Short answer: Start with each creator's expected relevant reach, not their follower count. For two creators, add their comparable reach and subtract the people shared by both. For larger rosters, pairwise overlap cannot fully account for people who follow three or more creators. Use qualified platform measurement when it covers the delivery you are buying. Otherwise show a low, base, and high overlap range instead of pretending you know one exact number.
A roster with five creators does not reach five separate groups of people.
Sometimes that is a problem. Sometimes it is the plan.
If a brand wants efficient awareness, paying repeatedly to reach the same small audience may be wasteful. If the product needs trust and several explanations, seeing it from three credible creators can be useful. The real question is whether the shared audience is doing a job in the plan.
Overlap is not inherently bad. Overlap you did not price or plan for is.
Define the audience before calculating overlap
"Audience" can mean:
- Followers
- Unique viewers in the last 28 days
- Viewers of the sponsored assets
- Newsletter readers
- Podcast consumers or downloads
- People in the target geography
- People who match an age, role, or interest requirement
Do not compare Creator A's followers with Creator B's monthly viewers. Choose one comparable base.
For organic creator planning, a practical definition is:
Unique people in the target market who are reasonably expected to receive the contracted content during the stated campaign window.
That number will usually need to be modeled from recent, comparable content because future organic delivery is not guaranteed. Use a range. A channel's last 28 days can be distorted by one breakout video, a quiet month, or content that looks nothing like the contracted post.
Name the input expected relevant reach, not audience size. It keeps follower count from quietly sneaking back into the math.
The exact two-creator formula
For two creators:
Deduplicated reach = Reach A + Reach B − Shared audience A∩B
Incremental reach from B = Reach B − Shared audience A∩B
Illustrative example:
- Creator A expected relevant reach: 100,000
- Creator B expected relevant reach: 70,000
- Estimated people reached by both: 25,000
Then:
Deduplicated reach = 100,000 + 70,000 − 25,000 = 145,000
Incremental reach from B = 70,000 − 25,000 = 45,000
The combined gross reach is 170,000. The estimated unique reach is 145,000. The 25,000 shared people may still receive useful repetition.
State which overlap rate you mean
Working with finance creators? Creators Agency manages a focused roster of verified finance and business YouTubers. Book a free strategy call to see who fits your brand.
Two vendors can report "30 percent overlap" and mean different things.
Common definitions include:
Share of the smaller audience
Overlap rate = shared audience / min(reach A, reach B)
In the example:
25,000 / 70,000 = 35.7%
This asks: how much of the smaller creator's audience is already in the larger creator's audience?
Jaccard overlap
Jaccard overlap = shared audience / deduplicated union
In the example:
25,000 / 145,000 = 17.2%
This asks: what share of the combined unique audience belongs to both?
Neither definition is wrong. A percentage without the denominator is unusable.
Why a five-creator roster is harder
For three audiences, exact inclusion-exclusion is:
|A∪B∪C| = |A| + |B| + |C| − |A∩B| − |A∩C| − |B∩C| + |A∩B∩C|
That last term matters. Pairwise overlap tables count a person who follows all three creators in each pair. Without the triple-overlap term, subtracting all pairwise overlaps removes that person too many times.
As the roster grows, higher-order intersections multiply. This is why an attractive heatmap of pairwise percentages is not an exact deduplicated-reach calculator.
Use one of three methods.
Method 1: platform or measurement-provider deduplication
When the campaign runs inside one paid media system, use the platform's unique-reach and overlap reporting where it fits the question. Google Ads can aggregate selected video campaigns into a deduplicated reach total through its reach-reporting tools. Display & Video 360 can report overlap, exclusive reach, and duplicate reach across supported paid-media dimensions.
That solves paid campaign measurement within the platform's scope. It does not automatically deduplicate the organic audiences of several creators before launch.
Ask:
- Does the measurement include organic creator delivery, paid delivery, or both?
- Which devices and placements are included?
- Is the target geography applied consistently?
- Is the measure planned, modeled, or observed?
- Has the methodology changed since the last report?
Method 2: privacy-safe matched audience analysis
Where creators and the brand have a lawful, consented, privacy-safe process, a clean room or approved measurement partner may estimate overlap from shared identifiers or platform data.
This requires legal, privacy, security, and data-governance review. Creators should not hand raw follower or customer lists to a brand. The useful output is an aggregate estimate, not a list of who follows whom.
Do not build a shadow identity graph in a spreadsheet because the media team wants a cleaner Venn diagram.
Method 3: low, base, and high scenarios
This is the best approach for many organic creator plans.
For each new creator, estimate the share of their relevant reach that is incremental to the roster built so far. Rank the evidence before turning it into a percentage:
- Strongest: platform or measurement-partner deduplication that covers the same delivery, period, geography, and audience definition
- Useful but directional: survey or panel data and shared traffic or affinity signals
- Historical evidence: deduplicated reach from comparable creator combinations measured under the same scope
- Creator-provided clues: crossover in YouTube's "channels your audience watches" report, audience geography, and other native analytics that match the target
- Qualitative clues: recurring names and relevant job titles in both comment sections, frequent collaborations, topic proximity, and the same communities appearing around both creators
- Weak on its own: a follower-overlap score with no explanation of the denominator, time window, or data source
YouTube says its "channels your audience watches" report covers channels viewers consistently watched outside the creator's channel over the previous 28 days. That makes it a useful signal of crossover, not a count of people the campaign will reach twice.
