Find the messages that convert. Then build everything around them.

The Playbook is how we run advertising: learn the exact words your market uses, build ads and pages around its biggest pain points, and let the results decide what runs next.

Topic: Advertising strategy System: The Playbook For: Brands spending on advertising Read: 7 minutes

The fundamentals matter more than ever.

The Playbook is built on direct response basics that predate every ad platform:

  • Clear message for every pain point.
  • Congruency from ad to page: they look the same and talk about the same thing.
  • More real angles in market, not more duplicates of one ad.
  • Decisions made from spend data, not from the brand deck.

What changed is how hard Meta now rewards this. Their delivery system has been adapting to exactly this style of advertising:

  • In late 2024 Meta rebuilt ad delivery around Andromeda, a retrieval engine that shortlists which ads even get considered for each person about to scroll.
  • It reads the creative itself and routes each message to the people most likely to respond. The new models handle roughly 10,000x more complexity, and Meta reported a 6% recall improvement.
  • Interest stacks and lookalikes matter less every quarter. The message is doing the targeting.
Meta reads the ad to decide who sees it. Which makes the message the targeting.

Andromeda didn't create this strategy and it isn't the point of it. But distinct messages with pages that match them are exactly what a content-aware delivery system rewards. The rest of this post is the process.

Start where your customers already talk.

Before we write a single ad, we go read. Reddit threads, Google reviews, Trustpilot, the category's review sites. Real customers describing the problem in their own words, with nobody selling to them:

  • Find the top five pain points in the category. The problems that come up again and again.
  • For each one, capture the desired outcome. Not what the product does, but what the customer wants to feel on the other side.
  • Collect the language. The exact words and phrases people use when they describe the problem to each other. That language goes into the ads and pages nearly verbatim.

Example: plantar fasciitis insoles

Pain point: the first steps out of bed in the morning. Sufferers describe it the same way over and over across Reddit and reviews: "like walking on glass," "a nail in my heel," "I dread getting up." Desired outcome: getting out of bed and just walking. No bracing on the nightstand, no shuffling to the bathroom, feet they've forgotten about by breakfast. Language: the phrases above, nearly verbatim. An ad that opens with "Dread your first steps in the morning?" doesn't read like an ad. It reads like the customer's own thoughts.

Then scrape the ad libraries.

Customer research tells you what to say. The ad libraries tell you what's already working. Both Meta and TikTok publish every active ad, and we scrape them for the competitors in the category:

  • Pull every active ad each competitor is running from Meta's Ad Library. It's all public: the creative, the copy, the start date, every version.
  • Sort by how long the ad has been running. This is the whole trick. Brands kill losing ads fast, so an ad that's been live for 90+ days is almost certainly profitable. Longevity is the closest thing to a public ROAS number.
  • Check the variant count. When a brand runs many versions of the same concept, that's a winner they're actively feeding. One-off ads are tests; ten variations of one angle is a scaled winner.
  • Tear the winners down. For each long-runner: which pain point is it hitting, what's the hook, what's the format, what's the offer. Not to copy it, but to see which angles the market has already paid to validate.
  • Do the same on TikTok. TikTok's Commercial Content Library shows every running ad, and Creative Center's Top Ads goes further than Meta ever will: it ranks ads by actual performance with real engagement numbers attached.

Then cross-reference. When a pain point shows up in both the customer research and a competitor's longest-running ads, that's a validated angle before you've spent a dollar.

This is the foundation for everything below. Five pain points, five desired outcomes, a file full of the market's own words, and a swipe file of the ads the category has already proven.

Build ads and pages for every pain point.

Each of the pain points gets its own funnel, and congruence is the rule that holds it together:

  • Every pain point gets its own ads and at least one dedicated page. Not the homepage, not the shared product page.
  • The ad and the page look the same and talk about the same thing: same promise, same language from the research, same visual world. The click should feel like the ad kept going.
  • This is message match, the oldest rule in direct response and the one most accounts break daily.

Test all pain points. Let the numbers decide.

All pain points go into market, and each side of the funnel gets judged on its own scoreboard.

On the ads:

  • Watch CPMr. It's a market signal, not just a cost line (plain-terms breakdown below).
  • Look for scaled spend with a ROAS that sticks. A great return on $4 a day proves nothing. The winners are the ads that hold their return while the spend climbs.

On the pages:

  • Conversion rate: is the page turning the clicks into buyers?
  • Bounce rate: a high bounce on a congruent page means the ad is making a promise the page isn't keeping, or you're just getting bad traffic.
  • Session recordings: watch real visitors use the page. Where they stall, what they skip, where they leave.

CPMr, in plain terms

CPMr is what you pay to reach 1,000 different people: spend ÷ reach × 1,000. Regular CPM counts the same person seeing your ad five times as five views; CPMr only counts unique people.

We use it because Meta's auction charges less to deliver ads people respond to. A low CPMr means Meta is spreading your message to new people efficiently. Strong creative often helps drive that efficiency, especially when CPMr stays low as spend increases. It also catches what CPM hides: an ad quietly dying of frequency while its CPM stays flat.

