Blog

By Jared DeValk · Last reviewed 2026-09-08 · Local SEO

The March 2026 Core Update Probably Didn't Hit You. Here's Why It Still Matters.

If you run a local service business and your traffic didn't move in March, that's the expected outcome. The update landed hardest on affiliate sites, comparison content and template-generated pages — categories most plumbers and law firms aren't in. Reported losses in the affected categories ran from around 20–35% at the mild end to well over half for some sites.

But the reason those pages lost is worth understanding, because it's the same reason most local content strategies quietly fail. It just fails slowly instead of overnight.

What actually got hit

The pattern across the reporting was consistent:

Lost badly Held or gained
Template comparisons written without using the products First-hand testing and original measurement
Mass-produced "best X for Y" listicles Narrow, deep topical authority
Pages assembled from other people's articles Original data nobody else has
Thin location pages differing only by city name Genuine local substance
High-volume AI-generated content Content with an identifiable author and point of view

The common thread isn't AI, and it isn't affiliate links. It's whether the page contains anything that only its author could have written.

A template comparison of two CRMs isn't bad content. It's accurate, readable, well-structured — and identical in substance to forty other pages produced by people who also read the vendor's pricing page. There's nothing in it that required the writer to exist.

Why this matters to a local business

You're probably not publishing listicles. But the same test applies to the content you do publish, and most local sites fail it:

Location pages that swap the city name. "Plumbing in Franklin" as a find-and-replace of "Plumbing in Brentwood" is the local version of a template comparison. It's also the oldest doorway-page pattern there is, and it carries real manual action risk on top of the ranking problem.

Service pages assembled from competitors' service pages. If your HVAC maintenance page could belong to any HVAC company in Tennessee, it contains nothing.

Blog posts summarising industry news you didn't cover. Someone else already wrote it, usually better and earlier.

The honest test: could a competitor have published this exact page without changing anything but the logo? If yes, it will not earn a ranking or an AI citation, whatever else is true about it.

What survived, and what that costs

The pages that gained had things that are genuinely hard to produce:

  • First-hand experience. Actually running the thing you're writing about
  • Original data. Numbers from your own operation that don't exist elsewhere
  • Narrow authority. Being demonstrably deep on a small topic rather than shallow on a wide one
  • Real attribution. A named person with a reason to be credible

None of that scales, which is the point. It's expensive precisely because it can't be mass-produced, and that's what makes it durable.

This is why our software section covers seven categories instead of forty — we only write about tools we've actually deployed for paying clients, and that rules out most of the market. It's also why our location pages stop at the cities where we have something real to say, and why the research section has one report in it instead of five.

Fewer pages that pass the test beat more pages that don't. That was true before March and it's now expensively demonstrable.

What to actually do

If your traffic dropped: identify which pages lost, and be honest about whether each one contained anything original. Adding word count to a page that says nothing produces a longer page that says nothing.

If it didn't: don't take that as permission to start producing volume. The threshold moved; the pages that would have been marginal two years ago are now the ones that lose.

Either way, the practical move is the same. Pick your ten most important pages and put something in each that only you could write — a number from your own jobs, a photograph of your own work, a thing you got wrong and fixed, a genuine opinion about a common practice in your trade. That's slower than publishing forty pages, and it's the only version that holds.

More on this: SEO for service businesses and how to get found in AI search.

Sources


Written by Jared DeValk, founder of Nashville Digital. Published September 8, 2026.