Make the Ad Rep Prove It: The Prove-It Protocol for Buying Martial Arts School Advertising

Before you buy any advertising, price what a lead, an appointment and an enrollment already cost you in the channels you run today. That number is the bar. Then ask the seller to carry part of the risk on their own claim, run a small tracked pilot, and write the kill criterion before you write the check.

The session above is a group coaching call I run with school owners, and this article is built from the media-buying thread inside it. Every member name, school name, market and individual revenue or enrollment figure has been removed, and no publication, vendor or agency is named. The numbers used below are our coaching benchmarks and worked examples, not any one school’s books.

Three Pitches, One Week, Same Missing Piece

In a single call, three owners brought me three different offers.

One had a rep from a local coupon mailer in his lobby with a binder of martial arts ads from around the country. One had been approached by the advertising rep at a big-box gym selling proximity-based phone advertising — a beacon in the gym, ads served to those members’ phones for thirty days, a hundred thousand impressions guaranteed, eighteen hundred dollars for the year. One had a company on the phone promising to get her school found in voice search — ten keywords, a geo-targeted area, four hundred dollars.

Not one of the three pitches contained a single number about response. Impressions, distribution, demographics, guarantees, other schools who “seem really happy” — all present. Calls, appointments, enrollments, cost per any of those — all absent.

That is not an accident, and it is not usually dishonesty either. It is simply how media is sold. The seller is compensated on insertions, not outcomes, so the pitch is built out of the only inventory they control: reach. Your job is not to argue with the pitch. Your job is to change what the transaction is about.

I have bought a great deal of advertising since 1975, including plenty that did not work, and the discipline I want to hand you is a five-step sequence I call the Prove-It Protocol. It is not a way to avoid paid media — paid media is a permanent column in a healthy school’s marketing plan. It is a way to stop buying confidence and start buying tested response.

The Prove-It Protocol

Step	The question	What it produces
1. Price	What does a lead, an appointment and an enrollment already cost me?	Your own baseline, in dollars
2. Par	What would this channel have to beat to be worth owning?	A maximum bid per lead
3. Proof	Will the seller take any risk on their own claim?	A performance offer, or a real read on their confidence
4. Pilot	What is the smallest tracked version of this I can buy?	A dedicated number, a dedicated URL, a control offer
5. Pull	What result makes me stop, and when?	A written kill criterion and a capped loss

Run all five, in order, every time. Skipping step one makes steps two through five impossible, which is exactly why most owners skip it.

Step One — Price: You Cannot Evaluate an Offer Without Your Own Numbers

Here is the conversation that repeats itself in this industry constantly. An owner tries a coupon mailer — say ten thousand pieces for around $380 — and reports that it “didn’t work.” I ask what happened. Four calls and one enrollment.

Ten thousand pieces is not information. Four calls is information. Distribution is the seller’s number, not yours. It tells you how many pieces went into mailboxes, which is a fact about a printing press, not about your business. Response is your number.

And once you have your number, “didn’t work” becomes a calculation instead of a feeling. One enrollment for $380, at a premium school charging a $400 initial enrollment investment and roughly $375 a month, means you were cash-positive on that student inside the first thirty days. That is not a failed campaign. That is a campaign that paid for itself before the second tuition payment cleared.

One member on that call had sponsored a community charity walk for about $500 and enrolled one student out of it. He described himself as “down a hundred” until we walked it through: $400 collected at enrollment, plus $375 the following month, against a $500 cost. He was ahead by the end of the first billing cycle and enormously ahead over the life of the student. He had been about to cancel a channel that had actually just worked.

So step one is unglamorous bookkeeping, and it is the whole game:

• Cost per lead, by channel, for every channel you currently run
• Cost per booked appointment, which is cost per lead divided by your booking rate
• Cost per enrollment, which is the only number that connects marketing to money
• What you collect in the first thirty days from a new student — enrollment investment plus first month
• What a student is worth over their tenure — our target is $7,000 to $9,000, and never below $5,000

Most owners have none of these written down. If you have none of these written down, you are not qualified to evaluate any advertising offer, because you have nothing to compare it to. That is not an insult; it is a to-do list.

