Nowadays, the customer buying journey has become more complex. As a result, businesses are turning to multi-touch attribution to make sense of it all.
For instance, let’s say a customer first hears an ad for your company while listening to a podcast on the subway. Then, they see several videos of your influencer affiliates talking about your brand on TikTok. A few days later, they see your Google PPC ad. They click and sign up for your email list. You send them a series of welcome emails and a 10% off code on their first purchase. Finally, they decide to make a purchase and use the promo code.
Now, which channel is actually responsible for attracting the customer and converting them to a sale? Your answer might depend on which attribution model you use. Last-touch would say it was the email. First-touch would say the podcast, but that’s not the full story, is it? It may have contributed to brand awareness, but did it impact the eventual conversion?
A conversion event rarely happens based on one interaction. Today, a customer may interact with your brand seven times or more before making a purchase. This is where multi-touch attribution comes into play.
Multi-touch attribution helps tell the entire customer journey from the first interaction to the last and everything in between. In this guide, we’ll cover multi-touch attribution models, and how to set up multi-touch attribution for your business.
What Is Multi-Touch Attribution?
Marketers use attribution to identify which channels and touchpoints are responsible for conversions. Multi-touch attribution (MTA) is the process of understanding the success of your marketing by analyzing every touchpoint that leads to a conversion.
Single-Touch vs. Multi-Touch
Multi-touch provides a complete picture of each campaign you are running and assigns a percentage of credit to all the touchpoints in a buyer’s journey. It can help you optimize your marketing spend allocation by identifying the strategies that are most effective.
Previously, marketers had to depend on single-touch attribution, which gives credit to only one channel for conversions. For example, if a person reads a blog post and converts, you’ll credit that piece of content for the conversion while ignoring other channels that may have influenced them.
Conversions and sales usually happen after a mix of marketing touchpoints like posts on social media, blog articles, webinars, Google Ads, content on your site, in-person events, email marketing, and so on. So, single-touch has major drawbacks when considering that customers today interact with you online and offline many times before converting.
Last-Touch vs. Multi-Touch Attribution
Last-touch is one example of a single-touch attribution model. It gives 100% credit to the last touchpoint of a consumer journey. It’s the easiest for analytics platforms like Google Analytics to track because it is where the conversion happens. However, it doesn’t paint the whole picture.
It may give one channel more credit than it should get, and lead to incorrect optimization decisions.
Why Do You Need Multi-Touch Attribution?
There are several reasons that multi-touch attribution is ideal when assessing your marketing efforts. However, there are two primary reasons it is becoming more popular right now—third-party cookie deprecation and rising advertising costs.
In their early days, the CPA on Facebook and Instagram was a steal. Now, over 10M marketers in the US use Facebook for ads. As more people are bidding on digital ads, due to increased competition the cost has increased over the years. Even more so after the pandemic, people’s media consumption patterns shifted. They were commuting less and browsing online more. Retailers flocked to meet their customers where they were—online. As a result, digital marketing became necessary.
Secondly, Apple’s iOS 14 update has changed the digital marketing landscape. Apple users can now opt out of ad targeting. For Facebook, this made it harder for marketers to track conversion events for Apple users, specifically using tracking pixels. Facebook users that opt out still get ads, but they aren’t targeted. So, if you visited the Nike website, they wouldn’t be retargeting you with Nike ads on Facebook after.
On top of that, third-party cookie tracking is being phased out. Third-party cookies enable a third party that is not the website you visited to track what you do on the Internet, like what sites you visit. Advertisers would use these to identify which ads to serve a customer, but third-party tracking is going away with privacy regulations.
All of these factors together are pushing marketers to get smarter about their attribution. Digital advertising is still effective in reaching customers, but tracking what works and what doesn’t is harder.
That’s why multi-touch attribution is critical. It gives you a clear picture of how your touchpoints and channels influence every conversion.
Let’s take a look at the different types of multi-attribution models, and which to use for your business.
Multi-Touch Attribution Models
There are various multi-touch attribution models. The main difference between these models is how many credits they assign to every point of interaction in the customer’s journey.
- Linear
- Time Decay
- U-Shaped
- W-Shaped
As a marketer, you can either adopt one of these models as it is or use them to develop your custom models.
Linear Attribution Model
The linear attribution model gives equal credit to every touchpoint involved in the customer’s journey.
However, the linear model doesn’t recognize whether a touchpoint was more influential than another. For example, in this model, a first paid touchpoint is as valuable as a last organic search touchpoint. But maybe a consumer would not have known about your brand without the first touchpoint.
Linear attribution captures the entire customer journey, but it is considered the simplest form of MTA. You know what touchpoints your customer interacted with, but you don’t know to what extent each touchpoint influenced their decision. If you are just starting to use MTA, and don’t have enough data to set a baseline of how the touchpoints in your business perform among your audiences, this model may be ideal.
Time Decay
Time decay assigns more credits to the touchpoints closest to the conversion. In other words, the closer the conversion event is, the more effect it has in conversion, so they get more points. This model prioritizes touchpoints that drive conversions over those at the top of the funnel.
The time decay attribution model gives the first touchpoints fewer credits. Unfortunately, the time decay model fails to analyze the nature of each touchpoint. For example, let’s say you run a YouTube influencer video campaign and a banner ad. If the banner appeared after the video, based on the time decay model, it would receive more credit.
