Giving Data Driven Attribution a Chance to Shine
You’ve decided to move away from Last Click attribution and to utilise a custom Data Driven Attribution Model (DDAM) instead - hurray! Nam showed us more about DDA models and Harriet explained about the move from a last click model in previous blogs. Here are a few reasons to make the switch:
- Optimise performance based on the true incremental value of a channel.
- Reduce under / over weighting channels that are higher / lower in the conversion funnel.
- Optimise spend based on the data in DDA conversion reports by incorporating a DDA CPA.
- More holistic view of channel performance.
Visuals speak louder than words so a dashboard is a great way to show the impact your DDA model is having on the attribution of channels; people will appreciate its simplicity and it will help senior stakeholders to understand attributed conversions.
It can be challenging and tedious to export attributed performance out of platforms due to there being no native connections, which can lead to lots of data exporting, manipulation and formatting: in GA there is no API for the Model Comparison Tool data (or any of the other attribution reports). Data Studio (DS) allows you to seamlessly integrate data from platforms within the Google Marketing Platform, making it a good choice to start utilising your data and building out dashboards.
Getting Data Driven Attribution Conversions into Data Studio
Now that you have a DDA model in one of your platforms, you’re probably wondering how best to present this information. DS a great place for attribution dashboards. Let’s say you have a DDA model set up in your Campaign Manager (CM) account and you want to report on campaign performance for each channel. DS uses connectors to connect to platforms, ranging from the Google stack - e.g. SA360, DV360, BigQuery - to other partner connectors, such as Adobe Analytics.
To set up a connection:
- Select the CM connector
- Select the profile and advertiser(s) that you want to import from
- Press ‘connect’!
A popular choice is to add your data via Google Sheets as this give you the flexibility to manually manage your data. At present, getting DDA attributed conversions out of the Model Comparison Tool on a daily basis would be incredibly manual and time consuming. Here at Merkle | Periscopix, we can build a bespoke script that can pull unsampled GA360 DDA reports and export this to BigQuery, making daily DDA insights an automated process. Easy-peasy.
Once you have created a connector to your DDA data, you will see that it appears as a data source in DS. From there you can add multiple data sources to your blank report and start creating magic!
Data Blending within Data Studio
Data blending is a new feature that allows you to merge up to four data sources. Using data blending, you can align different fields to create a more holistic report - for example, blending campaigns and placements from one data source with the respective cost in another source. This means a more meaningful DDA dashboard using key KPIs such as Return on Investment (ROI).
Transforming Numbers into Visuals
Once the data sources are in your dashboard report, consider how best to visualise the data. Don’t keep your fantastic visuals all on one page – split them out! Use multiple tabs to make the story of your data flow better; you can create navigation buttons to make the report sleeker and comprehensive. For a DDA dashboard, you could create the following tab structure:
- DDAM conversion report tab: which drills down into placements, campaigns and keywords.
- A “typical customer journey” tab: which could look at path lengths, time to conversion and more attribution focused analyses.
For DDAM conversion reports, you can look at the way DDA has re-distributed conversions between the channels, and across more granular dimensions, such as placements and ad groups. Using the date range comparison, you can look at month on month changes, as well as changes between DDA and last click.
The Customer Journey
Within the customer journey tab, you can create a funnel positions graph that shows the position of a channel in a converting path, in relation to the different stages of the marketing funnel (i.e. First, Middle, Last, and Immediate). Where a channel appears most in one position, it usually means it is most effective at that position. The underlying data can be gathered from the Top Conversion Path report in CM, split out into funnel positions per channel using formulas, and visually represented using a horizontal bar chart in DS.
The heatmap chart type can be used to represent cross-sell opportunities. One of the use-cases for this is to show how the different digital channels overlap one another.
Data-driven attribution can be presented in several different ways, all of which are dependent on your reporting needs. Data Studio is a great tool to create clear and engaging DDA reports, but there are other solutions out there. If your data driven dashboards aren’t dashing enough, or automation has you in a pickle, please don’t hesitate to get in touch with our team.