Before You Pay for Ads: The Checks That Come First

There is a moment in every quiet launch when ads start to look like the answer. The page is live, sales are not happening, and a hundred dollars of traffic feels like the difference between a product and a hobby. Sometimes it is. Usually it is a hundred dollars spent learning, at premium prices, what free traffic would have taught you for nothing.

The problem is not that ads are bad. Ads are a multiplier. They take whatever your page already does and do it more times. If the page converts strangers into buyers, ads buy you more strangers and more buyers. If the page does not convert, ads buy you faster, more expensive silence, and the silence arrives with no explanation attached.

So the question is never "should I run ads." It is "has this product earned ads yet." Here are five checks that answer it.

Check one: at least one sale from a stranger

Before paying for traffic, you need evidence that free traffic converts. The cleanest form of that evidence is one sale to someone with no relationship to you. Not a friend, not a former colleague being kind, not someone who followed your build in public and wanted to support you. A stranger who found the page, read it, and decided the product was worth their money.

One stranger sale proves the whole chain works: the promise is legible, the price is survivable, the checkout functions, and at least one person in the world converts on this page without knowing you. That is the thing ads multiply. Zero stranger sales means the chain is broken somewhere, and ads cannot tell you where. They just run more people into the same wall while charging you per collision.

If you have not had that sale yet, the fastest route to it is not spend, it is direct outreach to people who already have context with the problem, followed by useful posts in the communities where your buyer lives. That work is free, it is slower than ads, and it comes with something ads never include: replies. Replies tell you why people do not buy, which is precisely the information you need next.

Check two: a sales page that answers real objections

The second check is whether your page has been rewritten at least once against real objections, meaning things actual humans said or asked, not things you imagined they might think.

If nobody has ever asked you a question about the product, your page is a first draft, and first drafts are the worst possible thing to send paid traffic to. The questions you collect from early conversations, "what format is it," "does it cover my situation," "how is this different from the free stuff," are a punch list for the page. Every question a real person asked is a place the page failed and can now be fixed for free.

A page ready for paid traffic states the problem in the buyer's own words, shows exactly what is inside (a real table of contents beats adjectives), says plainly who it is for and who it is not for, and answers the objections you have actually heard, ideally in a short FAQ. If you cannot fill an FAQ with questions people really asked, you have not talked to enough people to be spending money yet.

If you have not collected real questions at all, the ten-conversation test in how to test demand in one weekend is the fastest free way to gather them; run that loop until the page has survived contact with real readers.

Check three: the margin math at your actual price

Now the arithmetic that most first-time ad buyers skip. Say your product sells for 19 dollars. Your platform takes its cut per sale; check your platform's current fee page and write the real number down. What is left is your gross margin per sale, the most an acquisition is ever allowed to cost you.

The cost of an ad-driven sale is your cost per click divided by your page's conversion rate on cold traffic. You will not know either number precisely before spending, but the structure of the equation is the point: at a low ticket price, clicks are expensive relative to margin, and the conversion rate a 19 dollar product needs to break even on paid clicks is a rate most first sales pages do not reach on cold traffic.

Run your own version honestly, with your price, your platform's real fees, and a pessimistic conversion guess, and see what a click is allowed to cost before the math turns negative. For many first products the honest answer is "less than clicks cost," which means ads at this price are a structural loss, not a growth channel. That is not a tragedy. It just means the free channels are not the budget option, they are the only option that adds up, until either the price rises or the product ladders into something bigger. If the price itself is the weak link, fix the price before spending a dollar on traffic.

Check four: a way to count results by hand

Ads without measurement is spending with a blindfold on. Before any money moves, you need to be able to answer one question precisely: how many sales came from the ad?

You do not need an analytics suite for this. At small scale, hand-counting works: know your baseline (orders per week as counted in your store's admin, right now, without ads), use a link you can distinguish for the ad if your platform supports it, and log daily spend next to daily orders in a plain text file. If your store currently gets close to zero orders a week, attribution is nearly free: whatever happens during the ad window stands out against a flat baseline.

The check fails if you cannot state your current weekly baseline from your own admin screen. Not knowing your baseline means any post-ad number is uninterpretable, and "uninterpretable" plus "money spent" is the worst combination in early marketing. Set up the counting ritual first, run it for a week or two without spend, then decide.

Check five: free channels genuinely exhausted, not just tried once

"Free did not work" usually means "I posted twice and nothing happened." That is not exhaustion, that is a flinch.

Genuinely working the free channels looks like: direct messages to every person who had context with the problem while you built. Useful posts, the kind that teach something real, in the two or three communities where your buyer gathers, over weeks, not days. A product listed where marketplace search can find it, with a title and description written in the buyer's search language. Each of those channels also produces feedback, and the feedback compounds: every reply improves the page that any future ad would land on.

The honest test: write down what you actually did. Ten messages or fifty? One community post or eight useful ones over a month? If the written list looks thin, the cheapest growth move available is finishing it, not paying to skip it. The starting playbook is in how to launch a digital product with no money.

If every check passes, start embarrassingly small

Suppose all five pass: a stranger has bought, the page has survived real objections, the margin math has room, the counting ritual exists, the free channels are truly worked. Then a paid test is a legitimate experiment, and the way to run it is small enough that the worst case is a cheap lesson.

Set a fixed budget you would be comfortable calling tuition, decided before the first dollar moves, with a calendar end date. One channel, one audience, one ad, so that whatever happens is attributable to something. Let it run the full window without fiddling. Then close the loop against the file from check four: spend, clicks, orders counted in the admin, margin per order. The ad either paid for itself or it did not, and either way you now own a real number instead of a hope.

And if it did not pay: stop, keep the learning, go back to the channels that were working, and revisit ads when the price, the page, or the product has materially changed. The worst outcome in paid marketing is not a failed test. It is a failed test that gets rerun monthly out of optimism.