8 Ad Copy Frameworks: Choose by Goal, Then Test by Channel
I’m Stefano Mazzei, a performance marketer at Cosma, and here is the short answer: start with AIDA, PAS, Before-After-Bridge, 4Ps and Hook–Story–Offer. Choose between them by the campaign objective and by whether the message should lean on emotion or on rational proof. Everything else is detail you can test.
TL;DR:
- For awareness, use APP or StoryBrand; for consideration, try AIDA or PAS; for conversion, favor FAB, 4Ps, or BAB when buyers weigh specifics.
- Use PAS or 4Ps with evidence to reassure new buyers, but give repeat buyers a concise FAB line focused on what has changed.
- For responsive search ads, make every headline work alone because Google recombines them; pair social copy with a matching opening visual, not a disconnected logo.
- Hold the visual, audience, offer, and budget constant; test one framework, wait for enough conversion events, and judge sales by conversion rate and acquisition cost.
Table of Contents
- Which ad copy frameworks are worth learning first?
- How do you choose the right framework before you start writing?
- What do copy-ready templates look like for social, search and video?
- How do you test frameworks and measure which copy wins?
- What does the research say about gain versus loss framing?
- My three rules for choosing and writing ad copy
- FAQ
- Sources
Which ad copy frameworks are worth learning first?
Most frameworks are variations on the same job: get attention, make the problem or desire clear, and give a reason to act. The shortlist below covers the formats I see ad writers use most. Each entry has a plain definition, a best use, and a one-line example you can adapt.
- AIDA (Attention, Interest, Desire, Action): The classic direct-response ladder. Best for consideration and bottom-funnel ads where the reader already knows the category. Example: “Tired of flimsy water bottles? This one is double-walled steel, keeps drinks cold, and ships free today.”
- PAS (Problem, Agitate, Solution): Names a pain, makes it feel real, then offers the fix. Works for pain-point products and attention-grabbing offers. Example: “Your back aches after a long day. Sitting on a hard chair can make it worse. Our lumbar-support seat pad is designed to support your posture.”
- Before-After-Bridge (BAB): Show the better “after” state, then bridge to it with your product. Strong for aspirational claims and hero creatives. Example: “Mornings used to mean a hunt for matching socks. Now every pair finds its mate. The bridge: our sock subscription.”
- 4Ps / PPPP (Picture, Promise, Proof, Push): Paint the scene, state the promise, back it with evidence, then push to buy. Use it when you actually have proof, such as reviews, results or a guarantee.
- FAB (Features, Advantages, Benefits): Translates a spec into what it does for the buyer. Ideal for product-detail ads and technical offers. Example: “Merino blend (feature), stays odor-free between washes (advantage), so you pack lighter for a week away (benefit).”
- StoryBrand (Character, Problem, Guide, Plan, CTA): Casts the customer as the hero and your brand as the guide. Better for brand campaigns, longer scripts and hero videos than for a short headline.
- APP (Awareness, Problem, Positioning): An early-funnel structure. It builds awareness of a problem, then positions your product as the sensible answer, without pushing a hard sale.
- Hook–Story–Offer: A micro-structure for social video. A quick hook, a short story or demonstration, then a clear offer. It suits fast creative loops where you ship many variants.
How the frameworks bend to different formats
The framework stays the same while the format squeezes it. For search headlines, keep only the strongest beat: a PAS headline might be just the problem (“Back Pain at Your Desk?”) and a FAB description carries the benefit. For responsive search ads, write each headline so it works alone, because the system recombines them. For static social, use primary text for the narrative (AIDA or PAS) and let the headline carry the promise. For video, the hook has to land quickly, so the framework usually collapses into Hook–Story–Offer.
If you want a broader primer on persuasive writing before you pick a framework, this SEO copywriting guide covers headline fundamentals that carry over to ads.
Frameworks across customer personas and product types
The same framework reads differently depending on who is on the other end. A first-time buyer of a skincare serum needs the problem named and the risk lowered, so PAS or 4Ps with proof tends to fit. A repeat buyer of a pantry staple already trusts you, so a short FAB line about what’s new beats a long story. For high-consideration products like mattresses or supplements, I lean on proof-heavy structures. For impulse and low-price items, a fast Hook–Story–Offer or a one-line BAB usually does the job. Segment first, then pick the structure that matches what that persona still needs to believe.