Keep an evidence note beside every assumption. Then model scenarios rather than one false-precision number.
Illustrative four-creator model
Creator | Expected relevant reach | Low-overlap incremental share | Base incremental share | High-overlap incremental share |
|---|---|---|---|---|
A, anchor | 100,000 | 100% | 100% | 100% |
B | 70,000 | 85% | 65% | 45% |
C | 50,000 | 80% | 55% | 35% |
D | 40,000 | 75% | 50% | 25% |
Estimated deduplicated reach:
- Low-overlap scenario: 100,000 + 59,500 + 40,000 + 30,000 = 229,500
- Base scenario: 100,000 + 45,500 + 27,500 + 20,000 = 193,000
- High-overlap scenario: 100,000 + 31,500 + 17,500 + 10,000 = 159,000
Gross expected reach is 260,000. The decision should survive the plausible range from 159,000 to 229,500. The percentages are illustrative, not suggested norms.
The spreadsheet layout
Create these columns:
Column | Formula or input |
|---|---|
Creator | Input |
Expected relevant reach | Input range from comparable analytics |
Fee and landed cost | Input |
Incremental share, low overlap | Assumption |
Incremental share, base | Assumption |
Incremental share, high overlap | Assumption |
Incremental reach, low | Reach × low incremental share |
Incremental reach, base | Reach × base incremental share |
Incremental reach, high | Reach × high incremental share |
Cost per 1,000 incremental relevant people, base | (Landed cost / base incremental reach) × 1,000 |
Evidence note | Source and date |
Order creators deliberately. The first creator is the anchor and receives 100 percent incremental credit in the model. Add subsequent creators in the order the brand is considering them. If changing the order materially changes the result, the model is telling you something useful: the roster may contain close substitutes, or the assumptions are not yet grounded well enough to support the decision.
Use overlap to make four different decisions
Add
The creator contributes enough relevant incremental audience or a distinct role to justify cost.
Replace
Two creators reach nearly the same people and perform the same job. Choose the better fit, evidence, creative, or economics.
Keep for frequency
The overlap is intentional. The audience may need several touchpoints, and the creators can approach the problem from meaningfully different angles. That is especially useful when you are testing different theses, calls to action, or explanations. Track total frequency and the listener or viewer experience so useful repetition does not become ad fatigue.
Keep for validation
The audiences overlap, but the brand needs to learn whether trust transfers differently across creators. This is not a pure reach buy. Label it as a comparison or persuasion test.
Organic and paid overlap must stay separate
A person may see:
- Creator A's organic video
- Creator B's organic post
- Creator A's video again as a paid ad
- A retargeting ad from the brand
Gross campaign impressions can rise while unique reach barely moves. Preserve organic delivery, paid creator amplification, and business-as-usual paid media as separate components before trying to deduplicate them. A paid-media overlap report cannot tell you how much of the organic creator audience saw the original sponsorship unless that organic delivery is explicitly part of the method.
The paid amplification guide owns platform permissions. The creator MMM guide explains how reach and frequency can enter aggregate modeling.
Do not solve overlap by picking unrelated creators
Audience separation is not the only goal. A creator whose audience has no reason to care about the product adds incremental reach that is worthless.
Relevance comes first. Then use overlap to decide whether adjacent relevant creators create efficient breadth, useful repetition, or redundancy. The pre-deal analytics guide owns the evidence to request; the micro versus macro budget framework owns budget distribution by campaign job.
Reporting after launch
Show:
- Gross organic delivery by creator
- Paid delivery by campaign
- Unique reach where a qualified source provides it
- Estimated deduplicated reach with the method, range, and evidence behind the assumption
- Average or distributional frequency where available
- Outcome by creator and for the roster
- The share of outcome that cannot be assigned at creator level
Do not back-solve an overlap rate from conversions. Two creators can reach the same audience and influence different people inside it.
Price the audience you are actually adding
Roster planning gets better when gross reach stops pretending to be unique reach.
Define the audience. Keep the period and geography comparable. Use exact math when the data supports it and scenarios when it does not. Then decide whether shared audience is redundant or useful frequency.
If your team is building a creator roster and wants a realistic reach plan, talk with Creators Agency.
Primary sources
Frequently Asked Questions
For two creators, add their comparable reach and subtract the people reached by both. For larger rosters, exact calculation requires higher-order overlap or person-level/platform measurement. Otherwise use scenarios.
It is a clue, not the same measure. Many followers will not receive the campaign, and non-followers may see it. Use active or expected reach when possible.
There is no universal good percentage. The answer depends on whether the campaign needs incremental reach, repetition, validation, or several trusted explanations.
No. Repeated exposure can be intentional. It becomes waste when the media plan assumes incremental reach but buys frequency without recognizing or valuing it.
Some discovery, planning, and paid media tools provide similarity or reach estimates. Coverage varies, and organic cross-creator overlap may remain modeled. Verify the metric and scope.
Do not add each platform audience as fully unique. The same person may follow the creator on several platforms. Use a cross-platform overlap assumption or provider with deduplicated measurement.
Ready to reach an audience that actually converts?
Tell us what your brand needs. We will help you plan the campaign and find creators who fit.
Work With Our Creators →