Step Two — Par: The Number the New Channel Has to Beat

Once you know your costs, you can compute what you are permitted to pay. This is the part almost nobody does, and it converts a vague “is this worth it?” into a hard ceiling you can defend in a negotiation.

Start with the acquisition budget. A new student should cost you roughly $150 to $300 in advertising and staff time to acquire — that is the band we coach to, and it reflects the fact that acquiring a student costs five to seven times what it costs to keep one. Then work backward through your own conversion ratios.

Here is a worked example using ratios a well-run school should be able to hit:

Stage	Count	Conversion
Raw leads from the channel	100	—
Booked appointments	45	45% of leads
Appointments that show and complete an intro	32	70% of booked
Enrollments	20	62% of intros

Twenty enrollments per hundred leads means five leads per enrollment. Now the ceiling falls out of the arithmetic:

Leads needed per enrollment	Max cost per lead at $150 acquisition cost	Max cost per lead at $300
5	$30	$60
8	$18.75	$37.50
10	$15	$30
15	$10	$20

Read that table twice, because it contains the single most important insight about paid media in this business: your conversion ratios set your maximum bid. The school that turns five leads into an enrollment can pay twice what the school that needs ten leads can pay, for identical traffic. Which means the fastest way to be able to afford better advertising is not to find cheaper advertising — it is to fix your follow-up and your enrollment conversion. That is a topic in its own right, and it is the reason I put more coaching hours into the appointment-setting process than into the ad copy.

Now put that ceiling next to the value. At roughly $375 a month with a well-coached school’s attrition under 2% a month, a student stays long enough that our planning figure for lifetime value is $7,000 to $9,000. A $300 acquisition cost against a $7,000 lifetime value is better than twenty to one, and it is recovered in cash inside the first month. Once you have internalized that, you stop flinching at the number and start asking the only sensible question: can this specific channel deliver leads at or under my ceiling?

One honest caveat, because I do not want to hand you a rule you will misapply. A single channel can sit above the $150 to $300 band and still be worth keeping — a scarce, high-intent source that produces ten enrollments a year at $400 apiece is not a problem. What must land inside the band is your portfolio average. Individual columns vary; the building has to stand.

Step Three — Proof: Ask the Seller to Carry Some of the Risk

Now for the step that gives this article its title.

Years ago I sat with a rep from a parenting publication with a very large local circulation. I had tried that book before and never gotten it to produce, though the same company’s publications in other cities were working for us. The rep told me it was a perfect audience and it was going to work great.

I asked how confident he was. Very confident, he said.

So I made him an offer: fine — then I will pay you $50 for every call that comes in, because that is our comparable cost in the other media we run. Put a dedicated tracking number in the ad. If you are right and it produces a flood, I will happily pay you twice what you were going to charge me for the space. I will even give you our best control piece so the creative is not the variable.

What happens next, almost every time, is that the confidence quietly evaporates and the rep finds somewhere else to be. Which is itself the result you were looking for. The rep who will not take a nickel of risk on their own claim has just told you the truth about how much they believe it — and they told you for free, before you spent anything.

I have had reps say yes. Those deals have been some of the better media buys I have made, because a seller with skin in the game suddenly starts caring about placement, timing and frequency. And note that the $50 figure was not plucked from the air. Look back at the ceiling table: at five leads per enrollment and a $300 acquisition target, $60 a lead is the ceiling and $50 is comfortably inside it. The performance offer only works if step two came first. Otherwise you are negotiating with a made-up number and you will not hold the line.

The Softer Versions of the Same Test

Not every seller can do pay-per-call, and some genuinely are not authorized to. There is a ladder of weaker but still useful versions:

1. Ask for the annual rate, then buy two months at it. Grandmaster Jeff Smith uses this one relentlessly: ask whether there is a discounted rate for a twelve-month commitment. They will offer a third off, or half off. Then say you will run two months at the annual rate, and if it produces you will commit to the full year. You have just bought the volume discount without the volume risk.
2. Ask for makegoods. If the insertion misses its date, misses its position, or under-delivers distribution, what do you get? Get the answer in the contract.
3. Ask for the position, not just the space. More on this in step four — position is often worth more than a discount.
4. Ask what they are actually doing for the money. For the $400 voice-search pitch, that question is decisive. Have them put the deliverables in writing and send it to you. Frequently the entire service turns out to be something you can do yourself in an afternoon, which we will get to.