Pairing the words with the visual
Copy never runs alone on Meta. The first frame of a video or the dominant element of a static image decides whether anyone reads your primary text. Write the hook and the visual together: if the hook says “stop wasting money on razors,” the opening frame should show the razor pile, not a logo. When copy and visual say two different things, the ad feels confused and results suffer. My team at Cosma builds the creative concept first, then writes the framework around it, which is why you can see the pairing in our portfolio of creative examples.
How do you choose the right framework before you start writing?
Picking the framework after you’ve written a draft is how teams burn budget on trial and error. Decide first, using the funnel stage as your starting point.
For awareness, lean on APP or StoryBrand. For consideration, AIDA and PAS usually fit best. For conversion, FAB, 4Ps and BAB tend to work because the reader is ready to weigh specifics. Then adjust for creative format, how familiar the audience is with you, and how clear your offer is. A confusing offer needs FAB or 4Ps to spell it out; a well-known offer can run on a tight PAS.
Then decide the emotional versus rational question. Research on message framing (more on that below) suggests gain-framed messages tend to work for rational and positive-emotional appeals, while a targeted loss frame can serve higher-involvement, risk-reduction products when backed by solid information.
Answer these six questions before you write a word:
- What is the single campaign objective: attention, clicks, or purchases?
- Where is the audience in the funnel, cold, warm or past buyers?
- What is the one call to action you want them to take?
- What is the main barrier stopping them from buying: price, trust, confusion or timing?
- What evidence do you actually have: reviews, results, a guarantee, a demo?
- What is the creative format: static image, short video, carousel or search text?
A quick ecommerce example
Say you sell a weighted blanket to cold traffic on Meta, with customer reviews and a trial. The objective is purchase, the audience is cold, the barrier is “will I actually sleep better?”, and the proof is reviews plus the trial. That points to PAS for the hook (poor sleep named and agitated) with a 4Ps-style proof line closing the ad. If the same product ran to a warm retargeting audience, I’d drop the problem setup and use a FAB line plus the trial as the push. Same product, different framework, because the audience’s need changed.
For a deeper look at how I structure these tests across hooks and angles, see my breakdown of Facebook ad angles for ecommerce.
What do copy-ready templates look like for social, search and video?
Use these as starting points and swap in your own product details. Each template marks the hook, the proof and the call to action so you know what to change in a test.
Social ad templates
- AIDA: Hook: “[Specific audience], this is for you.” Interest: “[Product] does [one concrete thing].” Desire: “Imagine [result] without [annoyance].” Action: “Shop now and get [offer].” Headline: “[Result].”
- PAS: Hook: “Still dealing with [problem]?” Agitate: “It only gets worse when [consequence].” Solution: “[Product] fixes it by [mechanism].” CTA: “Try it risk-free.”
- BAB: Before: “Right now, [frustrating reality].” After: “Picture [desired state].” Bridge: “[Product] gets you there with [mechanism].” CTA: “See how it works.”
For each, I write three variants by changing only the hook line. That keeps the test clean.
Search ad guidance
Google’s own guidance says responsive search ads let you supply multiple headlines and descriptions, and the system tests combinations over time. Pinning reduces that flexibility, so reserve it for text that must appear, such as a legal line. Write headlines that each stand alone: one benefit, one proof point, one offer, one keyword-led line. The search campaign specs list the character limits and recommended counts, so check them before you draft. A good RSA mixes frameworks across its assets: a PAS-style question headline, a FAB benefit headline, a 4Ps proof headline, and a plain CTA.
Short video micro-scripts (Hook–Story–Offer)
- Product demo: Hook: “Watch this [product] do [surprising thing].” Story: show it solving the problem in one continuous shot. Offer: “Grab yours, free shipping this week.”
- Testimonial: Hook: “I almost didn’t try this.” Story: the customer explains the problem, the doubt, and the result in their own words. Offer: “Join other customers who switched.” (Use a real, verified number only.)
- Aspirational: Hook: “This is what [desired morning/routine] looks like.” Story: a few quick shots of the after state. Offer: “Start your version today.”
Annotating what to swap
In every template, the hook is the first line or first three seconds, the proof is the claim a skeptic would check, and the CTA is the last instruction. When a test loses, change one of those three, never all at once. For a template-by-template look at what separates strong creative from weak, my creative analysis framework shows how I score ads.
Pro Tip: Write the mobile version first: put the hook and the benefit near the start of the primary text, since the rest is often hidden behind “See more.”
