The Best Marketing Attribution Model For eCommerce?

If you’re wondering ‘What’s The Best Marketing Attribution Model For eCommerce?’ and clicked on the link to this page hoping for a short answer, you’re about to know what it’s like to be my mother; bitterly disappointed. Not only does this article have to be at least 500 words for SEO purposes, but the answer to the titular question is also relatively complex and there isn’t a one size fits all approach. Don’t let that put you off though. It’s actually surprisingly interesting and understanding attribution modelling is essential if you ever hope to gain a firm grasp of digital marketing and eCommerce.

What Is A Marketing Attribution Model?

Firstly, let’s quickly cover the basics to ensure everybody is up to speed. Marketing attribution modelling is all about assigning value and giving credit to the various marketing channels that feed a business. In the case of eCommerce websites, this means determining where those who placed an order have come from and what steps they took before their transaction took place. It’s not only about discovering how each buyer initially came to the site (though it can be that simple to some), but also looking at how many times they came and went before placing a purchase, whether they visited the site via different channels throughout their buying process, and how much time their considerations took.

Marketing attribution models allow businesses to more accurately assess the success of their marketing activities, however, what ‘success’ means can vary from business to business and even from channel to channel.

What Are The Marketing Attribution Model For eCommerce?

Before we begin looking at the different attribution models, let’s quickly go over a possible scenario. Please note that this scenario will be referred to in order to provide examples for the various attribution models.

‘You figure the weather is nice and fancy sitting in your garden, but soon realise you have no garden furniture (you silly goose). You hop online, google ‘rattan garden furniture’, then click on the first sponsored ad at the top. It takes you to ioliving.co.uk and you browse for a while but leave without placing a purchase. The next day, you’re on Facebook looking at memes while you’re supposed to be working and notice an ad from the furniture site you visited the day before. This ad has a voucher code for 20% off steel garden furniture, so you click on the ad and browse the steel garden furniture. Sadly, you have your heart set on rattan garden furniture, so you leave, dejected, without placing an order.

Four days later, you’re sat with your partner on the sofa and they say “you know what? I’d love some rattan furniture for our garden”. You’re just so in sync. You discuss this uncanny coincidence for a moment then grab your laptop to show them the site. You cannot remember the URL, so you Google something that you think is the name. The website appears at the top of the page, but there are no ads and you click on an organic listing that takes you to the site. You and your partner, still so in love, find the perfect rattan garden furniture set but it’s a little over your budget and for a second you think all is lost. BUT WAIT! You remember that there was voucher code available for 20% off steel garden furniture, so you go to vouchercodes.co.uk and see if there are any other codes available. You’re in luck! There’s a 15% off code and it’ll be automatically applied if you click the link to the site. You hurriedly click the link, your hearts bursting with joy, and finally place an order for £1275, down from £1500. The purchase has been made.’

Now, four channels (or ‘touchpoints’ if you’re feeling jargony) were used before an order was placed in this exciting and well-written example. Namely, Google Search (Paid), Facebook (Social Media), Google Organic (Organic), and vouchercodes.co.uk (Affiliate). The question we now have to ask ourselves is “how much credit does each channel deserve for the sale?” and that’s where your Marketing Attribution Model comes in.

Last Click Attribution

Perhaps the least insightful attribution model, Last Click Attribution gives all credit to whichever channel the site was visited through when the order was placed. In the case of our example above, 100% of the credit would go to the affiliate channel and every other step along the way would be essentially ignored. It is worth noting that no matter what attribution model you choose to use, affiliate marketing is always measured using Last Click attribution.

First Click Attribution

If the Last Click model is the least insightful then the First Click model is a close second. First Click Attribution gives all credit to the initial point of contact and essentially ignores the rest. Going back to our example, Google Search would get all credit for the transaction under the First Click Attribution Model and the other three would get nothing.

Linear Attribution

Easily the most diplomatic option of the lot, Linear attribution assigns an equal portion of the credit to each channel. If a buyer visits a site through 4 channels before placing an order, as is the case in the above example, then 25% of the credit is assigned to each channel along with 25% of the value of the transaction. If the site was visited through 5 channels prior to an order being placed, then 20% of the credit would be assigned, and so on.

Time Decay Attribution

With the Time Decay attribution model, channels used closest to the time an order is placed are given more credit than those the site was visited through earlier. The Last Click will always get the most credit, of course, but Time Decay Attribution does at least tip its hat to the other channels that contributed to the sale. Looking at our example, Organic and Affiliate would get the most credit as they were both used close together, Social would get less credit as the visit through Facebook took place 4 days before, and Google Search would get even less as it the Paid interaction happened 5 days prior to the sale being made.

Position-Based Attribution

Also known as 40-20-40, Position-Based Attribution gives equal credit to First Click and Last Click, usually 40% each, and divides the remaining 20% between the other channels that the site was visited through. The 4 channels mentioned in the example would be easy to divvy the percentages between, as Google Paid and Affiliate would each get 40% while Organic and Social would each get 10%. 40-20-40 is the default, but it is possible to customise the ratios. For example, you give the first click 10%, divide 30% amongst the channels in the middle, and give the last click 50%.

Why Do eCommerce Attribution Models Matter?

I am definitely over the 500-word minimum requirement for this article and could quite easily leave it at that. However, I’m regularly asked (relatively speaking) why any of this attribution modelling malarkey matters and frankly that vexes me. If there is one thing I’d like anyone reading this to take away with them, it’s that I don’t like to be vexed. No, sir/madam.

So! Why do eCommerce attribution models matter? Well, they matter because the majority of internet shoppers do not place an order during their first time to a site. This may be shocking to the impulse buyers amongst us (which, truth be told, I most definitely am), but a study carried out by digital services provide, Episerver, suggests that up to 92% of first-time site visitors don’t place a purchase. The hope is that at least some of these visitors will come back one day and place and place an order, and when that fateful day arrives, wouldn’t you want to know how they came to know of you? They may have placed an order after having clicked a link in an email newsletter you sent out, but wouldn’t you also want to know that they initially discovered your site through a paid social media campaign that you thought failed?

Effective attribution modelling provides contextual data that can be more readily analysed, and it’s not only helpful for eCommerce websites either. Any websites with even a single specified conversion goal can benefit from attribution modelling. As to which is the best marketing attribution model for eCommerce, it really does depend on the strategic goals of your business and website (why did you fool yourself into thinking the answer was going to be anything other than this? There’s no such thing as easy answers in this world).

If you’d like to discuss which attribution model would be best for your business’ website, please feel free to get in touch. You can contact a member of the Fuel Communications team via the contact form on our website or email me directly at alem@fuelcommunications.co.uk