Most Meta advertisers don't have a creative problem. They have a research problem.
They're spending hours in Canva, iterating on hooks, swapping out thumbnails, A/B testing colour variations, and wondering why none of it moves the needle. The honest answer is uncomfortable: the creative isn't the bottleneck. The thinking behind the creative is.
If you skip the research, every ad you make is a guess. You can guess well. You can guess for years. But at some point, your competitors stop guessing and start researching, and from that point on, you're outgunned on every single placement.
This is the part of the process most agencies don't show you, because most agencies don't actually do it. They open Ads Manager, they look at last month's winners, they brief an editor, and they call it strategy. It isn't. It's reaction.
Real ad research has three pillars: competitor, customer, and creative. Get these right and the ads almost write themselves. Skip them and you'll spend the next twelve months wondering why nothing scales.
Why Research Has To Come First (And Why Most People Resist It)
There's a reason most advertisers skip research. It doesn't feel like progress.
When you're researching, nothing is being built. Nothing is going live. The account isn't moving. To a results-hungry founder or media buyer, that feels like waste, so they shortcut to creation, get something into the account, and tell themselves they'll "iterate from data."
The problem is that the data they're iterating from is already corrupted. They're not testing whether the right angle works on the right customer. They're testing whether their guess at an angle works on Meta's guess at an audience. That's two layers of guesswork stacked on top of each other, and no amount of iteration cleans it up.
Research is the difference between throwing a dart at a dartboard with the lights on and throwing one with a blindfold. The throw is the same. The chance of hitting anything useful is not.
So before a single hook is written or a single creative is briefed, three questions need answers:
- Who is actually buying this product, and why?
- What are the best competitors in this market doing, and what are the worst doing wrong?
- What does creative that wins in this category actually look like right now?
Each of these maps to one of the three pillars. We'll take them in order.
Pillar One: Competitor Research
The mistake most people make with competitor research is scrolling the Meta Ads Library for ten minutes, screenshotting a few ads they like the look of, and calling it done. That's not research. That's mood-boarding.
Proper competitor research is structural. It asks: what is the offer doing, what is the site doing, and what is the ad doing, and how do those three pieces work together to win the click and the conversion?
Start with the site, not the ad. The ad is downstream of the offer. The site is where the offer lives. So we start there.
When auditing a competitor's site, the questions worth asking are:
- What's in their hero section? What's the headline doing, is it leading with a result, an identity, a price, or a pattern interrupt? What's the primary CTA?
- What CRO features are they running? Sticky add-to-cart bars, exit-intent popups, free shipping thresholds, urgency timers, social proof modules, video product pages, AR try-ons?
- What payment gateways and allowances do they offer? Klarna, Clearpay, Apple Pay, PayPal, and how prominently are these surfaced? Friction at checkout is invisible to most advertisers but devastating to conversion rates.
- What offers are they running? Bundles, BOGOs, subscribe-and-save, gift-with-purchase, sitewide percentage off, threshold-based incentives? Are these always-on or seasonal?
Then comes the analysis. It's not enough to note what they're doing. The question is: why is it working, and what can you do that beats it?
This matters more than most advertisers realise. A solid ad gets the customer in the mood to buy. But "in the mood to buy" doesn't mean "in the mood to buy from you." If a competitor is running a stronger offer than you are, what you've actually done with your ad is warm up a prospect who will then go and convert with someone else. You've paid for their customer.
An offer has to do two things: get someone to want to convert now, and convince them to convert with you rather than anyone else. That's the bar. If your offer isn't doing both, every pound of ad spend you put into the account is partly subsidising your competitors.
Then the ads library. Once you understand what competitors are selling and how, the ads library tells you how they're selling it.
The questions that matter here:
- What formats are working, static images, single videos, carousels, catalogue ads, reels-native creative?
- How long have specific ads been running? (Anything live for 60+ days is almost certainly a winner, Meta would have killed it otherwise.)
- What's the angle distribution? Are they leaning heavily on testimonials, founder-led content, problem-aware hooks, before-and-afters, or product demos?
- What's the creative diversity look like? Are they running 5 ads or 50?
For ecommerce brands in 2026, the ads library has become more useful than ever, partly because Meta's recommendation system rewards creative volume, and partly because best-in-class advertisers are now running 20+ creatives at a time inside Advantage+ Shopping campaigns. If your competitors are running 20 and you're running 4, you're not in the same fight. You're not even in the same weight class.
Pillar Two: Customer Research
Competitor research tells you what the market is doing. Customer research tells you who you're actually talking to.
Most advertisers think they know their customer. Almost none of them really do. They know the demographics. They might know the average order value. What they don't know is the emotional state of the person at the moment they encounter the ad, and that's the only thing that matters.
Every piece of customer research is trying to answer two questions: What does this customer actually want? And why do they want it?
The "what" is usually obvious. The "why" is where the gold is.
Take a skincare brand. The "what" is clear skin. The "why" might be confidence in photos, the ability to stop wearing foundation, an end to the morning-mirror anxiety they've felt for a decade, or a way to stop hiding from their own reflection. Each of those is a completely different ad. None of them is "clear skin."
