How to spy on competitor Facebook ads (ethically)
Competitor ad research is legitimate when you study patterns rather than lift work. Here is where the ethical lines sit, and a process for turning what you find into original creative.
Updated Oct 27, 2026

On this page+
"Spying" on competitor ads sounds shadier than it is. Marketers have always studied each other's work: flicking through magazine ads, walking competitors' shop floors, signing up to rival newsletters. Meta now publishes active ads in a public library, and ads delivered in the EU stay searchable for a year after they stop. Looking is not the ethical question. What you do with what you see is.
The short version: learn patterns, never lift work. This guide sets out where the lines sit, a process for mapping competitors without drowning in screenshots, and how to turn what you find into creative that is genuinely yours.
Where the ethical lines are
Competitor research becomes a problem in two ways. Either you take something that belongs to someone else, or you get information through means you shouldn't. Everything else is ordinary market research.
Fair game
- Browsing ads in public libraries such as the Meta Ad Library
- Noting the angles, offer structures, formats, hook styles and calls to action a competitor uses
- Tracking how often they launch new ads and which ones they keep running
- Reading public comments to understand the questions and objections their customers raise
- Visiting their website and buying their product the way any customer would
Off limits
- Copying creative. Reusing their images or video, re-cutting their footage, or recreating a shot so closely that it is recognisably theirs.
- Copying copy. Lifting their headline or script word for word, or rewording it so lightly that the source is obvious.
- Using their brand assets. Their name, logo, trademarks, product photos or distinctive packaging have no place in your ads unless you have clear legal footing, and rules on comparative advertising vary between countries.
- Implying affiliation. Anything that suggests you are them, are endorsed by them or sell their product when you don't.
- Scraping private data. Harvesting commenters' profiles, getting around login walls, joining private groups under false pretences or buying leaked data.
- Sabotage. Mass-reporting their ads, leaving fake reviews or engaging with their ads to distort their results.
This is not legal advice. Copyright, trademark and advertising law differ by market, so if a piece of creative feels close to the line, ask someone qualified before it goes live.
Learn from it, don't take it
| Element | Learn from it | Don't take it |
|---|---|---|
| Hook | The type of opening, such as a question, a bold claim or a demonstration | Their exact opening line |
| Visuals | Format, pacing, shot types, text overlay style | Their footage, photos or edits |
| Offer | The structure, such as a bundle, threshold or guarantee | Their exact wording or brand promise |
| Brand | Gaps in how the category is positioned | Logos, names, trademarks, packaging |
| Creators | The type of creator and delivery style | The same creator's face or script |
A process for mapping competitors
Unstructured browsing produces a folder of screenshots and very little insight. A simple process keeps the output usable.
1. Build the competitor set
Aim for five to ten brands across three groups. Direct competitors sell a similar product to the same customer. Indirect competitors solve the same problem in a different way. Aspirational competitors are larger brands in adjacent categories whose creative is further ahead than yours. The indirect group is often the most useful, because their ideas haven't been absorbed into your niche yet.
2. Pull their active ads
Search each brand in the ad library, note how many ads are running and record start dates. Our guide to using the Facebook Ad Library covers the search filters in detail. If the brand advertises in the EU, set an EU country to see inactive ads and reach estimates too.
3. Catalogue angles
An ad angle is the reason an ad gives someone to care. Tag each ad with one primary angle: problem and solution, social proof, price or value, identity, comparison, founder story or product demonstration. After twenty or thirty ads, patterns appear. Many niches lean heavily on just two or three angles, which is useful to know when you are looking for room to stand out.
4. Catalogue offers
Record the offer attached to each ad: percentage discount, bundle, free gift, free shipping threshold, guarantee, subscription saving or no offer at all. Note whether the offer sits in the creative, the copy or only on the landing page.
5. Catalogue formats
Note the format (image, video or carousel), the production style (creator-led, studio, founder to camera, motion graphics) and, for video, the rough length. A spreadsheet with consistent columns is enough:
| Competitor | Angle | Offer | Format | Hook type | Running since | Notes |
|---|---|---|---|---|---|---|
| Brand A | Demonstration | Bundle | Video, creator-led | Before and after | Three months ago | Several versions live |
| Brand B | Social proof | Free shipping | Image | Review quote | Last week | New test |
6. Flag what keeps running
Ads that have run for months, or that exist in many variations, are the ones a competitor is most likely funding because they work. They deserve the closest study, and they are the best evidence you have of which angles and offers actually resonate in your category.
Turning findings into original creative
Research only pays off when it changes what you make. A few principles keep the result original:
- Look for gaps, not templates. If every competitor leads with discounts, a guarantee or quality angle may stand out. If nobody answers the objection that keeps appearing in their comments, answer it.
- Start from your own customers. Reviews, support tickets, post-purchase surveys and sales calls give you language competitors don't have.
- Borrow structure, replace substance. If demonstration videos dominate, film your own product doing something theirs can't. The format is shared; the content is yours.
- Write a hypothesis for every test. "Competitors lean on price; we'll test a durability angle with a longer guarantee" is a test you can learn from. "Make something like Brand A's ad" isn't.
- Run the swap test. If you could put a competitor's logo on your ad and nobody would notice, it's too close. Go back and make it specific to you.
For help turning angles into openings, see our collection of ad hook formulas.
A quick check before launch
Before any research-inspired ad goes live, confirm:
- No competitor footage, images or copy appear in it
- No competitor name, logo or trademark is used without clear legal grounds
- Every claim is true for your product and can be supported
- Creators are your own, briefed with your own script
- The ad passes the swap test
How PixelSpyer helps
PixelSpyer is designed for the pattern-level research described here. You can browse advertiser brands to build a competitor set, follow them with the brand tracker so new launches arrive as alerts, and save examples to project boards in the swipe file, tagged by angle, offer and format. When it's time to brief, the angle finder and hook generator in the AI studio help you develop directions from your own product and claims rather than someone else's.
Written by
PixelSpyer Team
The PixelSpyer team studies millions of public Facebook and Instagram ads to work out what keeps ads running. We write up what we learn so you can test smarter. Questions or corrections? Get in touch.


