.d53b12e5.png&w=3840&q=75 "Trivas.ai")
by Trivas.aiWe just pulled the trigger on killing one of our data dictionary pages. Here's the thing: /resources/data-dictionary/cancellation-rate pulled fewer than 10 impressions and zero clicks over the last three months. Zero. That's not a slow page, that's a page nobody is finding, and it's time to talk about why it's going away and where the content actually ends up.
Let's just sit with the numbers for a second. Under 10 impressions. Zero clicks. Three full months of data. For a page that exists purely to define one metric, that's about as close to invisible as you can get in Search Console.
Here's what a lot of people get wrong about pages like this: they assume a quiet page is a neutral page. It just sits there, doing nothing, costing nothing. That's not how Google evaluates a site. A standalone glossary entry with near-zero engagement is a signal, and not a good one. Thin content, low dwell time, and intent that overlaps with a dozen other pages on bigger sites all tell Google something about the quality of the section it lives in. One weak page rarely tanks a whole domain, but a pattern of them drags down how the crawler treats your other resources, too.
To be clear, cancellation rate isn't an unimportant metric. DTC brands, especially subscription ones, live and die by it. The problem was never the topic. It's that a bare, isolated definition page is the wrong shape of content for how people actually search for this stuff. Nobody types "cancellation rate" into Google expecting a 200-word definition in isolation. They're comparing it to churn, refund rate, return rate, trying to figure out which metric tells them what. A standalone page can't answer that. A guide can.
The original page wasn't bad, just thin. It defined cancellation rate, gave the formula (cancelled orders divided by total orders), and explained why DTC brands, particularly subscription-based ones, track it closely.
There were a couple of things worth keeping. It had a short note on formula variations for subscription models versus one-time purchase models, since "cancellation" means something different depending on whether you're talking about a recurring subscription getting cancelled or a one-time order getting cancelled before fulfillment. It also had a rough benchmark range for healthy cancellation rates in subscription ecommerce, which is the kind of detail people actually want when they land on a metrics page.
None of this content is wrong. It's accurate, it's clean, it's just too small to carry its own URL. A page needs enough depth and context to earn a spot in the index on its own merits. This one never had that, and the traffic numbers back it up.
We looked at two options. Noindex the page and leave it live for internal reference, or merge it into a broader page and 301 redirect the old URL.
Noindexing felt like the lazy choice. It stops the page from competing with better content, sure, but it also just leaves a dead end sitting on the site. Anyone who still links to it, internally or externally, hits a page that Google has effectively given up on. That's not a great experience, and it wastes whatever crawl equity the URL has built up.
So we went with merge and redirect. The cancellation rate definition is getting folded into a broader ecommerce metrics guide, as one section among several related metrics: churn rate, refund rate, return rate. Instead of four or five separate thin pages competing for scraps of search volume, you get one stronger page that actually covers the category people are researching.
Merge and redirect wins here because it consolidates topical relevance onto a single page instead of scattering it. Google doesn't have to decide which of five near-identical pages is the "real" answer for cancellation rate questions. There's one answer, and it's a good one.
The destination is a consolidated metrics guide living under /resources/guides-reports. It covers cancellation rate alongside churn, refund rate, and return rate, since these four metrics get confused with each other constantly and brands usually need to see them side by side to understand the difference.
We're not copy-pasting the old definition in as a section. It's getting rewritten from scratch as a labeled subsection inside the guide, with its own header so it's still linkable and skimmable, just contextualized next to the metrics it's most often compared against. The formula stays, the subscription-versus-one-time distinction stays, the benchmark range stays. What changes is the framing: instead of standing alone, it's positioned as one piece of a bigger "which metric should I actually be watching" conversation.
Any internal links that used to point at the standalone page are getting updated to point at the new section anchor inside the guide. That part matters more than people think. A redirect handles external traffic and bookmarks fine, but internal links pointing at a stale URL that then bounces through a redirect chain is just sloppy, and it adds unnecessary hops for both users and crawlers.
If you're trying to figure out how cancellation rate fits next to other metrics you're already tracking in your dashboards, that's exactly the kind of cross-metric view our BI reporting tooling is built to surface automatically, rather than making you piece it together from five different glossary pages.
Here's the actual rollout, step by step:
None of this is complicated. It's just easy to skip steps, and skipped steps are how you end up with broken internal links six months later pointing at a page that no longer exists.
A data dictionary with dozens of single-metric pages, each pulling single-digit impressions, isn't a resource. It's a thin content pattern, and Google treats it that way. Every near-empty glossary entry you leave live is a small vote that your resources section doesn't have much depth, and that can suppress crawl budget and rankings across pages that actually deserve to rank.
Consolidating low-traffic definitional pages into fewer, stronger pillar pages does the opposite. Instead of diluting topical authority across ten thin URLs, you concentrate it into one page that can actually compete for the comparison-style searches people run. One strong page beats five weak ones, almost every time, for this type of content.
This won't be the last page we prune this way. The cancellation rate definition was one of several candidates sitting in our "merge or kill" bucket for manual content review, and more data dictionary entries are going through the same audit in the coming weeks. If a page can't earn its own clicks, it doesn't get to keep its own URL.
If you want to see how we're restructuring the rest of the data dictionary as this continues, it's worth keeping an eye on our guides and reports section, or just subscribing so the bigger consolidations land in your inbox instead of you having to check back manually.
explore more insights
3 min read
3 min read
3 min read