Advertising opportunities are often compared using numbers that are easy to understand.
Annual attendance. Daily traffic. Event registrations. Membership. Impressions. Number of screens. Number of rounds played.
These measures help establish scale. An event attended by 50,000 people reaches more people than one attended by 5,000 if those figures actually represent unique attendees. A roadway carrying 80,000 vehicles a day offers more potential exposures than one carrying 8,000. A venue hosting hundreds of thousands of visits each year operates at a different scale from a neighborhood business serving a few thousand.
Scale matters.
It just doesn’t answer the question advertisers are usually trying to answer.
An advertiser rarely wants to reach the largest number of people regardless of who they are. It wants to reach enough of the people relevant to a particular objective, under conditions that make the opportunity worth considering.
That makes audience size and audience match fundamentally different characteristics of an advertising opportunity.
Consider a regional home-services company trying to reach homeowners within its service area. A large event attracting 100,000 visits may initially look more attractive than a collection of community venues generating 25,000 visits.
But suppose only a small portion of the event’s audience appears to come from the company’s service territory, while the community venues draw heavily from the markets the company serves and have strong homeowner representation.
The event is still larger. That fact hasn’t changed.
The larger audience is still larger. The smaller audience may be more relevant.
Understanding the difference requires being more precise about what audience evidence can—and cannot—tell us.
Audience Fit is different from audience scale
In the Audience Intelligence methodology, Audience Fit describes how well the audience characteristics supported by evidence correspond to the audience an advertiser has defined for a particular objective.
Article 6 introduced an important distinction inside that target. Some characteristics are Required: they have to be satisfied for the audience to be relevant to the objective. Others are Preferred: they strengthen the match without determining relevance by themselves.
Suppose a residential contractor requires geographic compatibility and homeowner representation while preferring higher household incomes and a particular age range.
An opportunity that aligns strongly with the Preferred characteristics but clearly draws its audience from outside the contractor’s service territory has not overcome the failed geographic requirement. Several attractive preferences cannot compensate for evidence that a Required condition is not satisfied.
Missing evidence is different.
If another opportunity has strong homeowner evidence but does not report customer geography, we cannot conclude that the geographic requirement fails. We also cannot assume that it passes.
The condition is unknown.
That distinction matters because Audience Fit should preserve three very different situations: evidence supporting alignment, evidence showing non-alignment, and insufficient evidence to determine either.
Audience scale answers another question entirely. It may describe unique people, attendance, visits, impressions, traffic, memberships or another measure of potential exposure. Whatever the measure, it tells us something about how much audience or exposure exists.
It does not tell us how well that audience corresponds to the advertiser’s target.
Profile alignment is not target prevalence
Suppose a venue reports the following audience characteristics:
| Audience characteristic | Reported share |
|---|---|
| Age 35–54 | 60% |
| Homeowners | 80% |
| Household income $150K+ | 30% |
| Healthcare professionals | 20% |
Now suppose an advertiser is interested in homeowners between 35 and 54 with household incomes above $150,000.
The venue clearly has evidence that each of those characteristics is represented in its audience.
What it does not necessarily know is how often those characteristics occur in the same people.
The 60 percent who are between 35 and 54 may not be the same people as the 80 percent who own homes. The 30 percent above the income threshold may overlap substantially with both groups, only one of them, or neither to the extent an advertiser might assume.
The percentages cannot simply be added. Nor can they be multiplied together unless there is evidence supporting the assumptions required to do so.
If the venue has respondent-level survey data in which age, homeownership and income were measured for the same respondents, it may be possible to calculate the proportion satisfying the combined definition directly. Customer-level records can sometimes support the same analysis.
Aggregate distributions cannot reconstruct those relationships after the fact.
This creates an important distinction in audience analysis.
An opportunity may show strong profile alignment because several evidenced characteristics correspond well to the advertiser’s target. That does not automatically tell us the target prevalence—the proportion of the audience satisfying the complete target definition.
That is the difference between profile alignment and target prevalence.
The distinction becomes more important as targets become more specific. A business might want homeowners of a certain age, above a particular income threshold, in selected communities, working in certain occupations and displaying particular interests. Separate evidence may exist for many of those characteristics without establishing how frequently the complete combination occurs.
Target specificity does not create joint evidence that was never collected.
A responsible audience analysis should preserve that distinction rather than allowing a matching score to quietly turn into an estimated population.
Evidence coverage matters too
There is another problem hiding inside audience matching.
Suppose two opportunities are being compared against a target containing five meaningful characteristics.
Opportunity A has evidence for only one: age. Its age distribution aligns extremely well with the advertiser’s target.
Opportunity B has evidence for four of the five characteristics. Its age alignment is somewhat weaker, but it also provides useful evidence about homeownership, geography and occupation.