Everything sold by a rep who physically shows up is negotiable — coupon mailers, community papers, radio, television, sponsorships. The rate card is an opening position. If you accept the rate card, you have volunteered to be the least sophisticated buyer they will talk to that month.

One Rough Heuristic Worth Holding Loosely

Here is a rule of thumb that is not literally accurate but produces the right posture: when someone walks into your lobby to pitch you a marketing channel, it is usually not a channel that works well. The genuinely productive channels rarely need a door-to-door salesforce, because they are already spoken for.

Hold that loosely — I have made money on rep-sold media and I have wasted money on things nobody sold me. But it should set your default from “interested” to “skeptical, and asking for numbers.”

And here is the related trap. The rep will show you a binder of ads from other schools who “seem really happy.” Those schools were not tracking either. Social proof without data is not data. You are looking at ads designed by owners who copied ads designed by owners who copied ads, none of whom measured a single call. That binder is a chain of guesses, and the rep is asking you to add a link to it.

Step Four — Pilot: Small, Tracked, and Placed Where It Can Be Seen

Assume the seller passed step three, or failed it in a way you can live with because the price is genuinely trivial. Now design a test that can actually produce an answer.

Tracking Is Not Optional

A media buy without a dedicated tracking number and a dedicated URL is not a test. It is a donation.

This is easier now than it has ever been. A tracked phone number costs a few dollars a month and forwards to your existing line. A dedicated landing page URL takes an hour. Use one number and one URL per channel — not per campaign, per channel — so that when the front desk takes the call you know exactly which line item produced it, without relying on a sixteen-year-old asking “how did you hear about us?”

Everything else about the pilot follows from wanting a clean read:

• One offer. My default for a cold audience is a stacked free offer — two free weeks, a free semi-private lesson with a black belt instructor, a free uniform — because free removes the price objection at the exact moment you are asking a stranger for a first step. If you want to test a paid trial against it, that is a second test, not a variable inside the first one.
• Your control creative. Use the ad you already know performs elsewhere. If you let the rep design the ad from the binder, you are testing their design department, not their audience.
• Enough volume to read. A test that can only produce three calls cannot distinguish between a good channel and a bad one. If the smallest buy they offer cannot plausibly generate ten or more tracked responses, either buy more or do not buy at all.
• A defined window. Start date, end date, decision date.

Placement Is Part of the Test

If you are buying print — a coupon vehicle, a community paper, a parent publication — the position you buy is frequently a bigger variable than the publication itself. Buy the back cover, the front cover, or the inside of either. Those are visible in the half-second between the mailbox and the recycling bin, which is the only impression most of these vehicles ever get.

A buried half page inside the book and the back cover are two different products that happen to be sold by the same person. If you can only afford the buried half page, you are not testing the audience; you are testing the worst version of the audience and will draw the wrong conclusion.

Two more print observations worth holding on to. Vehicles with editorial content generally out-pull pure coupon packs — the local paper with the high school scores and the heartwarming middle-school story gets read in a way an envelope of coupons does not. The exception is the grocery insert that goes into the daily paper, because a large share of subscribers buy the paper specifically for those ads. Readership is the thing you are buying; a coupon pack that people actually opt into beats a newspaper section they skip.

Applying This to the Digital Pitches

The proximity-and-beacon pitch fails step four on mechanism, before you ever get to price. Ask how the notification reaches the phone, which app has to be installed, and which settings the consumer must have enabled. In every version I have looked at, the answer requires a chain of conditions to all be true at once. And consider what a hundred thousand impressions is worth in a world where most people have dozens of apps on their phone and meaningfully use a handful. We have researched that category repeatedly over the years and never seen it produce.

I have also personally bought the ad on the back of the grocery receipt at the store next door to a school — a hundred thousand receipts a month — and generated approximately one call in a year. Impressions are not response, and a guarantee of impressions is a guarantee of nothing.

