How to make a truly effective measurement plan
Are you aware of the measurement of your marketing campaigns?
The most common question when it comes to marketing campaigns using ADV tools is:
Which campaigns really contribute to growing my revenue?
To answer this question, we need to understand that data measurement, from this perspective, is much more complex. It is not enough to know how many people clicked on the ad to define successful or unsuccessful a part of that campaign, composed, then, of a series of marketing actions.
The result achieved, in terms of conversion, depends on a multiplicity of factors and they all need to be analyzed. That is why it is necessary to keep attribution models in mind in order to understand what the merit of a conversion in a given campaign comes from.
Attribution models and conversion awareness
To determine the success or otherwise of a campaign, we need to analyze all target behaviors, whether explicit or not. It is important-for example-to understand whether a user arrives at our site in a direct way (i.e., by manually typing our site into Google) or in an indirect way (i.e., via advertisement).
Similarly, we should be clear whether the target arrives on our e-commerce because of a display campaign, on which, for example, he did not click, but saw in passing on facebook. Or even if the user comes to us via a retargeting campaign, perhaps following a period of time ranging from 15 to 30 days (time needed to economically mature the purchase decision).
All these factors cannot be considered of secondary importance, as they contribute to the design of an effective conversion strategy.
Let us now explore all the attribution models from which you can choose with Google ads:
- Google Ads Last click non-directed;
- Google Ads Click;
- Facebook ads Last interaction 7 days, 1 day view;
- Time decay model;
- Data-driven model.
What if I measure the conversion with Google Analytics 3?
In that case you need to know that there are other limitations. One of the most important is that Analytics 3 uses a tracking model based on cookies, consequently, if the target visits a site from PC, but decides to proceed to purchase a product from mobile, the tracking data will be different.
Thus, if your user purchases from mobile, but had previously viewed the site from PC, it will show that the ads campaign was successful from mobile.
Actually, however, today it is also possible to cross device with Google Analytics 3, that is, it is possible to see from which devices the user passed by, before arriving on our site. True, this feature of Google Analytics 3 is yet to be implemented and is much less effective than FB Analytics and Google Analytics 4.
Let's say that to get the most out of Google Analytics 3 you need to implement it.
Different was Facebook Analytics.
Good old Facebook Analytics, being it people based, used to be able to give you the exact name of who had generated conversion by reconnecting the exact cookie. With the IOs 14.5 update, this functionality was lost.
Conversion, as you can see, is based on the knowledge that you have to weave together a multiplicity of data. It has always been believed that data was the basis from which to analyze the success or otherwise of a campaign, but the real question that must be asked is:
How can data, with all its limitations, help me improve my performance?
By checking all key quality metrics, never leaving anything out. Only in this way can you figure out how to improve your future campaigns.
Use Google Analytics' data-driven, time-decay attribution tools to get a better idea of which channels contribute to conversion at various stages of the funnel, always remembering, however, that these are not absolute truths.
The accuracy of measurement also depends on the volume of traffic:
the more traffic, the more accurate the data will be.
The most important thing is to ask functional questions that will help us make practical decisions without wasting time and energy.
Trying, for example, many different marketing approaches without having carefully studied all the data we have at our disposal, does not help us understand how we can act in the future to be successful with our campaigns.
Moreover, we only risk not building the right offer for our target audience if we do not know their path specifically.
Below is an example that may help you better understand what we are talking about.
Let's say the case of an A/B test on two types of creative in a Facebook campaign.
The drill down consists of looking in detail at CTR of all clicks, CTR of clicks on the outbound link, time on page, additions to cart and Purchases.
This is how we are able to assess whether a creative creates interest but does not convert to clicks or whether a campaign brings in acquisitions but has a low CTR and therefore a more "attractive" variant can be thought of.
These individual details help us understand how best to structure our campaign.
.svg)
