Are you using the correct attribution model?
Sales attribution and user journey conversion path optimisation are words that every online marketeer loves. Yet, not many companies manage to implement a clean and fully operational attribution model for their online <performance> marketing. The main purpose of marketing attribution is to quantify the influence each advertising impression has on a consumer’s decision to make a purchase decision, or convert.
Which attribution model is the best fit?
Whether you are using a Time Decay Model, Last Click, First Click or a “Classical” U-Shaped Distribution Model <40-20-40 rule>, you need to know this: if you do not track your conversions with ONE system – you are not going to get it “right”. In essence there is not real right and wrong when it comes to attribution modelling, there is merely the notion of understanding the limitations of all attribution model variations – and knowing how to apply them.
Whether you are using Google Analytics, Webtrekk, Omniture Sitecatalyst, or your own in-house data warehouse solution, make sure that the data is easy and clear enough to understand for everyone in your organisation. You need to be able to communicate clearly with the management team where the opportunities lay in your business model and your marketing teams need to be able to work with this data on a daily basis.
A multi-channel and multi-screen approach is essential to understanding how the customer makes his decision to purchase your product and finds his way to your product offering
Whenever you add a new marketing channel or change the initial setup of the ones that are currently running, you need to review the attribution model that you have in place. truly understanding your data will bring you much closer to your customers than you have ever been, and will open up entirely new opportunities and ways of addressing your (potential) customer base.
Why is it so difficult to pick one model?
Most companies work with specialist teams in paid search, SEO, affiliate, display, PR and social media marketing. All of which love and prefer different attribution models. The affiliate and paid searchers among your team will probably prefer a last click attribution model, whereas the display, PR and SEO teams will prefer a first click “wins” attribution logic.
The campaign managers who are there to inspire potential customers <at the front-end of the user flow / click path> will surely prefer a model which is first click based, or at least attributes a substantial amount of value to the initial touchpoint with the user. Sales driven channels, those in which the customer has a clear intent to purchase a product already, will strongly prefer a last click attribution model.
Interestingly enough, most management discussions around attribution models in this day and age, still revolve around the simple discussion of first click vs last click. A <sole> last interaction (click) model will attribute the highest value to the click that is most closely <right before> the transaction takes place. More advanced models such as Time Decay, will attribute a part of your sales or registration data to each contact point throughout the customer user journey.
Budget allocation through Sales Attribution
As an online marketing manager or CMO, it is essential for your budget allocation decisions to understand which combination of channels brings you the customers that you truly wish to have. If you have understood the limitations of your model, then that is great – and basically all you need to run your daily operations.
There are a few freebie tools out there that can give you a good indication of whether your channel mix is wide enough to justify a more thorough run through of various distribution models. Google Analytics (even in the free version) can for example do just that. You have to be careful when using this data as your single point of truth though, as the data in the free version of GA do not rely on raw data input, but on estimates.
The smaller your data set is, the more “unlikely” it is that the data that you see here can really be used. On the other hand, the larger your data set is, and the traffic volumes that you would like to look at, the more inconsistent Google Analytics tends to get. Larger time lags of up to 72 hours and jumps data statistics (on a daily basis) can occur when you use significantly larger traffic volumes.
If you are looking to grow your business and invest in new, innovative ways to address your customers, then I suggest you have a look at your attribution model more carefully. Automatic bid management based on a specific attribution model is great, but can also become quite a challenge to handle if it is not set up correctly.
You need to stay on top of your numbers and understand exactly what is going on and why. … As a performance marketeer; attribution modelling needs to be an essential part of your daily work routine. I am sure that there are a number of hidden gems in the numbers that you have right in front of you. Just dive right on in…!
GANDT Ventures: attribution modelling
Whether you have just gotten started with your business, or you are running a full-fledged Multi-Million $$$ turnover online shop, attribution modelling is key to your long term success. If you would like us to have a look at your current setup and help you understand the impact of your current attribution model, feel free to reach out to us at any time and we would be more than happy to help out.
