By Sofia Collins

Paid Social doesn't lend itself easily to cross-channel attribution, making it difficult to measure Social's true influence on a user. How to determine its full worth has become a recognised problem, especially as the average user interacts with six touchpoints before purchase (Marketing Week, 2017). This blog outlines some ways in which you can report more holistically to incorporate Paid Social into the picture.

Back to blog home

Paid Social channels

When it comes to attribution, Social often remains somewhat ‘anti-social’ – it doesn’t integrate well with other channels. You will have noticed that when reviewing your marketing across channels in Google Analytics (GA), DoubleClick Campaign Manager (DCM) or other media and analytics platforms, the numbers just don’t match. 

If you’ve read our blog on the reasons for this discrepancy, you will no doubt want to know how to work around these limitations to consolidate cross-channel measurement within one single source of truth. If you haven’t yet, we would definitely recommend having a quick read through first.

The Solutions

The solutions can be broadly categorised into two methods: (1) Leveraging Google Analytics to report on Social more holistically, and (2) Advanced statistical analyses to understand the true incremental value of your paid social activity. 

1. Leveraging Google Analytics to begin integrating Social with other channels

a) Conversion contributions in GA

Paid Social attribution in FB is the equivalent to looking at channels’ ‘contributions’ in GA: contributions are the number of converting journeys each channel took part in. You can get channel contributions data from the Top Conversion Paths report in GA. We can therefore use this view to compare Paid Social performance with all other channels’ applying the same logic:

Paid social conversion contribution in Google Analytics

However, if we were to report on Social’s contributions solely in GA, impressions would not be included and volumes would therefore be deflated (only click-through conversions).

As a workaround, Social view-through conversions data can be imported from Social bidding platforms instead of using the click-through conversions. This will provide a more accurate, like-for-like comparison between Social’s and other channels’ contributions, which can help inform cross-channel budget strategies:

Paid social conversion attribution including view-through conversions

If you include cost per channel, this provides an interesting metric - Contribution CPA, or cost per contribution.

b) Paid Social’s position in the conversion funnel 

GA's Top Conversion Path report can also be leveraged to gain insight into where Paid Social tends to sit in the conversion funnel.

This data allows you to identify how many times a channel was the first, middle, last, or only (immediate) touch point in a conversion journey:

Paid Social Position in Conversion Funnel

The position a channel appears most in usually means that is where it is most effective in driving conversions: those predominantly used as ‘last’ interactions are the lower funnel channels, whereas those most often a ‘first’ or ‘middle’ interaction sit higher up in the funnel. This provides interesting insights into the ways in which your customers interact with your ads and can be used to inform the types of creative messaging they should be displaying. 

And, whilst you may be most accustomed to viewing those channels that are most often last interactions as best, this perspective gives insight into the conversion paths that would have likely not begun if you were to remove those ‘first’ interaction channels. Each channel should be viewed as having a different kind of worth, with activity and budget planned in accord with this.

2. Taking things further with a custom Merkle solution outside of GA

At Merkle | Periscopix we have a custom solution in our econometrics toolbox to help you better understand the effectiveness of your Paid Social campaigns and navigate the data sharing limitations between GA and FB. 

We can perform Media Mix Modelling (MMM), a holistic, accurate, and actionable marketing analytics solution. MMM is a statistical analysis which involves multivariate regression on conversions to calculate the impact of spend variation and other marketing tactics, in order to optimise spend levels. It can then also be used to forecast the impact of these tactics on future sales. This will allow you to understand the true incremental value of your various Paid Social campaigns with more rigour and less joining of diverse datasets.

There are some caveats to be aware of with this more complex analysis:

  • It’s not possible without sufficient variance in spend, as the model won’t be able to confidently ‘understand’ how spend impacted conversions
  • There needs to be sufficient conversion volumes (>300-400 per month)
  • Significant correlation is not guaranteed
  • MMM does not measure long-term brand equity, hence it is limited to short to medium-term forecasting

So, if you are interested in understanding the incremental impact of your Paid Social activity with the aim of making its spend go further, reach out to the team.

Share this article