Understanding last click attribution and where it fits into data-interpretation strategies is crucial for any marketing manager.
Making decisions without studying crucial data can lead you to incorrect assumptions which, in turn, could lead to loss of revenue. This is because a poor understanding of what creates conversions will always result in creating campaigns that are largely hit and miss.
On the other hand, utilizing last click attribution can often enable you to quickly get a clear overview of what works and what doesn’t in your efforts to convert prospects into customers. This is probably the reason why last click attribution is the most popular attribution model in existence.
However, popular doesn’t always mean “best” or “ideal”. In fact, to understand the pros and cons of last click attribution, you need to understand attribution models in general. So, let’s dive into that for a bit.
What is an attribution model in the first place?
An attribution model is basically a strategy that determines the way you assign credit for conversions across various customer touch-points.
An attribution model may encompass paid search, email, social media, organic search, referrals and other digital channels. It also attempts to measure the impact that each of these channels has on the achieved conversion.
Last click attribution is the most popular channel because its implementation and usage are incredibly simple. This might be okay for a small business, but if you are in charge of a medium-sized or big business, you really need to utilize the different models when appropriate. So, let’s look at some of them right now.
Different Types of Attribution Models
Designed to tell you which channels are most effective in increasing awareness of your brand or services. It does this by attempting to let you know where your client first came in contact with you.
This model doesn’t care if the client has gone through countless other touch points prior to the first purchase. It assumes that the first touch point is the crucial one, and that your marketing dollars should go there.
LCA assigns 100% of the generated revenue to the last click point made before the purchase.
Let’s say that you send out a newsletter to your list and you announce your latest service. You include a banner announcing a promotional discount for the first 50 businesses that sign up. If people click that banner and purchase a plan, the banner gets all the credit.
The beauty of last-click attribution lies in its elegant simplicity and the fact that it provides you with a rather decent way of knowing which campaigns are most effective in converting prospects into purchasing customers. It’s also very easy to track and set up.
The downside of last-click attribution is that it completely steals the spotlight from earlier touch points which might have been just as crucial, or even more crucial in encouraging a person to make a purchase.
Let’s say that you get a really big business as a customer. Last-click attribution tells you that they clicked on an AdWords ad right before purchasing your enterprise plan.
However, what if this decision maker has actually been reading your newsletter for months? Your email marketing encouraged them to purchase. It’s just pure convenience that they were reminded by your AdWords ad to get it done. In such a scenario, LCA would falsely attribute credit to a channel that had little to do with their decision.
Some marketers prefer using linear attribution
Linear attribution provides an overview of the process that prospects go through before they make the decision to buy. This model aims to tell you which campaigns/channels are most used and most effective in moving prospects along the purchase funnel.
It does this by mapping out all of the touch points that happen between first click and last click, and assumes they are all equal in value.
For example, let’s say that your customer approached you 4 times before scheduling a sales meeting. Once via email, another time via social media, a third time via an AdWords ad, and a fourth time via third party content syndication. Linear attribution would credit each of these with 25% of the credit for the sale. Or in this case, for creating that sales lead.
There’s also the positional model
This is a model that combines the logic of all previously discussed models. It accepts that the first click and last click are indeed important. In fact, it assumes they’re the 2 most important points. But, it also assumes that other touch points in between deserve some credit, and it gives them credit in a linear way.
The exact percentage will depend on your analytics software. In general, though, a common formula is 40-40-20. This means that 40% of the credit is given to the first click, another 40% is given to the last click, and 20% of the credit is distributed evenly across all the other interactions.
The crucial difference between first-click and last click attribution
In essence, these two simple models have a different aim. First click aims to easily provide you with the information on how someone learned about your brand in the first place. On the other hand, last click intends to reveal what closed the deal. If you consider both metrics to be just as important, then the positional model is right for you.
The “Time Decay” attribution model
This is a model which assumes that the closer (in time) a touch point is to the sale, the more influence it has on that sale.
There is a notable difference between this model and the linear one. The older the marketing campaign, the less credit it will receive under a time-decay model. This is unlike linear distribution where each point gets an equal amount of credit.
The time decay model is valuable if you consider each touch point important, but after a certain point, you want to devalue past marketing efforts or give higher value to more recent ones. This model might be better for certain kinds of long campaigns. It really depends on your business model.
In some cases, the “last non-direct click” variant will be your attribution model of choice
This is a neat compromise that takes the simplicity of last click attribution while adding an additional element. It assumes that the most important touch point is the last one the customer made before they came to your website to purchase. Let’s illustrate this with an example.
Imagine you are running a LinkedIn campaign promoting your SaaS platform to potential customers. Let’s say a developer clicks on that ad, scans your pricing page for a couple of minutes and leaves. Then, a couple of days later, they decide to google your name because they want to make a purchase.
The last-click attribution model would credit the sale to that Google search. The last non-direct click model would, however, credit it to your LinkedIn campaign.
The “death” of last click attribution
As of late, there has been much talk about the “demise of last click attribution” and how other “superior” models are going to kill it. But a lot of that talk is nothing more than unnecessary hyperbole. It all depends on your business size and marketing power.
Some marketing campaigns can get away with relying on last-click attribution alone. Other types of campaigns, however, demand a mix of models if they are to have any chance of performing well.
While we wouldn’t say that last click attribution is “dead” or “completely useless”, there is something important you should consider.
If you’re in charge of marketing for a big business, you absolutely can’t rely on using LCA all the time. If you’re running serious campaigns with serious budgets, you are required to go beyond simple models. A small business may get away with keeping operations simple, but a big business cannot afford this luxury.
You need to understand all of the models within a broader context. And if you want to fully optimize your budget, you need to understand your customer’s typical buyer’s journey. Yes, it might sound a bit “vague” and general. Unfortunately, this is a complex subject.
A lot of this will depend on your particular business model and the product or service being marketed. There are literally thousands of possible scenarios and going into each one is beyond the scope of this general overview.
It doesn’t have to be overwhelming and confusing
Monitoring and analyzing all these metrics and data points can actually be easy if you ask for professional help. Whether you want to completely outsource this process or just need help implementing it, consider talking to a specialist.
Sifting through the data generated by your digital marketing efforts is often a strenuous task if you’re new to advanced modeling and analysis. Having professionals analyzing the data and using it to apply cutting-edge strategies to high-performing campaigns is often invaluable. In fact, it’s probably what your top competitors are doing right now.
At DemandZEN, we have many years of experience helping big businesses create and monitor digital marketing campaigns. Chances are, we’ve already worked with a business just like yours.
Whether you need help analyzing and improving your existing campaigns, or want to outsource any aspect of your digital marketing strategy, feel free to reach out at any time. Our team might just be that crucial piece which finally puts your marketing into overdrive.