However, most of the buyers are loyal subscribers of your YouTube partner. Some may even use a UTM link or discount code provided by the influencer. In that case, should the banner ad get more credit?
Giving more credit to the banner ad simply because it appears after the video doesn’t always make sense. SaaS companies often use the time decay attribution model because they tend to have longer sales cycles. You can look at the time in between interactions to see how they influenced each other, and it is less likely to skew credit.
That said, the drawback of time decay attribution is that it discounts how much top-of-funnel marketing efforts contribute to conversions.
U-Shaped or Position-Based
The U-shaped attribution model gives extra credit to both the last and first touchpoints. It allocates 40% for each and shares the remaining 20% across the intermediary touchpoints. In other words, when mapped on a chart, it forms the shape of a “U”.
The U-shaped model is known as a position-based model. It focuses on the bottom and top-funnel touchpoints and less on mid-funnel.
It helps marketers understand which strategies are great at brand awareness starting the customer journey and which are better at converting. Also, it acknowledges the intermediate touchpoints but gives them less credit. The U-shaped is the first model that recognizes that all touchpoints don’t deliver the same value.
This model is suitable for short sales cycles. Also, it’s ideal for marketers who want to focus on the impact of the first and last touchpoints. It emphasizes less on the intermediary touchpoints that support the customer’s journey.
W-Shaped Attribution
The W-shaped model takes the U-Shaped further. It credits the first touchpoint as well as the lead creation and opportunity development touchpoints. In this model, every primary touchpoint earns a credit of 30%. And the rest of the 10% gets distributed among the intermediary touchpoints.
For example, let’s say a person is searching for a travel destination for an upcoming holiday. They click on several Pinterest ads, PPC ads, and read travel blogs for different locations. Then, they click on a sponsored content link in one of those blogs for a particular hotel chain. They loved the idea of going for a holiday in Miami, so they searched for the hotel’s website. They go ahead and register with the site and include the destination in their wishlist, which is called lead creation. However, they do not book the holiday.
After some days, they received an email offering a 10% discount for a stay in Miami. But there is a catch, a user has to download and install their hotel app. They open the link sent, which takes them to the App Store, download the app and complete the purchase. The email offer sparked them to take action. In a W-shaped model, it would get a higher percentage of credit than one of the PPC ads they visited.
This model is good since it considers identifiable stages. Its drawback is that it’s not always easy to differentiate between the stages. Therefore, it makes the process complex.
Custom
In some cases, you might find that none of these models work for your specific business and marketing needs. With multi-touch attribution, you can also create a custom model. You decide which channels to track and how much credit to give each. Creating a custom attribution model is a lot of work, so if you are new to attribution, it is better to choose one of the above models to implement first.
How To Do Multi-Touch Attribution
Setting up multi-touch attribution is not an easy thing. The customer’s journey is more complex day after day. For instance, according to Google, 90% of consumers switch between devices to complete tasks.
When it comes to multi-location businesses and franchises with physical and digital presences, it can get even more complicated. A customer might see a billboard for a new store opening in their neighborhood while out walking. Then, they watch an Instagram post about it and a TV ad that directs them to download the store’s app. While nearby the store location, they get a push notification on their phone with a discount and make a visit. How do you account for all those touchpoints when looking at your marketing strategy?
Making sure that all these touchpoints across multiple channels, devices, and platforms communicate with each other is a challenge in implementing multi-touch attribution. It is one of the main reasons why businesses seek out Goodway’s services.
However, you can get started with these beginner steps.
1. Set up tracking
First, you need to have proper tracking set up. This way, you can collect the right types of data to use in your marketing attribution model. You can do these in three ways.
JavaScript tracking code
This is a snippet of a code usually added to a company’s website. When a person visits your website, the code triggers and tracks their actions from one page to the other. JavaScript code helps you know who is interacting with your website and how. It includes calls such as page, track, identity, and inbound. In other words, you are enabling first-party tracking.
Setup UTMs
You determine which snippets populate at the end of URLs. UTMs provide data regarding where the consumer comes from, including information such as the source (e.g. Instagram), the campaign, and the creative type. You can integrate UTMs with JavaScript calls for a more precise and complete picture of your users.
APIs
Connect the tool you use and data to give a complete picture. Integrate it with CRMs, HubSpot, Salesforce, Google, email marketing, and so on.
2. Apply insights
Once you have all this data, what do you do with it? Ideally, you’ll have one dashboard set up to bring all your data together in one place. You can pull insights from all your channels and highlight what is working.
After some time, you’ll feel like you completely understand how your campaign has performed based on the data collected through the attribution model. That’s the right time to apply those insights in real-time to start course correcting.
3. Optimize and Test
This shouldn’t be a one-time exercise. Instead, marketers should be testing their MTA data more often to optimize and evaluate campaigns. Testing regularly will help marketers know the best tactics and marketing sequence to reach their customers at the perfect time.
If implemented well, multi-touch attribution is a robust process that businesses can use to know various factors that make a consumer convert. Also, it helps marketers understand which of those factors influenced their decisions. Therefore, marketers can use this information to enhance their buyer’s journey and marketing effectiveness.
Putting all these tools together takes work and technical knowledge. In the end, it may not be a complete solution for you. This is why many brands prefer to partner with a marketing agency like Goodway Group to manage attribution. Agencies tend to have the experience, resources, and tools needed to manage online and offline attribution across channels. Get in touch with us today!