Not all benchmarks are created equal. Here's how Makemi builds your peer group — matching segment, platform, geography, and audience size — and what happens when the ideal match is too small.
Every benchmark, every comparison, every “how am I doing?” answer in your Makemi report starts with one question: compared to whom?
Makemi actually gives you two answers — and they serve different purposes.
Your competitors are the organizations you choose. The brands you track, the rivals you care about, the accounts your boss asks about in Monday meetings. You tell us who they are, and we benchmark you directly against them — side by side, metric by metric.
Your peer group is the one we choose. Using an algorithmic matching system, we scan thousands of profiles in our database to find organizations that share your exact profile — across multiple dimensions — regardless of whether you’ve ever heard of them. This catches the rising stars, the quiet disruptors, and the segment peers you didn’t know existed.
The competitor comparison tells you how you’re doing against the names that matter to you. The peer group tells you how you’re doing against the market reality — which sometimes includes competitors you should be watching but aren’t.
Here’s how the peer group matching works.
The Four Dimensions of Peer Matching
Every organization in our database is indexed across four dimensions. When we build your peer group, we match against as many of them as the data will support:
| Dimension | What It Means | Example |
|---|---|---|
| Industry Segment | Your specific sub-sector (e.g., “Craft Breweries,” not just “Food & Beverage”) | A brewery is compared to other breweries, not restaurants |
| Platform | The social platform being analyzed | Your Instagram performance is benchmarked against other Instagram profiles, not mixed with TikTok |
| Geography | The country where your audience (or business) is based | A UK brand is compared to other UK brands when possible |
| Audience Size | Your follower count, within a sliding window | A 10K-follower account is compared to accounts between 3K and 30K |
These four dimensions work together to isolate the most relevant slice of our database for your specific profile. Because the audience size window is custom to your follower count — not a fixed bucket — every peer group is unique to the profile it serves.
The Sliding Window: Why Audience Size Isn’t a Bucket
Most social media benchmarking tools assign you to a fixed follower bracket — “5K–10K followers,” “10K–50K followers,” and so on. Tools like Socialinsider, Rival IQ, and Hootsuite all use this bucket approach in their industry reports.
There’s a problem with buckets: a profile with 9,900 followers and one with 10,100 followers end up in different peer groups despite being nearly identical in reach. The boundary is arbitrary, and your benchmark changes depending on which side you fall on.
Makemi uses a sliding window instead. For any profile, we include peers whose follower counts fall between 0.3× and 3.0× your own:
- If you have 10,000 followers, your window is 3,000 to 30,000
- If you have 100,000 followers, your window is 30,000 to 300,000
This means every profile gets a custom range proportional to its own size, and no one falls through a bucket boundary.
Logarithmic Weighting: Bigger ≠ More Important
Within your sliding window, not every peer gets an equal vote. A peer with 3,000 followers is meaningfully different from one with 30,000 — even though both fall within a 10K-profile’s window.
We apply a logarithmic weighting formula that gives the strongest influence to peers closest to your own size, then gradually reduces influence as the size gap widens:
| Peer Size (vs. You) | Weight |
|---|---|
| Same size (1×) | 1.00 (full influence) |
| 2× larger or smaller | 0.59 |
| 5× larger or smaller | 0.50 |
| 10× larger or smaller | 0.50 |
This keeps your benchmarks grounded in organizations operating at a comparable scale, while still allowing broader data to contribute when needed.
Confidence Tiers: When the Ideal Group Is Too Small
The four-dimension match is the ideal. But for niche segments in smaller countries, a perfect 4-dimension match might only surface 2 or 3 peers — not enough to draw meaningful conclusions.
When that happens, Makemi doesn’t fail silently or fabricate data. Instead, we relax the matching criteria in a controlled cascade and tell you exactly which tier was used:
| Tier | Dimensions Applied | Minimum Peers | Confidence | What It Means |
|---|---|---|---|---|
| Tier 1 | All 4 (Segment + Platform + Geography + Audience Size) | 8+ | High | Your peer group is statistically robust. Benchmarks reflect a tightly matched set of comparable organizations. |
| Tier 2 | 3 (Segment + Platform + Audience Size) | 5+ | Moderate | Not enough peers in your country alone, so we expanded to all geographies. Your benchmarks are still anchored to your segment, platform, and audience size — just with a global peer set. |
| Tier 3 | 2 (Segment + Platform only) | Any | Directional | A niche segment on a smaller platform. We dropped both geography and the audience size window to gather enough peers. Benchmarks are directional — useful for context, but interpreted with appropriate caution. |
Why Geography Drops First
When we need to loosen the criteria, geography is the first dimension to go. Here’s why:
Social media content strategy patterns — posting cadence, engagement dynamics, content mix — tend to be shaped more by industry and platform norms than by national borders. A craft brewery in Melbourne faces the same Instagram algorithm and audience behaviors as a craft brewery in Portland. The segment + platform combination is the strongest predictor of benchmark-relevant performance.
Audience size drops second because scale directly impacts engagement dynamics. A 5,000-follower account and a 500,000-follower account operate in fundamentally different realities — even within the same industry and platform. Preserving the audience size window matters more than preserving geography.
What You’ll See in Your Report
Every metric card that uses peer benchmarks will indicate which confidence tier was applied. Look for:
- “High confidence” — Your peer group is 8+ organizations matched on all 4 dimensions. These benchmarks carry the strongest statistical weight.
- “Moderate confidence” — Your peer group draws from a global (all-country) pool but stays locked to your segment, platform, and audience size.
- “Directional” — A smaller peer set from segment + platform only. Useful context, not a precision instrument.
In all cases, the peer group is displayed transparently — you can see exactly how many peers contributed and which dimensions were active.
Learn More
- Peer Group Benchmarking — How benchmarks appear in your diagnostic report
- The Vitality Score™ — How peer benchmarks feed into your overall health grade
Share this article
Written by David
David is the founder and CEO of Makemi. He's been a designer and marketer since the early 2000s and has driven organizational strategy since at least 2015. When he isn't working, David plays ultimate frisbee and spends time with his family. He wishes he still had time to play guitar.
