Ask an agency how it makes your ads and listen for the word "template". Most flooring companies that have already fired somebody were paying for the same ad every other account in that agency was running, logo swapped, city changed.
We do the slow version: offers and creatives built around one flooring company, put up against each other in that company's own market, judged on what lands on the calendar. Here is the method, so you can run it yourself or hold us to it. No number is attached to our tests, creatives or clients anywhere on this page. The client figures we publish live on the results page, each welded to its own window.
Why a copy-pasted ad fails in flooring
Flooring is not one market. A showroom on a busy road in a city full of new housing sells to a different person, at a different price, against different competitors, than a family install business two states away working off referrals. A template ad cannot hold any of that, so it says the safe thing: free estimate, quality flooring, family owned, call today. Every company in the feed says that.
The platform punishes it too, and Meta publishes why. The winner of an ad auction is the ad with the highest total value, and the total value is a combination of three major factors: Bid, Estimated action rates, and Ad quality. Two of those three are about the ad rather than your money, which is why Meta says on the same page that an ad that's more relevant to a person could win an auction against ads with higher bids. A better ad is not a nicer ad. It is a cheaper one.
The brief, and why it comes before any ad
Nothing gets made until four things are written down with the owner, rather than guessed at from the website.
- The brand. How you already sound to the people who buy from you, what your reviews say in their own words, and what the showroom looks like on a Saturday.
- The offers. Everything you can honestly put in front of a homeowner: the promotion you run, the product you want to move, the thing the shop down the road does not do.
- The market. Your service radius, the product your buyers ask for by name, who else is advertising there, and what a job is worth when it closes.
- Install capacity. The one nobody asks about and the one that decides the plan. Book more work than the crews can install and you have a review problem in six weeks.
That last item is why we will tell a shop the honest answer is not yet. The system those four inputs feed is on the flooring marketing agency page, and the questions readers ask next are on the FAQ.
Offers: a showroom and an installer sell different things
An offer is not a discount. It is the reason a homeowner stops scrolling and gives you a phone number today instead of thinking about floors again in spring.
For a showroom or a flooring retailer, the offer is usually built around the room. Samples in the house so they see the plank next to their own furniture and their own light. A booked time with somebody who measures properly. A price that covers the install rather than a per-square-foot number that doubles at the quote.
For an established flooring contractor who sells and installs, the offer is usually built around certainty. When you can start. How long the crew is in the house. What happens to the old floor. Whether the quote is the number they pay. Homeowners burned once by a trade are buying the absence of a disaster, and an offer that names what they are afraid of beats one that names a discount. So the same market gets two different tests depending on who is paying for the ads.
How a creative gets built
Once an offer is written, the creative is the delivery. Three parts, in this order.
- The hook. The first line and the first second. It names the buyer or the problem, and it never opens with the company. "Your floors are the last thing in the house nobody has touched since you moved in" is a hook. "Welcome to our family owned flooring store" is a sign on a building.
- The proof. Your work, shot on your jobs. Before and after in the same room from the same corner. The floor going down. Proof is what a template cannot supply, which is why a template account leans on stock.
- The showroom, and the people in it. The owner talking to camera in his own shop beats a designed graphic more often than anybody expects, and it is the one asset a rival agency cannot copy off your feed.
Several creatives get built for each offer, and they are built to be genuinely different rather than recolored. Same offer, different hook. Same hook, still photograph against video. Two ads that differ by a font are not a test.
The test ladder
The ladder separates a tested account from a busy one. Each creative gets a small budget of its own and a fair run against its siblings, and then it dies, stays in, or gets the budget.
Fair means the comparison is set up so the answer means something. Meta is blunt about what informal testing does: A/B testing helps ensure that your audiences will be evenly split and statistically comparable, while informal testing can lead to overlapping audiences, and it recommends using the same budget for both versions in a test to ensure a fair comparison. Flipping ad sets on and off by hand and reading the difference is two things happening at once and a story told about them afterwards.
A loser gets turned off rather than nursed, because fixing it is the expensive move: every edit restarts the clock, which is the next section. A winner takes the budget the losers were drinking and keeps it until its cost per booked estimate moves. A fair run is long enough for the platform to have stopped guessing and long enough for a flooring decision to happen, which is slower than most trades. Weekly numbers in this business are weather.
The learning phase, in Meta's own words
Every new ad set starts in a state Meta calls the learning phase: the period when the delivery system still needs to learn about how an ad set may deliver and perform. Ad sets leave it once they can deliver stably, which usually occurs after about 50 results in the week after the ad set's last significant edit.
