain, and the one I'm building now: an engine that does the work of an SEO specialist automatically. Most SEO is manual, repetitive, and follows well-worn patterns — which makes it exactly the kind of work software should be doing.
The problem
SEO is one of the highest-return growth channels there is, and most of the day-to-day is grunt work. Crawl the site. Run the same technical audit. Check the same list of known issues. Group pages by type so a fix applies everywhere. Watch the rankings. Notice what slipped. Do it again next week, and after every change. It's tedious, it's easy to be inconsistent, and it doesn't scale with a solo operator's time.
But here's the thing: the patterns are completely well established. Good SEO isn't a mystery — the rules for what a healthy page looks like, what to check, and how to fix the common problems are all known. The hard part was never knowing what to do; it was doing it consistently, across every page, forever. That isn't judgment work. It's exactly the kind of well-patterned, repetitive work that can now be automated.
My approach
Treat SEO as a system, not a series of manual projects. If the patterns are established, codify them: turn the specialist's checklist into automated checks that run continuously, across every page, without getting bored or missing one. Connect a site once, then let software do what a good specialist would do — every day, not just on audit day. Keep the human for the handful of genuinely strategic calls; hand the relentless, repetitive execution to the machine.
Like the rest of the toolchain, it starts narrow and real, then grows: detect first, then move up the chain from finding issues to drafting the fixes.
What I'm building
An SEO engine. You connect a site, and it crawls every page and groups them into templates — so a single fix applies to a whole class of pages instead of one at a time. It runs the technical audits a specialist would run by hand — performance, health, and the signals search engines report back — then checks each page against an established rulebook of known issues and flags what's wrong. It tracks how the numbers move over time, so you can see whether a change actually helped, and it surfaces what changed and what to fix next.
The heavy lifting runs continuously in the background, on its own. It's built to detect first; the next step is layering AI on top to move from finding an issue to drafting the fix — turning the engine from a diagnosis tool into one that does the work.
Why it matters
SEO is a compounding channel, and compounding rewards consistency — precisely what people are worst at and software is best at. Automating the established-pattern work turns SEO from a periodic manual chore into an always-on system that never skips a week. It's the three arms of the toolchain working together: the research engine finds the opportunity, Distro pushes the work out, and this makes sure it actually gets found.
What I'm learning
The clearest lesson is a pattern I keep seeing across the whole toolchain: well- established rules plus relentless, repetitive execution is the exact shape of work AI is now good at. The edge was never knowing SEO — plenty of people know SEO. The edge is applying it consistently, at a scale and cadence no person would keep up. Build the system that does that, and the shared knowledge everyone has stops being the thing that decides who wins.