Do you know how your newest customers heard about your product or service? Maybe you have a general idea, like they saw an ad and clicked on it, or that someone told them about it. But what if they saw an ad multiple times across different platforms? Or it was a combination of newsletter information, social media posts, and word of mouth?
Depending on how long your sales cycles and customer journeys are, attribution modeling can be the solution to all of these questions. Some companies only need a general idea of where their new customers come from while others need detailed results. There’s an attribution model for everyone that can help you get more out of your social metrics.
What is attribution modeling?
Attribution modeling in marketing is assigning a value for the customer’s interaction points with the brand that led to the purchase. Models give you an idea of where your customer came from. Attribution models work backward from the purchase: how did the customer first hear about you, which interactions were important and what did they see or do right before the purchase. They can be as easy as checking referral sources of a web purchase or as difficult as needing a data analyst to do it all for you.
Why marketing attribution models are important
Closely tied to ROI, marketing attribution helps you understand if your social strategy is working, what stages in the journey are most or least important and where you should be focusing your marketing efforts. ROI measures revenue while attribution measures impact.
In terms of the social media marketing funnel, attribution models credit the channels within the Awareness and Consideration stages that led to the Conversion.
Attribution modeling in social media doesn’t give credit to just the social channel as a whole. It breaks down which social media platform it was, what type of content like a post, video or white paper and how often the customer had to interact before the conversion. Models also help you understand if your social media campaign is actually working. Thankfully, there are tools out there that can make this all easier to understand.
Before you dive in, learn more about the different models to know which one works best for you.
The different marketing attribution models
There are three major types of marketing attribution models: first touch, last touch and multi touch. Within the multi-touch model are additional types that assign weights to the various touchpoints.
In the below models, we’ll use the following scenario: Nina has been thinking of purchasing new shoes. On Monday, her friend posts on Facebook about some new shoes that they bought with a link to the shoes’ product page. Nina visits the manufacturer’s website and then on Tuesday, begins to receive Instagram ads that show her the company’s shoes. She clicks on one of those ads, doesn’t purchase yet but signs up for the newsletter. On Thursday, she receives an email with a coupon from the company, which she clicks through and completes her purchase.
First touch attribution model
The first touch attribution model is pretty self-descriptive: it credits how the customer first heard about you. In our example scenario, the first touch model would credit her friend’s Facebook post as the source of the purchase. This is an easy model but also ignores all the other interactions that factored into Nina’s purchase.
It’s best used for campaigns and needs that are looking to focus on the top of the marketing funnel.
Last touch attribution model
The last touch attribution model does the opposite of the first touch. It credits the very last interaction the customer had before making the final purchase. In our example scenario, the last touch attribution model would credit the email newsletter as Nina’s source for purchasing.
Like the first touch model, it’s a simple one to track, but also ignores the previous interactions that Nina had. Last touch is often the default model and is great if you’re looking to analyze where your traffic is coming from, or if it’s the metric to measure campaign success. For example, it can reveal if one Twitter post’s link directed traffic to a landing page. It’s not the best option if you’re trying to measure the impact of your other marketing efforts prior to the last touch.
Multi-touch attribution model
Multi-touch models are more detailed, but also more challenging to track. To measure multi-touch attribution, you need to have the tools in place to track and analyze each touchpoint.
The linear multi-touch model gives equal weight to every interaction point. Nina’s interactions with the product: her friend’s Facebook post, web visit, Instagram ad and newsletter send would all have the same credit weight. While this is an improvement on first and last touch, it also assumes that every touchpoint is equal to the next when in reality, this probably isn’t the case.
The U-shaped model gives 40% to the first touch, 40% to the last touch and 20% to everything in between. This weight modeling recognizes that the first and last touch are most important to the customer but everything else still gets counted in, too. In Nina’s case, the Facebook post and the newsletter would get the most credit for her purchase. This model works well for short sales cycles and lower price point products, where the customer does not typically spend as much time considering the purchase.
The algorithmic model gives different weights for different touchpoints customized based on your own data and performance. While the most accurate, it’s also the most difficult to set up and might not be possible without the help of a data analyst. In our example scenario, Nina’s friend’s recommendation and the Instagram ads may have had more of an impact on her decision than the email newsletter. The model would adjust for this and you would be able to easily recognize which parts of the journey are most important.
Setting up a social media attribution model
If you’re just starting out on creating attribution models for your social media marketing,, then it’s best to take it one step at a time instead of diving straight into a complex multi-touch model.
1. Decide on your social media goals
What’s the point of modeling if you don’t have any goals to reach? Learn how to set and achieve your social media goals and determine the social media metrics that match up with them. With measurable goals, the attribution models will be able to tell you if you’ve achieved them. Without them, the models are just another set of data.
2. Set up UTM tagging
Urchin Tracking Modules (UTM) allow you to add tags at the end of a URL so you can track where the link click came from, what type of traffic it is and if it’s associated to a certain campaign. UTM tracking in social media is important because it gives you more information on which networks to credit instead of just the channel “social media.”
Sprout’s Premium Analytics offer the ability to edit and add tracking parameters right from the Compose window.
URL tracking in Sprout also allows for automation in rules. They’re easy to set up and you don’t need to leave Sprout to manage your campaign.
Other options, like Google Analytics and the Facebook Attribution Tool are also available to social media marketers. Note that Facebook’s tool requires the installation of their Pixel, which will also help you in your overall Facebook advertising strategy.
3. Use link shortening or vanity URLs
A link shortening tool like Bitly, takes a long link and shortens it to a unique one. With a paid plan, you can add a branded one so every shortlink that’s generated has your company name attached to it. Another major plus: Sprout integrates with Bitly so all of your links can be automatically shortened.
4. Use surveys
If you want to get started testing the attribution model waters out, use a survey. For every lead generation, include a question that asks how they heard about you. This will give you at least some idea of which marketing effort is working and what type of attribution model may eventually serve you best in tracking performance.
Marketing attribution models aren’t just for enterprise businesses. Small businesses can easily set up their own models to help analyze how their marketing is performing. While the multi-touch model is the most accurate, the first and last touch models also have their uses in social media marketing. If you want to dive further into attribution modeling, there are plenty of software options out there for you that focus on only that.
But for social media marketers that are just dipping their toes in, a management platform like Sprout that helps you set up tracking models, will be the best way to go. Sign up for a free demo today to see how Sprout can help you with your attribution models.
Reach vs. impressions: What’s the difference in terms?Published on February 20, 2024 Reading time 6 minutes
10 metrics to track influencer marketing success in 2024Published on January 24, 2024 Reading time 8 minutes