Read the second half again, because it governs the whole ladder. Only a significant edit causes an ad set to reenter the learning phase, and Meta's list of those includes any change to targeting, any change to ad creative, adding a new ad, and pausing for seven days or longer.
Two consequences, and they are why the ladder looks the way it does. First, Meta's own advice is to wait to edit your ad set until it's out of the learning phase, because during it performance is less stable, so your results aren't necessarily indicative of future performance. An owner who tweaks the headline every Monday has an account that is permanently learning and never delivering.
Second, you cannot test everything at once. Meta says that when you create many ads and ad sets, the delivery system learns less about each ad and ad set, and one that cannot get enough results ends up labeled learning limited, which Meta calls an indication that your budget isn't being spent effectively. A modest flooring budget split across a wall of ad sets buys activity and nothing you can read.
None of that is an argument against testing. The same Meta page says the opposite in one line: testing new creative and marketing strategies is essential for improving your performance over time. It is an argument against testing carelessly, which is what most accounts are doing when they say they test.
Judge the test on booked estimates, not leads
A creative that produces the cheapest leads in the account can still be the worst ad in it. Cheap leads are easy to buy: promise less, qualify nobody, and the form fills arrive. So a test is scored on cost per booked estimate: the money that went into the ads divided by the appointments that landed on the calendar, counted in your own CRM. The full argument is on the only marketing number worth arguing about.
One honest limit. The ad account cannot see every sale an ad causes: some buyers never touch it and call the store, and some call months later when the floor finally has to go. The three waves of flooring ad sales explains which of those the platform counts. A creative test reads the wave it can see, which makes it a tool for comparing ads against each other rather than a verdict on the advertising.
Who calls the lead: your team, or ours
A test is only as honest as the follow-up underneath it: a creative whose leads never get called looks exactly like a creative that does not work.
Your team can take the leads. Every inquiry lands in your CRM with its source attached, so you can see which ad, offer and hook produced it, and your office manager or closer works the list the way they already do. The tracking still tells you which creative earned the appointment.
Or you add our call center and we run that half: a person calls every lead within minutes, asks about budget range, timeline, whether they own the property and the product they want, and books the in-home estimate on your calendar. The notes travel with the appointment, so a creative is scored on estimates rather than on phone numbers. That is appointment setting. Both are real options, the call center is the optional one, and the choice is yours.
What the library lets us skip
We took the long route, one flooring company at a time, and what that built is a body of tested flooring offers and creatives across different markets. It is not a swipe file. It is a shorter list of things worth trying first: which shapes of offer hold up in flooring and which die on contact, which hooks stop a homeowner, what footage earns attention against a still, and which promises get a sales team into trouble later.
That is the basis of the claim this site makes, and also why the claim is checkable. What the accounts behind it did is written up in three case files: Hometown Floor Covering, MC Flooring & Tile and Floorworks, every figure printed with the window it happened in.
What we still test fresh, every time
The library shortens the list. It never writes the ad. All of this is tested from the ground up on a new account, because nothing in another company's market answers it.
- The offer this market wants. The promise that won two states away is a starting hypothesis here.
- The hook. Local competition, local prices and local housing stock decide which opening line stops somebody.
- The proof. Your jobs, your showroom, your voice. There is no shortcut here and there should not be.
- The product. What your buyers ask for by name and what your crews can install this season.
- The follow-up. Which questions qualify a buyer here, and how fast somebody has to call to keep them.
What to ask any agency about its testing
Run these on us as readily as on anybody else. An agency that tests answers all five in a sentence each, without a deck. The longer list, written for use on any agency including this one, is seven questions for any agency.
- Show me two ads you run for two clients in the same trade. If they are one ad with a different logo, the conversation is over.
- What changes between one creative and the next? It should be the offer or the hook. If it is the color, they are decorating.
- How do you decide a creative lost? Listen for a fair run and a number. "It just was not performing" is a feeling.
- What number do you score a test on? Cost per lead and nothing downstream of it counts the cheapest half of the problem.
- Whose ad account is it? Everything you paid to learn lives in there. If it is theirs, so are the tests.
Sources
- Meta Business Help Center, About the learning phase. What it is, what ends it, the editing advice.
- Meta Business Help Center, Significant edits and learning phase. Which changes restart the clock.
- Meta Business Help Center, About A/B testing. Even splits, equal budgets, informal testing.
- Meta Business Help Center, About learning limited. The ad set that never gets enough results.
- Meta Business Help Center, About ad auctions. Total value, and why relevance can beat a bigger bid.
