How agencies run competitor research for clients
A repeatable competitor research process for agencies: kickoff questions, building the competitor set, client swipe boards, monthly creative reports and a reporting template.
Updated Oct 27, 2026

On this page+
At many agencies, competitor research happens in bursts. Someone screenshots a handful of ads before a pitch, a strategist pulls examples for a quarterly review, and the rest of the time it lives in individual browser tabs. The work is useful, but it isn't cumulative, and its quality depends on who happened to do it that month.
Run as a repeatable process, competitor research becomes one of the most valuable things an agency delivers. It feeds briefs, explains creative decisions to clients and builds a record that survives staff changes. This guide sets out a process that works for most performance agencies, from kickoff to the monthly report, with templates you can adapt.
Why make it a process
- Consistency. Every client gets the same depth of research, whoever manages the account.
- Explainability. Clients see the reasoning behind creative recommendations, which makes approvals easier.
- Memory. Saved research outlasts the people who did it.
- Speed. New team members and new clients get up to speed faster with an established structure.
Step 1: the kickoff
Competitor research starts with questions for the client, ideally in the onboarding call. The answers shape everything after.
- Who do you consider your main competitors, and why?
- Which brands do customers mention when they compare options?
- Which competitors do you admire, and which do you think are overrated?
- What have you tried in ads that didn't work?
- What claims can and can't you make, for legal, regulatory or brand reasons?
- Which markets and countries matter most this year?
- What are your margin and discount limits?
Clients often name the competitors they worry about rather than the ones their customers compare them to. Both lists are useful, but the second usually deserves more weight.
Step 2: build the competitor set
A good set is small enough to track properly and broad enough to surface fresh ideas. Eight to twelve brands per client is a workable range for most accounts, split into tiers.
| Tier | How many | Purpose |
|---|---|---|
| Direct | 3 to 5 | Same product, same customer; the brands you're compared against |
| Indirect | 2 to 3 | Different product, same problem; a source of fresh angles |
| Category leaders | 1 to 2 | Larger brands setting expectations in the category |
| Out of category | 1 to 2 | Brands with excellent creative in an unrelated niche, chosen for format ideas |
Review the set each quarter. Competitors go quiet, new ones appear, and the client's priorities shift.
Step 3: set up swipe boards per client
Every client needs a dedicated board, so research doesn't get mixed between accounts. Inside it, use a consistent structure:
- Sub-boards by angle, format or funnel stage, whichever the team briefs against most often
- Tags for competitor, angle, offer, format, hook type and date saved
- A one-line note on every saved ad explaining why it was saved
The note matters more than people expect. Six months later, "strong demo in the first two seconds, running since spring" is useful; an untagged screenshot is not. Our guide to building a creative swipe file system covers naming and tagging in more detail.
Step 4: track continuously
Rather than starting from zero each month, keep a light ongoing watch:
- Set up alerts for each competitor so new launches are flagged automatically
- Run a short weekly sweep, around 20 to 30 minutes per client, to save notable new ads and long runners
- Log every launch with a date, so the monthly report can be built from the log rather than from memory
Give one person ownership. Often this is a creative strategist or a junior analyst supporting the media buyer. Without a named owner, the weekly sweep is the first task to slip.
Step 5: the monthly creative report
The monthly report turns observations into recommendations. A structure that works for most clients:
- Summary. Three bullet points the client should remember.
- What competitors launched. New concepts, offers and formats this month.
- Long runners. Ads still active after 30 days or more, and what they have in common.
- Angle and offer shifts. What changed compared with last month.
- Format trends. Movement in the mix of video, static, carousel and creator content.
- Implications. What all of this means for the client's creative.
- Proposed tests. Specific concepts for the next production cycle.
A summary table keeps the competitor section scannable. Copy this template into your report and fill one row per competitor:
| Competitor | New ads this month | Longest-running active ad | Dominant angle | Main offer | Format mix | What it means for us |
|---|---|---|---|---|---|---|
| Competitor A | ||||||
| Competitor B | ||||||
| Competitor C |
And a second table for the tests you propose:
| Test | Hypothesis | Inspired by | Format | Priority |
|---|---|---|---|---|
| Concept 1 | ||||
| Concept 2 |
The "inspired by" column keeps the link between research and briefs visible. It should point to a pattern, such as "demonstration angle used by three direct competitors", rather than to a single ad you intend to imitate.
Presenting insights to clients
How you present research decides whether it changes anything.
- Lead with implications. Start with what the client should do, then show the evidence.
- Separate observation from inference. "This ad has run for 60 days in four countries" is an observation. "This angle is probably profitable for them" is an inference. Label each clearly.
- Don't overclaim. You can't see a competitor's spend or ROAS. Say so, and present longevity and reach as signals rather than proof.
- Show competitor ads as reference only. Make it clear that proposed concepts will be original, and explain how they differ. Our guide to researching competitor ads ethically sets out where the lines are.
- Connect to the client's own data. Research is most persuasive when it explains something in the client's own results.
- Keep it short. Fifteen minutes on a monthly call is usually enough.
Common pitfalls
- The screenshot dump. Forty ads with no commentary isn't a report. Fewer examples with clear reasoning are worth more.
- A stale competitor set. If the set hasn't changed in a year, it probably no longer reflects the market.
- Research that never reaches the brief. Every report should end in proposed tests, and every brief should cite the research behind it.
- Copying instead of learning. Recreating a competitor's ad is an IP risk for the client and a reputation risk for the agency.
Running the process in PixelSpyer
PixelSpyer supports each step of this process. The brand tracker handles continuous monitoring with new-ad alerts, the swipe file gives each client their own boards for tagged, annotated examples, and the ad breakdown tool in the AI studio speeds up the analysis behind each monthly report. The Agency plan, at $179 a month, is aimed at teams managing several clients; see pricing for the full comparison.
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.