CPMr never gets read alone, though. CPMr tells you how efficiently the message is reaching new people. ROAS tells you whether that reach is turning into revenue. A cheap CPMr with a ROAS that holds or climbs while spend scales is the real winner. A cheap CPMr with a falling ROAS means the ad is pulling attention from the wrong people, or the page is losing them after the click. Either way, the pair tells you where to look next.

Expand the winning pain points into more page styles.

When a pain point wins, it earns more real estate. The same message gets wrapped in more page styles, because different buyers read differently:

  • Listicle: breaks down the reasons to buy one by one, so the reader can scan and land on the ones that hit.
  • Quiz: the visitor answers questions about their situation and gets a recommendation that fits them.
  • Advertorial sales page: an editorial story that builds the problem before the pitch, often written in the founder's voice.
  • Offer page: the deal front and center: bundle, discount, guarantee.
  • Compare page: your product against the alternatives, side by side.
  • Switcher page: built for people currently using a competitor and ready to leave.
  • Product page: the classic PDP, tuned to the pain point that sent the click.

Check out some of our page examples in the Page Lab.

Expand the ad formats around the winners.

Different personas respond differently to different ads. That's why we expand the ad types around a winning message: to reach those different people. Here are some of the styles we reach for.

UGC

A real customer on a phone camera, using the product and saying the message in their own words. Imperfect on purpose.

Yappers

One person talking fast, straight into the lens, no b-roll. The rant format. Retention comes from the talking, not the editing.

Video explainers

A walkthrough of the problem and the mechanism, animated or narrated. One of the biggest formats for us: it builds belief while it entertains.

AI-made

Generated spokespeople, scenes, and voiceover. Cheap enough to test ten angles in the time one shoot used to take.

Influencer

The message delivered by a face the feed already knows. Borrowed trust, and reach into an audience you'd pay years to build.

Statics with strong headlines

One image, one line big enough to stop the thumb. The cheapest format to produce and still one of the hardest to beat.

iMessage screenshots

A text thread where a friend recommends the product. Reads like gossip you weren't supposed to see, which is why it gets read.

Slack threads

The same trick for B2B: a workplace conversation screenshot where the product solves the problem mid-thread.

Apple Notes

The message typed into a plain note. Zero design, which signals zero sales pressure. Feels like advice, not advertising.

Testimonials

The customer's own words as the creative, verbatim, ugly punctuation and all. Proof carries what polish can't.

Promotional

The offer is the hook: a discount, a bundle, a buy-one-get-one. Strongest once the message is already proven.

Problem-solution

Opens on the pain point, then walks straight to the fix. The workhorse structure of direct response.

Comparison

Your product against the old way or a competitor, side by side in one frame.

Founder video

The founder on camera telling the story themselves: why they built it, what it fixes. Trust straight from the source.

Interview style

A back-and-forth conversation, podcast-clip energy. Questions pull the message out naturally instead of pitching it.

Break the winning pain point into new angles.

A winning pain point is bigger than one ad. It's a territory, and the last expansion is finding new ways into it:

  • Take the pain point that won and present it differently: new concepts, new scenes, new framings of the same problem.
  • Bloating alone can become "puffy by dinner no matter what I eat," "the jeans that fit this morning," or "canceling plans because of how my stomach feels." Same pain point, different doors.
  • This is where personas come back in, not for targeting but for how the ad itself gets made. A 20-year-old and a 60-year-old look at advertising completely differently, so the format, the pacing, the references, and what the ad even talks about all change with who it's for.
  • Same pain point, different lives: "surviving the 3pm slump at your desk" finds the 30-year-old, "keeping up with your grandkids" finds the 65-year-old.
  • Every angle that wins sharpens your picture of who's actually buying, which writes the next angle.

Iterate the winners. Bet on new pain points. Repeat.

The Playbook isn't a project with an end date. It's a monthly rhythm:

  • Iterate on the best pain points, concepts, and angles from last month. Press the winners harder.
  • Find and test new pain points alongside them. The research file always has a next bet.
  • Retire what the numbers say to retire, no matter who liked the creative.
  • Run the whole process again, with better evidence than last time.

Why the loop compounds

Ad platforms optimize toward the conversions your pages produce. A congruent page converts more of the same clicks, those conversions teach the delivery system who to find next, CPMs fall on the messages that resonate, and the winners carry more spend.

What to do this week.

You don't need the whole Playbook on day one:

  • Spend two hours reading your category's reviews and Reddit threads. Write down the five pain points, the desired outcomes, and the exact phrases people repeat.
  • Take the loudest pain point and build one ad and one dedicated page around it, in the market's own words.
  • Run it against whatever your ads point at today, and watch CPMr, conversion rate, and bounce rate.

That one test tells you how much the gap between your ads and your pages is costing, and everything else builds from there.

This is the system we run for clients.

The Playbook only works if someone actually runs it, every month. You need a dedicated team constantly doing the research, building around the pain points, finding the angles inside them, and hunting for the next ones. That's the reason to hire us: Growth Lab runs the ad side, Page Lab builds the pages, and the loop never sits idle.

Start with a free Growth Review: on a short call we go through your ads, your pages and your numbers, then send you a report with the changes we would make first.

Get Your Growth Review

Or email us directly: [email protected]