Aggregatable Metrics

When we talk about aggregatable metrics, we're referring to metrics which can be correctly aggregated over a given period of time. An example would be a custom metric to calculate cost plus VAT (cost x 1.2). We'll call this metric "Cost incl. VAT".

This metric can be summed up over a given number of days. (The calculated metric is visualised to the left, the calculation to the right):

Jan 01: Cost incl. VAT = 12                        (10 (cost) x 1.2 (VAT))
Jan 02: Cost incl. VAT = 15,6                    (13 (cost) x 1.2 (VAT))
Jan 03: Cost incl. VAT = 13,2                    (11 (cost) x 1.2 (VAT))

Jan 01 - 03: Cost incl. VAT = 40,8             (12 + 15,6 + 13,2) - Correct (Total per day)
Jan 01 - 03: Cost incl. VAT = 40,8             ((10 + 13 + 11) x 1.2) - Correct (Total per period)

Non Aggregatable Metrics

Non aggregatable metrics are metrics which cannot be summed up correctly over a given period of time. An example would be CPC (Cost per Click).

Jan 01: Cost per Click = 12                         (120 (cost) / 10 (clicks))
Jan 02: Cost per Click = 5,71                     (200 (cost) / 35 (clicks))
Jan 03: Cost per Click = 10                        (60 (cost) / 6 (clicks))

Jan 01 - 03: Cost per Click = 9,23             ((12 + 5,71 + 10) / 3) - Incorrect (Total per day)
Jan 01 - 03: Cost per Click = 7,4                ((120 + 200 + 60) / (10 + 35 + 6)) - Correct (Total per period)

Unique Metrics

These metrics are usually primarily tied to a unique user across time such as Reach and Frequency. This means that calculating such a metric will have to take one or more unique users and how they relate to that timespan into account. Let's look at an example of the difference between a unique and an aggregatable metric. 

You're running a campaign called 'My Campaign', in which you have included several ads. A user gets presented with one of your ads on a Monday and clicks it. The platform recognises the click and adds one click to the "number of clicks". The platform also recognises the unique user (most often by the help of cookies) and also adds one unique click to the "number of unique clicks". On Tuesday, the same user sees your ad again and clicks it once more. Now the platform you used to display your campaign and ad can present you with the following statistics:

Monday
No. Clicks: 1
No. Unique clicks: 1

Tuesday
No. Clicks: 1
No. Unique clicks: 1

Monday & Tuesday
No. Clicks: 2
No. Unique clicks: 1

Keep in mind that the platform is able to present the number of Unique clicks for the two days (Monday to Tuesday) with the help of the platform's own cookie tracking. Funnel, on the other hand, has no possibility to relate the unique clicks to platform specific cookies. Instead, it is built to import both the regular Clicks and Unique clicks segmented by day and thereafter store them in a database. 

This means that when requesting stats for the time period Monday to Tuesday we would show 2 regular Clicks, as well as 2 Unique clicks. (Remember: we store the metric values per day and aggregate them when looking at longer time periods, without the knowledge of what click that belongs to what user, just which dates they occurred on). 

Not only would unique metrics in Funnel include inaccurate values when looking at multiple days, but also when observing the performance of several campaigns (or any other dimensions such as affiliate, campaign, ad set, etc.) at once. This is because a user could have seen ads from multiple campaigns, something the advertising platform would be able to recognise thanks to their user data, but not Funnel. 

Did this answer your question?