Do martial arts schools on page one offer a free class? 59 homepages, measured

Ask anyone who markets martial arts schools what a homepage needs, and the first answer is almost always the same: a free trial class, front and center. The reasoning is that parents shopping for a program compare two or three schools, and the one that lets them try it without paying wins the visit. Plenty of owners assume every school ranking above them already does it.

We checked. On September 18, 2026 we captured the first two pages of Google for “martial arts classes” in 12 US cities, from Spokane to Jacksonville, screened out everything that was not a martial arts school, and read the homepage of every school that remained: 59 schools, every one of them on page one.

The belief does not hold. Only 22 of the 55 homepages we could read offer a free class. Another 16 lead with a paid or discounted intro offer instead, and 17 make no new-student offer on the homepage at all.

The benchmark: 3 of 6 basics

What the 55 readable page-one homepages carry:

Homepage element	Schools	Share
Mentions a kids or youth program	50	91 percent
Mentions adult classes	48	87 percent
Shows or links to a class schedule	46	84 percent
Makes any new-student offer	38	69 percent
Has a contact or signup form	33	60 percent
Links to a blog	30	55 percent
Offers a free class, lesson or week	22	40 percent
Mentions pricing or tuition, or links to it	17	31 percent
Has testimonials or reviews from students	14	25 percent

To locate yourself in one number, count six of these on your own homepage: an offer of any kind, a schedule, pricing or tuition, testimonials, a form, and a blog link. The median page-one school has 3. Seventy-five percent have at least 3, 49 percent have at least 4, and 15 percent have 5. None has all six. Four puts you ahead of half of page one.

The sites themselves are small. The median page-one school has 31 pages in its sitemap, 40 percent have fewer than 25, and 88 percent have fewer than 100. Most have no blog to speak of: 18 of the 27 schools that link to a blog from the homepage have 1 blog page or none in their sitemap.

The free class: common, not universal, and leaning toward the top

The 22 free offers are a trial class, a first lesson, a week, or a private lesson. Sixteen more schools make an offer that is not free. The most common wording was a “web special” or “limited time offer,” a discounted starter package advertised only online. The other 17 make no new-student offer on the homepage that we could find.

So a school without a free class is in the majority of page one, not behind it. Pricing is the same: 31 percent mention pricing or tuition, with no relationship to position (a rank correlation of -0.001 across 55 schools).

The free class does lean toward the top of the page. It appears on 10 of the 17 readable homepages in positions 1 to 3, 7 of 19 in positions 4 to 6, and 5 of 19 in positions 7 to 10. As a rank test that is -0.298 across 55 schools, which only just clears the significance bar, and it is one of eleven feature tests. One clearing the bar by chance would not be surprising, so treat it as a lead.

Two features lean the same way more firmly. A class schedule appears on 16 of 17 top-three homepages against 13 of 19 at positions 7 to 10 (-0.335), and adult classes on all 17 top-three homepages against 13 of 19 (-0.401). Site size moves the same way: the median school in positions 1 to 3 has 52 pages, positions 7 to 10 have 23 (-0.37 across 48 sites with a readable sitemap). Those four probably describe one thing: schools with a fuller website. Nothing here shows that a free-class offer affects ranking on its own.

“Best in town,” and the numbers behind it

We also counted two kinds of claim on each homepage: superlatives nobody can check (“best,” “premier,” “top rated,” “#1”) and numbers someone could (“25 years,” “500 students,” “120 black belts”). Ages are not counted as claims.

The median page-one school uses 2 superlatives and 0 checkable numbers. Fifty-five percent make no numeric claim at all, and 9 of 55 use three or more superlatives with no number behind them. Neither count tracks position: -0.098 for superlatives and -0.067 for numbers, across 55 schools. A sample this size can only detect a strong relationship, so this is not proof the copy does nothing. It does mean page one is not won by the loudest claims.

The count reads words, not intent: the second-highest score, 25, belongs to a school whose name contains the word “best.” The median is unaffected.

What page one actually holds

A school owner looking at page one is not looking at ten competitors. In the 10 cities where the capture worked, page one held a median of 6 schools, and about a third of the results were something else: Yelp listings in 9 cities, Reddit threads in 5, a lessons directory ranking “the 10 best” classes in a city in 5, three regional kids’ activity guides, and seven YMCA, community center, city recreation and university club programs. Those programs rank for “martial arts classes” but are not schools, and their homepages describe the whole organization, so we excluded them.

So your real competition on page one is about six schools, not ten, and several of the other results are places a school can appear inside: a Yelp profile, a directory listing, a parents’ activity guide.

What verification changed

This study was designed to compare page one with page two, as our roofing study did. Verification ended that design.

Page two failed to return for 3 cities and came back empty for a fourth. Where it did return, none of its 78 results was a martial arts school. They matched single words of the search instead: videos about school buses and yoga, a travel guide to the city, a software handbook’s page on “classes,” a game guide’s character classes, state government pages. Page one for Sacramento and Omaha had the same problem, returning songs and pages about “California” and “arts.” A person searching there would almost certainly see something different. We treated all of it as a failed capture, not as evidence about page two, and did not spend more searches on it. The study became a page-one benchmark in 10 cities.

Checking every homepage’s offer wording by hand changed three things. The first pass missed four real offers worded as “Try A Class,” a four-week trial, a new member special, and a second month free; the pattern was widened. A classroom app slipped through the relevance screen because its name contains “dojo”; it was removed. And our online booking measure, built on appointment wording for dentists and contractors, found 0 of 55 schools. A measure that finds nothing is usually measuring the instrument, so booking was dropped, which is why the score is out of six.

Limits: what this cannot tell you

Every school here ranks on page one. The study cannot say what separates them from schools that do not rank, because none of those are in the data. With 59 schools, only strong relationships are detectable, and a result that is not significant is not evidence of no effect.

We read homepage text only. An offer shown in an image, a popup or a script-built banner is not counted, so the offer rates are a floor. Two homepages did not respond and two returned too little readable text to measure; they are reported as unmeasured, not as schools without these features. We did not measure Google Business Profile strength, reviews, backlinks, domain age or proximity to the searcher, which are likely to explain most of the ordering.

Results were captured once, from one location setting per city, for one wording of the search. One school held two page-one positions in its city and is counted once. The 127 excluded results are in the dataset with their reasons.

The number to write down

The median school on page one for “martial arts classes” shows 3 of 6 basics and uses 2 superlatives and no checkable numbers, and 60 percent of those schools offer no free class. More than half make no checkable claim at all. A schedule, an offer and one real number is a short list, and most of page one has not finished it.

The full dataset, with every exclusion and its reason, is below.