That said, at $150 a month I would not fight an owner who wants to satisfy his curiosity for thirty days with a tracking number attached. Which brings us to the last step, and the one that actually protects you.

Step Five — Pull: Write the Kill Criterion Before You Write the Check

The beacon offer was $1,800 for the year, pitched as “about $150 a month, which is really nothing.”

It is not $150 a month. It is $1,800, and the risk is not the $1,800 — it is what the $1,800 did not buy. Eighteen hundred dollars is a solid direct mail drop to your existing prospect database in January, which is a channel we already know produces. Every dollar committed to an unproven channel is a dollar not committed to a proven one, and in a business where January is one of the two highest-yield enrollment windows of the year, that opportunity cost is the real number.

So before you sign anything, write down four things:

1. The capped loss. The exact dollar amount you are willing to lose on this test, decided now, while you are calm. Not the annual commitment — the amount you would shrug at.
2. The response floor. The minimum tracked calls or form fills, by the decision date, that would justify continuing. Derive it from step two: if your ceiling is $30 a lead and the buy is $600, your floor is twenty leads.
3. The decision date. On the calendar, with the test’s tracking number and URL noted next to it.
4. What happens on each side of the line. Below the floor, you cancel — no “let’s give it one more month,” which is how a $150 test becomes an $1,800 commitment. At or above it, you scale and negotiate a better rate from a position of evidence.

Signing a twelve-month contract for an untested channel violates all four at once. If the seller will only sell you the year, you have your answer about how the year is likely to go.

Before You Spend a Dollar: The Free Congruence Pass

Now back to the $400 voice-search offer, because it deserves a fuller answer than “no.”

The mechanism behind being found by a voice assistant, a map, or an AI answer is not mysterious and not proprietary. It is your business listing data. When someone asks a device for the closest school of your style, the device is reading the same directory infrastructure everyone else reads, and the anchor of that infrastructure is your Google Business Profile.

Here is the free work that almost every school I look at has left half-done:

• Identical business name, phone number and web address everywhere. Not “close” — identical. Same suite number format, same phone format, same domain. Congruence across your website, your Google Business Profile, and every secondary directory.
• The profile has to actually point at your site. A profile with no working link to your website is a dead end for both the consumer and the algorithm.
• Real photos, and a lot of them. Search engines index images far better than they used to. Put your school name and location in the file names and the metadata, not just in the caption.
• Video links. The largest search engine owns the largest video platform. Links between your site, your channel and your profile are all one connected asset.
• A recurring calendar reminder. Chief Master Greg Moody’s rule is to re-check these listings quarterly, or at minimum every six months, because the platforms keep adding new fields — services, attributes, keyword areas — and unfilled fields are free ground you are ceding.
• The secondary directories. Keep them accurate because some map applications pull from them rather than from the dominant one. Accurate and free is the goal; that is different from buying advertising on them, which I would not do.

If a vendor wants $400 to do this in a semi-automated way and yours is genuinely a mess, $400 is defensible. If yours is already in order, you are paying $400 for a screenshot. Either way, you find out by asking exactly what they are doing — step three, applied to a digital service instead of a print ad.

The order of operations matters more than any individual item on that list. Fix the free congruence problem before you buy paid traffic, because paid traffic sends people to check you out, and the first thing a meaningful fraction of them will do is search your name. If what they find is a listing with the wrong phone number and four photos from 2016, you paid to send them to a bad first impression.

The Timing Constraint on Any Pilot

One last constraint, and it invalidates a lot of otherwise well-designed tests.

There are two genuinely hot windows in this business: late August through September, and the day after Christmas through January. Back-to-school for the kids market; New Year for both kids and adults. The rest of the year is workable — I ran schools where December was routinely the biggest cash month of the year — but those two windows are when demand comes to you rather than being manufactured.

That creates two failure modes for a pilot:

Testing in a dead window and killing a good channel. A print buy that lands in mid-July may produce a third of what the identical buy produces in early January. If you cancel on the July read, you have thrown away a channel that would have worked.

Testing in a hot window and crediting the wrong thing. If your pilot runs in January and enrollments are up, some of that lift belongs to the calendar, not the channel. Which is exactly why the dedicated tracking number matters — it separates channel-attributable response from seasonal lift.

