What we won't sell you (and why)
The local marketing industry sells small businesses a lot of things that don't survive contact with the evidence. Here are four we've taken off the menu — with the receipts, so you can hold anyone to them.
When you publish your prices — ours are here — people ask why some familiar line items are missing. No "conversion rate optimization." No "weekly Google posts for SEO." No pages for every suburb in the valley. That's not an oversight. Each one is missing because we looked at the evidence and couldn't sell it to you with a straight face.
1. A/B testing on your website
A/B testing is real science — at scale. The problem is arithmetic. To reliably detect a 20% improvement in conversions at a typical small-business conversion rate, you need roughly 28,000 visitors through a single test. A local service business gets a few hundred to a few thousand visits a month. One valid test would take anywhere from nine months to several years.
It gets worse. Run the test anyway — as some agencies do — and statistics plays a cruel trick called the winner's curse: an underpowered test that reaches "significance" has, by construction, wildly overshot. In simulations of a typical small-business site, a 3-month test reported an average +66% lift for a change whose true effect was +10% — and about one "winning" result in nine actually pointed at the losing variant. That's the report an agency proudly emails you. The number is noise.
Don't take our word for it — the A/B-testing vendors themselves say this. Convert.com, which sells testing software, recommends around 1,000 conversions per month before testing makes sense. CXL, the best-known conversion-optimization trainer, puts the floor at the same place and tells low-traffic businesses to spend their effort elsewhere.
What we do instead: fix the known conversion killers (click-to-call above the fold, speed, reviews, a short form), watch real users, and put the effort into growing your traffic — which, at this size, is the actual constraint.
2. Google Posts as an SEO lever
Plenty of packages include "weekly Google Business Profile posts" sold as a ranking booster. The only controlled test we know of — Sterling Sky tracked 441 keywords across nine weeks of posting — found the effect on rankings was zero. Not small. Zero. Average click-through on the posts themselves was about half a percent.
We still keep profiles fresh, because a complete, current profile converts better when someone's already looking at it. But we won't charge you for a ranking effect that a controlled experiment says doesn't exist.
3. A page for every suburb
"Plumber in Sandy." "Plumber in Murray." "Plumber in West Jordan." Same page, twelve times, with the city name swapped. Google's own spam policy has a name for this — doorway abuse — and since the March 2024 updates it has been deindexing sites for it. The agency that built those forty pages got paid either way; the client is the one who disappears from search.
We build a city page when there's something true and specific to say about that city — real jobs done there, real photos, real service notes. Otherwise we don't, and we'll tell you why.
4. Lock-in
Read the terms of service of the big small-business website providers and you'll find some version of this: cancel, and you lose the website. The domain is sometimes registered in their name. The "affordable monthly plan" is a lease with a repossession clause nobody mentioned on the sales call.
Our version, stated plainly: your domain and your Google profile are in your name from day one. On our monthly plans, the site itself becomes yours outright at a stated month — printed on the pricing page, not buried in a contract. Cancel any time before that and you owe nothing; you just don't take the site with you. Buy it outright and it's yours on day one.
The point
None of this makes us anti-marketing. It makes us pro-arithmetic. A small business's money should go where the evidence says the leverage is: a fast site that gets you called, a complete Google profile, real reviews from real customers, and — when you're ready to buy traffic — ads measured by cost per booked job, not impressions.
If someone pitches you one of the four things above, ask them for their evidence. If they have a controlled study, we'd genuinely love to read it.
Want the version of this applied to your website?
Run your site through our free report card — Google's own tests, no sales pitch attached — or see our published pricing.