I wanted to know what makes a page turn visitors into customers, so I watched Matthew Stafford of Build Grow Scale on Edward Sturm's show. Stafford says his team has tested and mentored online stores for about 10 years, with about 300 stores in their network. The episode is called Why Landing Pages Don't Work, and most of it is about making pages convert. One caution before the lessons: almost all of his tests are on stores that sell physical products. My clients are local service businesses that want calls and form fills. So I read his lessons as ideas to try, not as proof.
Ask the people who almost did not buy
He starts with one question on the thank you page: "What is the one thing that almost made you not buy?" He says about 30% of buyers answer it. He credits it with six of their ten biggest wins in four to five years.
He also sends a survey after each purchase, with a gift card for answering within 48 hours. He says those buyers come back within 7 days and are worth "four or five times" the average customer.
For a local business, the same question fits after a job is booked. Ask the customer what almost stopped them from calling. Their answer tells you what your page is missing.
Clear beats clever at the top of the page
Stafford says a large share of visitors never scroll. He gave no source for his figure, so I take it as a rough idea. Still, it means the top of the page does most of the work.
His points on that top section:
- The home page builds trust and helps people find what they came for. It is not the place for the hard sell. A product page needs a headline, bullets and good images at the top.
- Plain headings win. He calls it the "Homer Simpson check." Made-up product names made people skip parts of the site. Plain, descriptive headings won his tests. Benefits go in the bullets and the description, not the H1.
- Trust signals depend on the buyer. Bigger purchases and older buyers need more of them. Copying another site's layout will not copy its results.
For a bath remodeler, a heading like "Tub to shower conversion" tells a visitor more than a clever brand name for the same job.
Pages for search are separate from pages for ads
Stafford says an H1 stuffed for SEO hurts the people reading it. Sturm adds his own view here. Pages built for search should be their own pages, one per keyword, with the keyword in the web address, title, meta description, H1 and first sentence. The page you send paid traffic to "is not for SEO," he says.
Stafford adds that SEO text can sit lower on the page, in sections that open when tapped, below the part that sells.
Sturm makes two more claims of his own:
- Search pages for buyers at the end of their search convert far above the 2% Stafford calls normal, because the searcher already knows what they want. He gave no numbers.
- A weak top of the page sends people straight back to Google, and he believes that costs rankings. He also passes on Bill Hartzer's view that engagement matters more than backlinks. That part is secondhand.
Down the page: reviews, FAQs and the phone number
These are the tests he says won further down the page:
- Repeat a winning element in more than one place. People move through a page in different ways. His team kept adding it until the lift stopped.
- No false bottom. Let the top of the next image show at the bottom of the screen, or add an arrow, so people know there is more.
- FAQs at the bottom. Buyer questions as headings, answers in plain text. He says these work well for search.
- Two star reviews get clicked most and convert best, if the business answered them well. A page of only five star reviews is not trusted.
- A reason next to each form field. His examples: "Email required for order confirmation" and "phone required for shipping notifications." He says field errors fell to about a third.
- A visible phone number lifts conversion, even though few people call. Put it near the buy button, with reviews that answer common worries.
- Say "we pay shipping," not "free shipping."
- The successful stores all have a forgiving return policy and good customer service.
Most of these carry over to a service business with small changes. A reason next to the phone field could be as simple as saying you will text to confirm the estimate time. Answering a two star review well, in plain words, may do more for a page than another five star review.
Traffic is usually not the problem
Stafford says most sites convert around 2%, so the first fix is usually the page, not more visitors. He believes his lessons are psychology, so they apply to any product, including software and info products. That is his opinion. His tests are on stores, and a lead form is a different action from a checkout.
One of his numbers does not match the figures I could find. He said online sales reached 33% of all sales at the peak. US Census figures put the peak near 16%, in 2020, so he may have used a different base. That is one more reason I treat his numbers as his own experience, not as facts to quote.
How we use this at eezyRank
Sturm's point about separate pages is the core of what we build. We build one page for each specific search a buyer types: a page for each service, each town the business really serves, and the town and service pairs that make sense, with question pages under them.
Here is where Stafford's lessons show up:
- Plain headings. Our pages are built around the words people search for the job, not a clever brand name.
- Search text below the selling part. When a page shows up for a search but sits below the top spots, we can add a short section built around that search. It goes below the page's main copy, not in front of it.
- Questions buyers really ask. The questions on each page come from what people really search.
- Reviews where they matter. A client's real reviews can be placed on the service and town pages they relate to.
The rest is in the owner's hands. Keep the phone number easy to see, answer the bad reviews well, and ask new customers what almost stopped them from calling.
You can see what we build for local service businesses at https://eezyrank.com.





