Use case #1: Paid acquisition WoW analysis
Category: Acquisition |Solution: DataMa Compare | Type : Recurring | Client: Web Utilities | Extension: None
Tags: Tags: #Media #PaidSearch #TrafficManagement
Thanks to Datama, we are able to break down our SEA performance by type, by campaign and by date in order to see the hot spots. This allows us to look at our stats from a different angle than buying platforms and saves us time thinking about optimizations.
TotalEnergies is a major multi-energies company in France. Their gas & electricity activity relies heavily on their website for inbound subscriptions(B2C & B2B). The analytics team helps the business, and more particularly the Paid Media acquisition team, access and understand their data in order to improve their performance.
The analytics team used to provide numbers and reports from Google Ads and Bing Ads. They managed to go the extra mile by proactively providing insights on top of just the numbers. They were already using DataMa for product analysis and therefore implemented an acquisition use case to expand the range of usage on a weekly basis
In order to keep things simple and understandable for the client, the team started first with a relatively simple market equation.
The idea was to make classic KPIs like CPC (cost per click), Cost, and CVR (conversion) fit into a market equation to explain the absolute conversion variation.
They ended up with the following market equation:
2 observations on this market equation:
- The CPC (Cost per click) appears in the denominator, which is not the classic way that DataMa use cases are set up. Another, maybe more intuitive option from an analyst perspective would have been to use Click/ Cost (i.e. the number of clicks you get for 1€ spent). However, since the business (acquisition team) is so used to looking at Cost per Click, it’s much better to match industry standards to help understanding and adoption
- Subscription/Clicks is considered a good proxy for CVR. Of course, because of gaps between the advertising platforms and the web analytics tool, and especially because of cookie consent limitations, you might want to add a Sessions/Clicks ratio as an intermediary indicator in your monitoring. But again, to keep it simple, it’s a good option to start with this.
For much more complete market equation on media performance, read our article: How to track your media campaign with DataMa?
In this use case, the data originally comes from Google Ads, Bing, and Google Analytics and was already aggregated in a Big Query table, using funnel.io ETL
From there, we obviously need the metrics used in our market equation: Cost, Clicks, and Subscriptions
As for dimensions, the team is used to analyzing their performance by platform (Google vs. Bing), Type of search (Brand vs. non Brand), Device, and Campaign
Here is an example dataset
We ingest the data using the DataMa Prep connector for BigQuery
In DataMa Compare, the use case allows us to explain variations of subscriptions generated by paid traffic from both Google Ads and Bing Ads in an impactful way.
In the particular example of the dataset above, we see that the number of subscriptions has decreased, due to lower spends and higher CPC.
We can then get into the dimension analysis, where we understand that the end of the “Moving Summer” campaign, with lower CPC than the “Special Discount” campaign, explains both the decrease of cost and the increase of CPC due to a mix effect.
2The “Moving Summer” campaign, which had a very low CPC, has been replaced by the “Special Discount” campaign, which had a much higher CPC, resulting in a bad increase of CPC average
The report has been shared with the paid media team, with very positive feedback, both for the effort of proactively making the data more insightful, and for the clarity of the charts and comments. They now plan to integrate the report automatically within the data visualization dashboard (Reeport/Data Studio) used for the Paid Media report.
The value of approaching the Media analysis with the holistic methodology of the market equation has also triggered a willingness to learn more and multiply use cases across the organization. In order to spread the word to other teams, the analytics team is planning to attend the “problem solving” workshops proposed by DataMa in the next few months (These workshops are part of the Coaching Analytics offer of DataMa Solutions).