• Web Design
  • Understanding Attribution Models for Social Media Marketing

    Feb 17, 2021 | Advertising, Business, Conversion, Marketing, Social Media | 0 comments

    How did your newest customers find your business? You could assume they saw your Facebook ad, or maybe a friend told them about you. Many new businesses overlook the importance of knowing exactly where new customers found their business. Did they see your ad across multiple platforms? Did they come across your Instagram posts on the explore page? Did they hear about you through word of mouth?

    Attribution modeling can answer all of these questions, depending on how long your sales cycles and customer journeys are. Some businesses only need to know the general details about where their customers came from, while others need to know exact and thorough details. There are attribution models for everyone that allow businesses to get the most out of their social metrics.

    What is attribution modeling?

    In marketing, attribution modeling is assigning a value to each interaction point a customer has with a brand that led to a purchase. Attribution models work backwards from the purchase: what did the customer see or do right before making the purchase? Where have they seen and/or interacted with your business? Where did they first see or hear about your business? 

    The importance of marketing attribution models

    Keeping track of your customer’s interactions with your business leading up to purchase provides valuable information about what aspects of your marketing strategy are working and what aspects are not. Attribution modeling in social media doesn’t just credit the social channel as a whole, it specifies which social channel your customer found your business on, which type of content they saw/interacted with (post, video, story post, user generated content, etc.), whether they purchased through the social platform or searched your website through a search engine like Google or Safari, and how long they interacted with your content or website before making a purchase. 

    The thought of collecting all of this data can be overwhelming. Luckily, there are tools out there that can make this process easier to understand and tackle.

    There are different types of attribution models. So before you jump in with both feet, take a minute to learn more about the different types of models, and which model will work best for you and your business.

    Types of marketing attribution models

    The three major types of marketing attribution models are first touch, last touch, and multi touch. Within the multi touch model are additional models that assign weight to each touch point.

    To better explain how each model works, let’s use the following scenario:

    Chase has been wanting a new hoodie. Nothing specific, just a casual everyday hoodie. On Thursday, Chase’s friend posts a photo on Instagram wearing an olive green hoodie with the hoodie company’s Instagram page tagged in the photo. Chase clicks on the company’s tag and scrolls through their page before exiting without making a purchase. On Friday, Chase begins receiving ads on Instagram promoting the hoodie company. He clicks on one of those ads, doesn’t make a purchase, but signs up for the company’s newsletter. On Sunday, Chase receives an email with a coupon from the hoodie company, which he decides to use. He is redirected from the email to the company’s website where he uses the coupon to purchase a new hoodie. 

    First touch attribution model

    This model credits the how the customer first heard about the company. In our example scenario, the first touch model would credit Chase’s friend’s Instagram post as the source of the purchase. This model is simple, but it ignores all other factors and actions that led to the purchase. 

    This model is best used for campaigns solely interested in increasing and expanding their brand awareness.

    Last touch attribution model

    As the name implies, the last touch attribution model does the opposite of the first touch. It credits the customer’s final touchpoint before making a purchase. In our example scenario, the last touch attribution model would credit the email Chase received as his source for purchasing. 

    Similar to the first touch model, the last touch model is simple, but ignores all interactions between the customer and the company that led to the customer’s decision to purchase. This model is best for determining where your customers are coming from, or measuring campaign effectiveness. For example, last touch attribution modeling can determine if a company’s Facebook ad successfully directed viewers to the company’s website. This model isn’t the best option when tracking the success of marketing campaigns prior to the last touchpoint. 

    Multi-touch attribution model

    These models are much more detailed, but much more difficult to track. There are different types of multi-touch attribution models: the linear multi-touch model, the U-shaped model, and the algorithmic model. 

    The linear multi-touch model gives equal weight to each touchpoint. Chase’s interaction with his friend’s Instagram post, the hoodie company’s ads, and the email Chase received, would all be given the same weight. This model is more accurate than the first touch and last touch models, but it assumes that each touchpoint is of equal value, which usually isn’t the case. 

    The U-shaped model gives 40% weight to the first touchpoint, 40% to the last touchpoint, and 20% to every touchpoint in between. This model recognizes that the first and last touchpoints are the most important, but still gives value to the rest of the touchpoints. In Chase’s case, the Instagram post and the email containing a coupon would get the most credit for his purchase. This model is ideal for lower price point products, where the customer doesn’t spend as much time considering buying the product. 

    The algorithmic model applies different weights to each touchpoint, and is customized using your company’s data and performance history. This model is the most accurate, but the most challenging to set up. In fact, you’ll likely require a data analyst’s help. The Instagram ad Chase received may have been the most impactful factor contributing to his decision to purchase the hoodie. The U-shapes model would account for this and adjust the weight distribution accordingly. 


    Attribution models aren’t just for enterprise businesses, they are incredibly beneficial for small businesses as well. Businesses of all sizes can benefit from setting up their own models to get a better understanding of how their marketing is performing. Although the multi-touch model is the most precise, the first and last tough models are most useful in social media marketing. With achievable and meaningful marketing goals in place, any of the attribution models would be incredibly beneficial to your business’s marketing efforts. 


    Submit a Comment

    Your email address will not be published. Required fields are marked *

    You may also like…