So, you’ve thought about using last-click attribution to measure the effectiveness of your marketing, but is using last-click attribution the most accurate way to see where conversions are coming from? Or does it only paint half the picture by ignoring the rest of your customer journey?
First things first, before we dive into last-click attribution it helps if I explain what an attribution model is.
An attribution model is a set of rules that determines how your analytics tool applies credit for clicks, form completions and conversions. Without an attribution model, it would be near impossible to measure the effectiveness of your marketing activity.
Attribution modelling allows you to calculate the ROI of your channels so that you can allocate your budget towards the areas of marketing that have the opportunity to generate revenue growth.
There are now many ways in which you can attribute credit to your marketing channels in analytics. I won’t go into this here as you’ve come to learn about the last-click model, but if you’re interested, I have an article on the most common attribution models which shares ways on how you can choose the right one for your marketing.
So, let’s take a look at last-click attribution and how it works in Google Analytics. Then, we’ll discuss the problems marketers face when relying solely on last-click attribution over other marketing attribution methods.
What is Last-Click Attribution?
Last-click attribution refers to the website analytics model where the last-click a user makes before converting is awarded 100% credit.
Tools like Google Analytics will reveal the last source a customer clicked-through before making a conversion, giving you some indication of the success of your marketing channels.
To demonstrate how the last-click model works, I’m going to refer to a sample of a buyer’s journey.
1. An anonymous visitor clicks on a Google ad and lands on a website.
2. They use a brand search in Google and visit a case study page where they sign up to a newsletter.
3. Two weeks later, they click through to a blog post via a newsletter.
4. After two months, they visit your website via direct. From there, they convert by clicking a button to schedule a meeting with sales.
As you can see in the diagram above, 100% of the credit has been assigned to the direct visit. Despite the visitor using other channels before converting into a lead.
If you were to use software such as Ruler Analytics, this is how the last-click attribution model would assign revenue:
Ruler Analytics Source Report
The good thing about last-click attribution is that it can show the final points in your customer journey. It tells you which channels are most effective at the bottom of the funnel.
You may have heard of the last non-direct click attribution model. This is essentially the same last-click attribution, but with a small exception. Last non-direct click attribution gives credit to the channel before direct because there isn’t much to learn from a user directly visiting your website.
For example, if a shopper clicks on your email newsletter, bookmarks your page and converts later by directly entering your web address, then the last non-direct click model will provide email with 100% credit for the conversion.
Marketers tend to use the non-direct model for longer customer journeys and non-impulse buys, as shoppers often bookmark a page because they need more time to make their decision.
For example, if you’re selling high ticket items the buyer may need to consult with their partner, work colleagues or conduct more research before purchasing.
Not using non-direct click attribution in these circumstances may skew your conversion data and misdirect your marketing team into thinking that only direct traffic converts and nothing else.
By default, Google Analytics uses the last-click attribution model. 100% recognition of any conversion is awarded to the last click before visiting your website.
According to the Google Analytics help page, “last-click attribution is most useful when your ads and campaigns are designed to attract people at the moment of purchase, or your business is primarily transactional with a sales cycle that does not involve a consideration phase.”
Google Analytics can award a last click conversion to the following traffic sources:
- Organic search
- Paid search
- Referral (other websites, directories and forums)
- Direct (user visits your site from a bookmark or directly enters your website URL)
- Other (Google Analytics is unable to recognise the traffic source)
What Is The Problem With Last-Click Attribution?
Data-driven marketing that uses the last-click attribution model is a flawed approach in today’s marketing landscape.
Deciding where to increase or decrease resources based on last-click attribution can have a devastating impact on revenue growth.
For example, let’s say I see an advert for a pair of boots on Instagram and click-through to the website to view more products.
I don’t buy anything on my initial visit because I need to do more research on the brand before I make my final decision.
Although, I do sign up to their email newsletter where I’m sent a 20% off voucher for my first order a couple of days later.
A week later, I enter the brand’s name into Google, find their organic listing and end up buying a product using their 20% off voucher.
If I were using the last-click attribution model in Google Analytics, then the credit for the conversion would go to organic search. Skipping both Instagram and the email newsletter which played a huge part in my customer journey.
If it wasn’t for the Instagram ad, I’d never have known about the brand and I wouldn’t have received the 20% off voucher sent by email.
Last-click attribution doesn’t paint a clear picture of where your sales are coming from or how your customers move through the funnel.
Below are two screenshots of the Ruler Analytics dashboard. They both contain data from the same account and have been set the same date range.
The only difference is that the first screenshot is assigning revenue using the first-click attribution model (first-click attribution model assigns 100% credit to the initial touchpoint). The second screenshot is attributing revenue using the last-click model.
Ruler Analytics Source Report – First Click
Ruler Analytics Source Report – Last Click
When you compare both screenshots, you can see a clear difference in revenue for both models.
Email works better near the top of the funnel, whereas Google Paid plays a bigger part near the bottom.
If you were just operating last-click attribution then you wouldn’t have this level of insight.
On the topic of attributing revenue, we have a framework that can tell you exactly how to measure and attribute sales revenue back to your marketing channels, campaigns and keywords which you may find useful.
You can download that here.
Should You Use Last-Click Attribution?
Thanks to the Internet and smartphones the customer journey has changed, 56% of online shoppers read online reviews before purchasing, for larger ticket items the buyer may need to consult their partner or research similar products to find the best solution to their problems.
Last-click attribution does not value earlier touchpoints in your funnel which may have been pivotal to getting the conversion. This can lead to skewed metrics and result in placing your budget in the wrong areas.
I’ve seen clients remove keywords and audiences from their funnel because their analytics service (using last-click attribution) said they weren’t converting – only for their sales revenue to fall off a cliff.
These keywords did a wonderful job at bringing prospects into their site but had a non-existent conversion rate.
These prospects still needed to be shown social proof and time to consider their options.
All that said, the last-click attribution model does have its place in marketing.
If your ads and campaigns are designed to attract people at the moment of purchase, or your business is primarily transactional with a sales cycle that does not involve a consideration phase, then it does make sense to adopt the last-click model.
Although, if it takes your customers several hours or days to come to a decision, then it’s important that your last click model is used in conjunction with other attribution models so that you can get a clearer understanding of your sales process.