Spy on Competitors' Reviews: Mine Them for Ad Angles

To spy on competitors' reviews, read what their customers write on Amazon, Trustpilot, Google, Reddit and in the comments under their ads, and copy down the exact phrases people use about their problem, their hopes and their complaints. Then sort those phrases into pains, desires and objections, turn each cluster into an ad angle, and check the Meta Ad Library to see which angles your competitors already run and how long they have kept them live.

I do this before I write a line of copy for a new account. Reviews are a free focus group made of your competitor's paying customers. Below is the process, the sources, and the template I log everything in.

Why are competitor reviews so useful for ad copy?

Because customers describe the problem better than marketers do. A brand's own ads tell you how the brand wants to be seen. Its reviews tell you why people actually bought, what they expected, and where the product let them down.

That second part is the gold. Every complaint about a competitor is a promise you might be able to make. Every "I almost didn't buy because..." is an objection your ad can answer before anyone has to ask.

There is also the language itself. When a customer writes "I was sick of waking up with a stiff neck", that line is often a stronger hook than anything a copywriter would invent. This is what people mean by voice of customer copy: you borrow the words, not the product claims.

Where do I find competitors' customer reviews?

Start wide. Different platforms attract different kinds of reviewers, and each one gives you something the others miss.

A few notes from doing this a lot.

The brand's own site is curated. Those reviews are usually sorted in the brand's favour. Use them for desire language, not the full picture.

Amazon is where the detail lives. Even for a DTC competitor, check whether they or a similar product sells there. Amazon reviewers write about specifics: fit, durability, what broke.

Trustpilot skews toward service. People go there when delivery was late or a subscription was hard to cancel, which makes it the best place to find trust objections.

Reddit is the most honest. Search the brand name plus "reddit", or search niche subreddits for "alternative to". Comparison threads show what actually made someone choose.

Ad comments are the live feed. When a competitor's ad shows up in your feed, open the comments, and read their Page posts too. You will see doubt, price pushback, and customers answering questions on the brand's behalf.

Which reviews should I read first?

Read the middle first. Five-star reviews are often vague ("love it!") and one-star reviews are often about one bad delivery. The three-star reviews explain what people wanted and where the product fell short, which is the gap your offer can fill.

My reading order on any platform:

1. Three-star reviews, all of them if there are not too many.
2. Two-star and four-star reviews, looking for repeated specifics.
3. The longest five-star reviews, for the desire language and the "before" story.
4. One-star reviews, scanning for recurring failures rather than one-off disasters.

Sort by most recent as well as most helpful. Recent reviews tell you whether a complaint is still current or whether the brand already fixed it.

How do I turn review phrases into ad angles?

Most review research dies here: someone reads two hundred reviews and writes generic copy anyway. The fix is to log every useful phrase in a table as you go and let the patterns tell you the angles.

Here is the template I use. One row per phrase, copied word for word.

The phrases above are illustrative examples of the kind of thing you will find, not real quotes. Fill your own table with what your competitors' customers actually wrote.

A few rules that keep the table honest:

• Copy the phrase verbatim. The awkward wording is often what makes it feel real.
• Tag pain, desire or objection. Pains become problem-led hooks, desires become outcome-led hooks, objections become reassurance lines, FAQs or guarantees.
• Group before you write. Once you have a few dozen rows, sort by the angle column. The angles with the most rows are your strongest candidates.
• Only promise what your product delivers. If the review complains about something your product also does, that angle is not yours.

If you want a deeper walkthrough of the angle step itself, our guide to finding winning ad angles covers how to define and test them, and the post on pulling hooks from competitor ads shows how to turn an angle into opening lines.

How do I check which angles competitors already run?

Now cross-check your list against the Meta Ad Library at facebook.com/ads/library. Search each competitor, read through their active ads, and mark which of your angles they already use.

The useful signal here is longevity. The Ad Library shows "Started running on" for every ad. Advertisers keep paying only for ads that work, so an ad on the same angle that has been running for months is a strong hint that angle sells. Our explainer on Facebook ad longevity goes into how to read that date sensibly.

When I cross-check, each angle lands in one of three buckets:

That third bucket is why review mining pays off. Ad libraries show what competitors already say. Reviews show what customers want said that nobody is saying.

For reading the copy of the ads you find, the competitor ad copy analysis guide gives a simple framework for breaking down hook, body, proof and CTA.

How do I save the ads I'm comparing against?

Ads disappear from the Ad Library the moment the advertiser stops them, so I save anything relevant on the day I find it.

Here is how I do this step. I use our free extension, the Meta Ad Library Downloader, which adds a Download button to every ad in the Library. For a full competitor, its "Find all ads" mode scrolls the search and packs every running ad into one ZIP, along with a searchable swipe board (an index.html you can search by hook, CTA or advertiser) and a CSV of every ad. I paste the CSV next to my review table so the angles and the real ads sit side by side.

If you want other ways to keep copies, the full guide to downloading Ad Library videos covers the options, including screen recording and manual methods.

How do I turn the research into actual ads?

Lead with the customer's words. Put the strongest verbatim pain phrase, or a close paraphrase, in the first line of copy or the first two seconds of video.

Answer the top objection inside the ad. If reviews keep mentioning hard-to-cancel subscriptions, say how easy yours is to cancel. If they mention sizing, show the size guide.

Use your own reviews as proof. Once your ads run, collect your own customer reviews and feed them back in. Our roundup of testimonial Facebook ad examples shows formats that put real customer words on screen.

Test angles, not just headlines. One ad per angle, similar structure, so a winner tells you the message worked.

What should I avoid when mining competitor reviews?

• Quoting a competitor's customer in your ad. That review was written about another brand. Write your own line or use your own customers' words.
• Naming the competitor's flaws directly. Speak to the pain, not the rival.
• Treating one angry review as a trend. Look for repetition across platforms first.
• Stopping at research. Ship at least one ad per top angle.

FAQ

Is it legal to read and use competitors' reviews for ad research?

Reading public reviews to understand customer needs is standard market research. What you should not do is copy those reviews into your ads or present them as your own customers' words. This is general guidance, not legal advice.

Which reviews are most useful for ad angles?

Three-star reviews are usually the most useful, because they combine a clear desire with a specific complaint. Long five-star reviews are good for desire language, and recurring one-star themes reveal objections you should answer.

Can I quote a competitor's customer review in my own ad?

No. That review was written about another product, and using it as proof for yours is misleading. Use it to shape your angle and write your own copy, or quote your own verified customers instead.

How many reviews do I need to read before I see patterns?

There is no fixed number, but patterns usually start repeating after a few dozen useful phrases logged across two or three platforms. When new reviews stop adding new rows to your table, you have read enough for a first round of angles.

How do I check whether competitors already run the angle I found?

Search each competitor in the Meta Ad Library and read their active ads for that angle. Check the "Started running on" date: an angle in an ad that has run for months is probably working, while an angle nobody runs may be an opening worth testing.