Competitor ad analysis can help you decide what to test in your own campaigns. Start by comparing how other brands describe the buyer’s problem, what they promise, and the proof they show.
Those details reveal ideas to investigate. Public ad libraries show creative and some delivery information, but they do not reveal whether an ad earns a profit.
This guide shows how to choose comparable ads, record what you observe, and turn one finding into a creative brief. Your next test should answer a question about your own buyer and product.
The short version: Compare ads aimed at the same buyer and purchase decision. Record the source, date, hook, promise, proof, and destination, then separate what you observed from what you infer. Public libraries cannot tell you whether an ad is profitable. Turn one insight into a testable creative brief for your product, with a defined outcome.
Which competitors and ads should you compare?
Compare advertisers competing for the same buyer and purchase decision. Keep the market, product tier, and campaign purpose as similar as possible. Separate direct alternatives from brands you admire for their execution.
Start with a manageable sample. Three competitors can reveal useful differences without turning the exercise into an open-ended collection project.
A scheduling product for individual consultants should not copy an enterprise platform’s sales motion without examining the difference. Both sell scheduling. One might promise immediate setup, while the other needs to satisfy an IT buying committee.
Write the research question before opening an ad library. For example: “How do competing tools explain the handoff from a sales call to a booked onboarding session?” That question sets a useful boundary.
According to Meta’s description of its Ad Library, the library expanded to cover active ads from all Pages. Review the information available for your selected market and category.
Google’s Ads Transparency Center also supports advertiser research across Search, YouTube, and Display. Google describes regional, format, and delivery-date information. Those fields provide context, not a performance report.
What belongs in the evidence record?
Record the original source and observation date alongside the ad’s message. Capture the opening, promise, proof, format, and destination separately. Those elements answer different questions and can change independently.
Keep observations apart from interpretations. Another reviewer should be able to inspect the same material and see which conclusions need testing.
Save the library URL or available ad identifier. Add the country filter and visible status. An observation date means you saw the ad then; it does not establish continuous delivery or a particular spending level.
Use a compact record like this:
| Field | What to capture | Why it matters |
|---|---|---|
| Source | URL, date, market | Repeatable observation |
| Hook | Opening words or scene | Attention mechanism |
| Promise | Outcome and conditions | Reason to consider buying |
| Proof | Demo, evidence, testimonial | Support for the promise |
| Format | Screen demo, creator, static | How the argument is delivered |
| Destination | Offer and requested action | What happens after the click |
Do not collapse these fields into “good creative.” A polished demo may communicate little about the buyer’s problem. A simple static ad may make a precise promise that your current videos avoid.
Review the destination too. A free-trial ad leading to an immediate product experience serves a different decision from an ad inviting a consultation. Compare the whole promise before borrowing its opening.
What can a public ad library tell you about performance?
A public library can show creative and the delivery information available for that record. It cannot establish the business outcome behind the ad. Longevity, repetition, and apparent production effort are clues to investigate.
Keep unknowns visible. The advertiser’s objectives, customer value, conversion quality, attribution settings, and reasons for retaining an asset may remain unavailable.
An ad might remain active on a small budget. Several similar versions might reflect localization or placement requirements. A repeated message might be a brand decision rather than the result of a controlled creative test.
Record “three observed versions share this claim” when that is what the evidence shows. Avoid turning it into “this is their winning angle.” The second statement requires information the first does not provide.
Your campaign data answers whether a related idea works for your buyer and offer. Check its economics too. Our creative testing framework covers that validation step.
How do you turn an observation into a useful hypothesis?
Connect the observed pattern to a specific buyer concern, then identify what your product can demonstrate credibly. Propose a change that answers that concern while keeping the rest of the test interpretable.
The strongest hypothesis may come from missing evidence rather than a missing format. Ask what the buyer still needs to believe.
A public Notion ad shows how to separate observation from interpretation. We reviewed this verified Notion Labs, Inc creative on September 29, 2026, with Google’s region filter set to “Shown anywhere.”
Google labeled it Video and showed September 29 as its last-shown date. The preview stayed blank, so this record covers the visible headline and destination, not the video’s opening or demonstration.
| Observed element | Public evidence | Interpretation to examine |
|---|---|---|
| Visible headline | “Build, Plan, and Scale with Notion” | Broad work and growth promise |
| Format | Video label; preview unavailable | Opening and proof remain unknown |
| Destination | Enterprise page; “Request a demo” | Sales conversation as next step |
The linked Notion enterprise landing page connected the broad headline to an enterprise offer and demo request. That message-to-destination connection is observable. Its effect on conversion quality is not.
For a competing product, one hypothesis might compare a broad planning headline with a specific workflow demonstration. Use your own product evidence. This proposes a test; it does not establish a weakness in Notion’s campaign.
Write the proposed opening, the supporting demonstration, and the expected buyer response. Keep the audience, offer, and destination stable where practical. Define the primary outcome before production begins.
Do not copy a competitor’s testimonial, identity, distinctive artwork, or unsupported promise. Use the research to understand the question. Build your answer from evidence and assets your team can use.
What should the resulting creative brief contain?
The brief should connect the source observation to one buyer question and one proposed test. Include the evidence your product can supply, what stays fixed, and the outcome that would change your decision.
Name an owner for the next step. If you cannot assign the work, revisit the customer question.
A proposed brief could read: “Test our broad planning headline against a demonstration of one workflow. Keep the audience, offer, and destination unchanged. Compare qualified demo requests using the agreed measurement window.”
That example needs real product evidence and a planned test before launch. It does not predict that either version will win. Our AI creative briefing guide explains the remaining production inputs.
Use the creative approval workflow to assign review of claims and final exports. Competitor research does not replace your own product checks or approval record.
If you’ve collected the ads but aren’t sure what to test, we can help make that choice. Our AI performance creative team connects competitor research to ideas your own product can support.
Through the Free Creative Competitor Analysis, we’ll review up to 3 competitors and bring 3 test ideas to your working session. Together, we can think through which idea best fits your buyer, offer, and next creative brief.