And build lead time into the plan. Digital creative changes in thirty seconds. Print vehicles routinely close six to eight weeks before they land, which means the piece that hits mailboxes the first week of January is bought and finalized in November. If you decide in late December that you want a January print presence, you have already missed it. Sit down with your marketing calendar now and back-date every print deadline for the next twelve months.

What This Protocol Actually Changes

None of this is about being difficult with salespeople. Some of the best marketing relationships I have had started with a rep who was willing to be measured. The point is that you are the only person in the transaction whose money is at risk, so you are the only person who can insist that the transaction be about response.

Price your own channels. Compute the bar. Ask the seller to carry some of the risk and listen carefully to the answer. Buy the smallest tracked version. Write down what makes you stop.

Do that consistently and something quietly changes about your school: you stop having opinions about marketing and start having a portfolio. Some columns produce at $22 a lead and some at $55, some work in January and die in July, and you know which is which — which means when someone walks into your lobby with a binder, you are no longer guessing. You are shopping.

Related reading: When Is Your Martial Arts School Ready for Paid Ads? The Paid Traffic Readiness Ladder and The Four-Link Ad Chain: The Only Four Numbers That Matter in Martial Arts School Paid Ads.

Frequently Asked Questions

How much should a martial arts school pay for a lead?

Your maximum bid is set by your own conversion ratios, not by an industry average. Work backward from a target acquisition cost of $150 to $300 per enrollment. If your school converts five raw leads into one enrollment, you can afford $30 to $60 per lead. If it takes ten leads, your ceiling drops to $15 to $30 for identical traffic. That is why improving your booking rate and enrollment conversion is the fastest route to affording better media — the school with tighter ratios can outbid the school with loose ones and still hit the same acquisition cost. Compute the number, write it down, and treat it as a hard line in every negotiation rather than a feeling you renegotiate with yourself in the rep’s presence.

Should I ask an advertising rep to work on a pay-per-call basis?

Yes, and make the offer specific: a dollar figure per tracked call, justified by what you already pay in other channels, with a dedicated tracking number in the ad and your own proven creative. Most reps will decline, and that refusal is valuable information you obtained for free — it tells you what their confidence is actually worth. Some will accept, and those buys tend to perform because a seller with skin in the game starts caring about your placement and timing. If pay-per-call is genuinely outside their authority, fall back to the weaker versions: ask for the twelve-month discounted rate but commit only to two months at that rate, with the option to extend if the tracked numbers justify it.

Is it worth paying a company to get my school into voice search and AI results?

Only if your listings are genuinely a mess and you will not fix them yourself. Voice assistants, maps and AI answers read the same underlying business-listing data, anchored by your Google Business Profile. The work is free and unglamorous: identical business name, phone number and web address across every listing, a profile that actually links to your site, plenty of good photos with your name in the file data, and a recurring reminder to revisit every quarter as the platforms add new fields. Before you pay anyone, make them put in writing exactly what they will do. If it is only what you could do in an afternoon, do it yourself. And do this before you buy paid traffic, because ads send people to search your name.

Your Next Step

If you cannot state your cost per lead, cost per appointment and cost per enrollment for every channel you run right now, that is the constraint — not the ad rep in your lobby.

Two things to do this week.

First, get the free book. Six Simple Steps to Add 100 Students lays out the lead-generation columns that consistently produce for martial arts schools, so you have proven channels to compare any new pitch against. Get it at FillYourSchool.com.

Second, book your Free Personal Evaluation — a $1,297 value, at no cost. We will go through your actual acquisition numbers, identify which channels are earning their place and which are quietly draining cash, and build the tracking and pilot structure so the next rep who walks in has to prove it. Schedule through the Marketing hub.

About the Author

Stephen Oliver, MBA and 10th Degree Black Belt, is the Founder and CEO of Mile High Karate and Martial Arts Wealth Mastery, CEO of NAPMA (National Association of Professional Martial Artists), and Publisher of Martial Arts Professional magazine. A martial arts school owner since 1975, he and his coaching team — including Grandmaster Jeff Smith and Dr. Greg Moody — have helped school owners across the world build $1M+ schools.