How do you test frameworks and measure which copy wins?
A framework test only tells you something if the copy is the sole variable. Hold the visual, audience, offer and budget constant, and change only the framework. If you change the image too, you won’t know which change moved the numbers.
Match the metric to the objective:
- Awareness ads: click-through rate and thumb-stop or hold rate tell you whether the hook earns attention.
- Direct response ads: conversion rate, cost per acquisition and return on ad spend decide the winner. A high CTR with a weak conversion rate usually means the hook over-promised.
- Brand video: view-through rate and engagement show whether the story holds people.
- Search ads: add multiple RSAs per ad group, supply many unique headlines and descriptions, and avoid over-pinning so the system can learn.
On sample size, I avoid calling a winner on a handful of clicks. My rule of thumb comes from my creative-first testing method with the conversion-event testing method: wait until each variant has gathered enough conversion events to be meaningful before cutting anything, and keep tests running through pacing swings rather than reacting to early noise. When one framework wins, don’t stop. Take the winner and test a new hook against it, because the first win is only the baseline for the next round.
Why test at all: Message reframing research reports that reframing can outperform traditional techniques on attention, novelty and favorable ad attitudes without harming perceived credibility, which is a good reason to treat framing as a testable variable and not a settled question.
What does the research say about gain versus loss framing?
Research gives a useful anchor for these choices.
A marketing journal study on gain versus loss framing. It finds that gain-framed messages produce more positive ad and brand attitudes for rational and positive-emotional appeals, while loss framing can work for higher-involvement products when paired with strong informational evidence.
The second is a large-scale analysis in PLOS One of linguistic features in Instagram ads. It reports that language effects on click-through rate depend on product category: supplements respond to risk and loss framing, while cosmetics respond to gain and aspirational language. The authors recommend category-specific testing.
A framing that lifts clicks for one product category can underperform for another, so test within your category instead of copying a rule from a different one.
The practical rule: match the frame to the appeal type and the product category, and when you’re unsure, run a narrow A/B test. Here’s how that turns into two variants for a supplement:
- Loss frame: “Low on magnesium? Poor sleep may be the first sign. Don’t let another night slip away.”
- Gain frame: “Wake up rested. Our magnesium blend supports calmer evenings and better mornings.”
Same product, same offer, same creative. Only the frame changes, so the result tells you which one your audience prefers.
My three rules for choosing and writing ad copy
These are the rules my team at Cosma follows when we pick a framework and write copy:
- Decide the framework before drafting. Objective, audience stage and barrier come first; the words come second.
- Change one thing per test. Hook, frame or proof, never all three.
- Let the data overrule your taste. The ad I like least has won more than once.
Two mistakes I see constantly: teams pick a framework because it’s familiar and never test an alternative, and they declare a winner before enough data has accumulated. Both waste spend.
My team at Cosma runs framework-driven creative tests for ecommerce and DTC brands, and you can see how we work in our case studies. If you want help setting up your own tests, you can book a consult with my team at Cosma. For the system behind how we judge what will win, read my performance creative framework.
— Stefano Mazzei
FAQ
What is an example of an ad copy?
A simple PAS example: “Your back aches after a long day. Our lumbar-support seat pad is designed to support your posture. Try it risk-free.” It names the problem, adds a consequence, and offers one clear solution with a single call to action.
What are some effective copywriting frameworks?
AIDA, PAS, Before-After-Bridge, 4Ps and Hook–Story–Offer are the ones I’d test first. Choose by campaign objective and by whether the message should rely on emotion or rational proof. Research on gain versus loss framing supports matching the frame to the appeal type.
How do I know which ad copy framework is winning?
Hold everything constant except the framework, then judge by the metric that matches your goal: CTR for attention, conversion rate and cost per acquisition for sales. Wait for enough conversion events per variant before calling a winner, then test a new hook against it.
Is AI replacing copywriters?
AI speeds up drafting and variant generation, but it can’t decide which objective, proof or audience barrier matters. The human job shifts toward choosing the framework, supplying real evidence and judging test results.
Sources
- Gain vs. Loss Framing in Rational and Emotional Appeals: Effects on Consumer Attitudes, Reactance, and Brand Choice. | Published in Marketing Management Journal
- The impact of linguistic features on CTR in Instagram ads: A study of supplement and cosmetic products | PLOS One
- About responsive search ads - Google Ads Help