This is where the 16 core emotions framework comes in. Most buying decisions trace back to one of a small set of fundamental emotional drivers, and identifying which one is dominant for your customer is the single highest-leverage research output you can produce. Once you know the dominant emotion, every creative decision downstream gets easier: the hook, the body, the visual, the call to action. They all align around one core feeling.
The post-purchase survey. The single most underused tool in ecommerce is the post-purchase survey. A customer has just bought. They've handed over money. They're as engaged as they will ever be with your brand. And in that window, typically the order confirmation page or the first transactional email, they will answer questions honestly that no focus group, no Instagram poll, and no ad comments section will ever surface.
The questions worth asking:
- Where did you first hear about us? (Attribution truth, separate from your tracking.)
- What nearly stopped you from buying today? (Objection identification, pure gold for creative.)
- What was the single biggest reason you went ahead with the purchase? (The actual buying trigger, in the customer's own words.)
- If you had to describe what this product does in one sentence to a friend, what would you say? (Free copywriting from people who have already paid.)
- Did you consider anyone else before choosing us? (Direct competitor intelligence.)
The point isn't to run this once and learn everything. The point is to run it always-on, and to build up a corpus of first-party voice-of-customer data that becomes the source material for every ad you'll ever write.
This data doesn't just feed ads. It feeds your landing pages, your email sequences, your sales scripts, your product copy, and your offer construction. Customer research is the only kind of research with compound returns.
Building the Ideal Customer Profile. The output of customer research is an Ideal Customer Profile, but not the one most agencies hand over. Not the one with a stock photo and a demographic summary. A real ICP answers:
- Who are they? Beyond demographics, what's their identity, their self-perception, what tribe do they belong to?
- What are they thinking right before they encounter the ad? What's the trigger event that has them ready to buy today and not three months ago?
- What makes them feel seen? What language do they use to describe their own problem? Not your category language, theirs.
- What's their vernacular? How do they actually talk? What words would never appear in their feed, and what words signal "this is for me"?
- What organic content are they already engaging with? This tells you the visual and tonal register you need to hit to belong in their feed.
A good ICP is a document you can hand a creator and they instantly know how to talk. A bad ICP is a slide that lists "25-45, female, household income £40k+, interested in wellness." One of those is useful. The other is wallpaper.
Pillar Three: Creative Research
Creative research is where competitor and customer research collide. Now that you know what the market is doing and who you're talking to, the question becomes: what does winning creative actually look like in your category, right now?
Meta's a visual platform. It's an entertainment and connection platform. Ads that ignore that lose to ads that don't.
When mapping format opportunities, the categories worth tracking are:
- UGC-style video (creator-led, native-feeling, low-fi by design)
- Founder-led video (authority, story, brand voice)
- Testimonial content (in carousel, video, and static forms)
- Press and review-led static (third-party validation as the core hook)
- Us vs Them (positioning, often static or carousel)
- Product demo / mechanism reveal (showing how it works)
- Trend-jacked organic-style creative (formats borrowed from what's working organically, then mapped to your offer)
Most accounts only run two or three of these. The best accounts run all of them, then double down on whichever the algorithm rewards for that specific customer.
Why trend mapping matters more than most people realise. Here's a stat worth sitting with. Every day, around 3.2 billion people use the Facebook family of platforms. And every day, around 3.5 billion reels are shared. That's more reels shared than there are users. The platform isn't just visual. It's short-form video-native now, in a way most advertisers haven't fully adjusted to.
What this means in practice: the gap between organic and paid is closing. The creative that wins is the creative that doesn't feel like an ad, the creative that borrows the visual and tonal codes of the organic content people are already engaging with, then applies a clear conversion mechanism underneath.
This is why "trend mapping" is its own research output. The question isn't just "what are good ads in our category doing?", it's "what's trending organically right now that we can adapt to our offer without it feeling forced?"
A clumsy trend-jack feels worse than no trend at all. A well-mapped one is the closest thing to a free ride in performance marketing.
The Output Of Good Research
If the research is done right, by the time you sit down to brief creative, you should be able to say:
- Our ideal customer is X, driven by emotion Y, currently feeling Z.
- Their language sounds like this, not like this.
- The top three competitors in our market are running offers that look like A, B, and C, and ours beats them on D.
- The top-performing ad formats in our category right now are E, F, and G.
- The trending organic content angles we can credibly adapt are H and I.
That's a brief. That's the difference between creative that ships and creative that wins.
Most agencies will hand you the second half of that, the creative, and call it strategy. We treat the first half as the actual job. The creative is just the execution.
A Final Test Before You Move On
Once your research is done, there's one final test before any of it becomes an ad.
It's a question Ryan, one of our team, asks constantly: "Would you convert on the ads you're coming up with?"
Put aside your biases. Look at the creative through an objective lens. If the answer is no, the research either isn't deep enough or hasn't been translated properly into the brief. Either way, you're not ready to ship.
Research isn't the part of the process that wins awards. It's the part that wins accounts.
