When planning advertising campaigns across different digital media, results are often reported separately for each channel. This can lead to duplication in reach figures, since the same person may see the ad on multiple platforms. This happens because it's common for different campaigns to reach the same people, creating redundancies in how reach is calculated effectively.
To address this challenge, Bunker uses the most accurate way to calculate net reach while avoiding counting the same person multiple times, by applying the Sainsbury formula.
How does the formula work? 🧪
Sainsbury starts from the idea that the probability of a user seeing an ad across different media can be represented simply: if they see it, it's 1; if they don't, it's 0. So when we combine multiple media (such as TV, internet, etc.), we calculate how many people actually saw the ad without counting duplicates.
The formula combines the reach of all the media a user was exposed to and subtracts the overlaps, as follows: 1 - [[1-A]*[1-B]*[1-C]*[1-D]*...[1-n]]
Here, A, B, C, D... are the reach percentages of each medium, calculated over a fixed population or target audience — not over the sum of individual reach figures. This formula helps us determine how many unique people saw the ad in total.
What universe is it calculated over? 🎯
Sainsbury is applied over a fixed population number: the target audience size defined in the settings.
This target can be defined at two levels:
At the brand level, in the brand settings section, in the "Target audience size" field. By default, the population of the configured country is used.
At the campaign level, in the "Potential audience" field within the campaign's general information. This value allows you to adjust the audience ceiling for campaigns with specific targets (for example, a regional activation), improving the accuracy of the estimate for that particular case. Using the corresponding checkbox, this field can be automatically adjusted to match the target defined at the brand level. If no value is entered, the previous Sainsbury formula is used (the sum of reach figures as the universe).
When a campaign target is defined, that campaign's Total Reach is recalculated using that ceiling as the fixed population.
Protection rule 🛡️
If the sum of the individual reach figures from the accounts or media exceeds the defined target, the total result defaults to the highest individual reach among the accounts, rather than returning a number higher than the target (which would represent an impossible reach).
Where do we apply it?
The Sainsbury's formula is applied in:
In the «Analytics» module (brand) to calculate the total Reach of Google Ads in the Paid Media tables but only at the Total Campaigns level since the Google API does not provide the Reach data at the Ad Group and Ads levels.
In Paid Media for the Paid account summary at brand level. Now you can see the total Reach between media.
Campaigns, when a Potential audience (target) is defined at the campaign level — that campaign's Total Reach is recalculated using that ceiling as the universe.
The calculation uses the reach reported by each medium as input, without additional corrections for possible overestimations by the platforms.
🚨Important: Google Ads does not provide the reach for the Search network so the network is not considered for the calculation of the Google Ads Total Reach or between media.