Calling Opportunity A the stronger audience match because its one known characteristic happens to align more closely would reward missing data.
That is why Audience Fit needs to be considered alongside evidence coverage: how much of the advertiser’s target can actually be evaluated from the evidence available.
Coverage does not tell us whether the audience is a good match. It tells us how much of the target we are in a position to judge.
An opportunity can therefore show strong alignment on the characteristics we can observe while still having limited evidence coverage. Another may provide broader coverage with a more mixed pattern of alignment.
Those are meaningfully different situations.
Required characteristics make the distinction especially important. If a Required characteristic is known to fail, that is evidence of non-alignment. If it has not been measured, the uncertainty should remain visible rather than being converted into either a pass or a failure.
Preferred characteristics work differently. Strong evidence across several Preferred characteristics can strengthen the case for alignment, but those matches should not conceal a failed Required condition.
The point is not to penalize opportunities simply because some audience information is unavailable. Real-world audience evidence will often be incomplete. The point is to prevent incomplete evidence from creating unwarranted confidence.
The evidence available to evaluate the audience is part of the decision.
Target prevalence requires the right evidence
When joint evidence is available, target prevalence can add something genuinely useful.
Suppose an advertiser’s complete target is defined by three conditions and a respondent-level survey shows that 40 percent of an opportunity’s surveyed audience satisfies all three. A second opportunity, measured comparably, shows 5 percent.
If the first opportunity has an estimated unique audience of 10,000 people and the second an estimated unique audience of 50,000, a rough calculation would imply approximately 4,000 and 2,500 target-matching people respectively, subject to the quality and representativeness of the underlying evidence.
The larger audience is still larger. The smaller one contains the larger estimated number of people satisfying the defined target.
But even this calculation requires care.
An annual attendance figure is not automatically a count of unique people. Ten visits by the same person are ten visits, not ten audience members. Impressions represent exposures rather than necessarily distinct individuals. Golf rounds measure rounds played. Traffic counts may measure vehicles passing a location rather than identifiable people.
A prevalence estimate describing people cannot simply be multiplied by an unrelated exposure measure and relabeled as a count of unique target people.
The units have to remain visible.
In some settings, the most defensible statement may therefore be that a particular share of surveyed respondents satisfies the target while the opportunity generates a separate number of annual visits, rounds or exposures. Combining those measures may require assumptions the evidence cannot support.
Complete target prevalence is powerful when the underlying evidence genuinely allows it to be calculated. It should not be manufactured when it does not.
Uncertainty should remain visible
Real-world advertising environments will rarely provide identical audience evidence.
A professional association may have unusually detailed information about members’ occupations and industries but limited household information. A golf course may have good evidence about customer geography, household characteristics and recurring relationships. A community venue may have survey evidence across several demographic characteristics. A roadside advertising opportunity may provide large exposure estimates while knowing relatively little about the individuals represented by them.
That variation does not make comparison impossible.
It means the comparison needs to preserve what kind of claim the evidence supports.
For some opportunities, we may be able to say that several characteristics align strongly with the advertiser’s target. For others, we may know the prevalence of a combined target directly. Some may have excellent evidence coverage but mixed alignment. Others may appear promising on the limited characteristics available while leaving important Required or Preferred characteristics unknown.
The objective is to make the uncertainty legible.
This is why audience size, Audience Fit, evidence coverage and target prevalence should remain distinct.
Audience scale tells us how large the measured audience or exposure is.
Audience Fit tells us how well the evidenced audience corresponds to the advertiser’s Required and Preferred characteristics.
Evidence coverage tells us how much of that target can actually be evaluated.
Target prevalence tells us what share of the audience satisfies the combined target definition—but only when the evidence supports evaluating those characteristics jointly.
Each answers a different question. None should quietly stand in for another.
And even together, they do not determine whether an advertising opportunity is the right one to buy.
An opportunity with strong Audience Fit may be too expensive. It may offer insufficient exposure, inconvenient timing or limited creative possibilities. Another opportunity may have somewhat weaker audience alignment but provide repeated exposure, stronger geographic compatibility, a more useful advertising format or economics that better suit the campaign.
Audience Fit evaluates the relationship between an evidenced audience and an advertiser’s target. It is one part of evaluating the opportunity, not a verdict on the opportunity as a whole.
That leaves another characteristic of real-world advertising that audience profiles alone cannot describe.
The same person can encounter advertising while commuting to work, attending a professional conference, participating in a community event or spending an afternoon at a golf course. The person’s age, occupation, household and other audience characteristics may be unchanged.
The circumstances of the exposure are not.
Understanding that difference requires moving beyond who the audience is to examine where, when and in what relationship to the environment that audience is reached.