Google Search Console has been telling me something for weeks and I finally got around to actually building for it.
A while back I noticed one query showing up over and over in my impressions: "flexitarian vs intuitive eating diet." Then "flexitarian vs AIP diet." Then a dozen more variations, all comparing one diet against another. Nawtch had individual pages for a bunch of diets already, but nothing that actually answered a "diet vs diet" question. So I built Diet Arena.
What it does
Pick any two diets and get a side by side comparison. Same categories for both, no verdict on which one is better. Typical foods, foods to avoid, eating pattern, macro emphasis, who it's targeted at, how strong the actual evidence is behind it, that kind of thing.
The part that took real thinking
With 15 diets, that's 105 possible pairs. I went back and forth on whether every single pair should get its own real page, or whether I should only build out the ones I actually had search data for and let everything else fall back to some generic dynamic view.
I ended up going all in. Every pair gets a real page, same template, same level of care. Part of that was practical (Astro can generate all 105 from one template and one data file, so it's not actually 105x the work), but part of it was just that the point of this thing was never only "capture search traffic." I also wanted an actual tool people could use to compare any two diets, and a two-tiered system where some comparisons get the full treatment and others get a lesser one didn't sit right with either of those goals.
To make that work without rewriting the same information 105 times, I had to go back through my diet pages and pull everything into a structured format first. Name, typical foods, eating pattern, evidence strength, all of it, the same shape for every diet. Once that existed, generating every pair was just running a build step, not writing 105 pages by hand.
How it's actually doing so far
Honestly mixed, and I think that's worth being upfront about instead of only posting when everything looks good. Out of 105 pairs, about 10 are actually indexed by Google right now. Those are the ones I manually requested indexing on through Search Console, the same shortlist I'd already linked from the "popular comparisons" section on the picker page, since those are the pairs I actually have real search demand for.
Everything else is sitting in "discovered, not currently indexed," which is Google's way of saying it knows the page exists and just hasn't gotten around to crawling it yet. Manually requesting indexing on 95 more pages isn't really practical either. There's a daily limit on how many you can push through that way, and doing it in bulk for a batch of similar new pages doesn't actually change why Google deprioritized them in the first place.
That's the real tradeoff of going all in on all 105 instead of a curated dozen. A newer site only has so much crawl attention from Google at any given time, and dumping 105 pages on it at once means most of them are just going to sit in a queue for a while, regardless of what I manually push through. The rest will get crawled eventually on Google's own schedule. I'd rather have a complete, honest tool now and wait on the indexing than fake completeness with a worse version of the feature.
If you want to poke at it
Nawtch's Diet Arena, no signup needed. Compare whatever two you're curious about.
Top comments (0)