How did you hear about us? Trying to answer that question is the earliest practice of marketing attribution.
Answering it used to be easy. "From a friend," "saw your ad on TV," or "got your brochure." Now it's difficult even if you could ask every customer.
"Well, I saw you mentioned in a LinkedIn post, then I visited your website and signed up for your email newsletter. That led me to a blog post a few weeks later, and then I think I clicked a referral link and bought something. But it might also have been in your mobile app. I'm not sure..."
Today, a typical customer journey unfolds across many touchpoints.
Welcome to the wonderful world of the omnichannel customer experience. While consumer convenience has increased in lockstep with the touchpoints available to them, so have marketers' headaches in trying to figure out which of those channels drive sales and which don't.
What is marketing attribution?
Marketing attribution answers the question: which channels led someone to become our customer? With this knowledge, marketers can increase spending on the touchpoints that drive sales (and stop wasting cash on those that don't). Such optimization results in a higher marketing ROI per dollar spent.
Why attribution is so important for marketing teams
Along with increasing the ROI of your marketing spend, attribution affects several other areas as well, like:
Targeting and personalization: You can zoom in on your most effective touchpoints for more granular insights on customer behavior, personas, and preferences–and adjust your campaigns accordingly.
Product development: Attribution data can reveal customer preferences and insights that are helpful for product teams. You also know where to best reach your customers so you can increase user research participation and quality.
Brand and creative: You get more accurate data on the effectiveness of your messaging and creatives by testing them on your customers' preferred channels. That way, their metrics are not affected by an inherent customer bias against the marketing channel you're running them on.
Seven marketing attribution models
A marketing attribution model calculates which touchpoints in the customer journey get credit for a sale (and how impactful they were). There are seven types of attribution, which we can divide into two categories:
Single-touch attribution: One channel gets all the credit for a sale.
Multi-touch attribution models: Several channels get credit for a sale.
The first two models below belong to the single-touch category, while all others are multi-touch attribution. You can click each model's name for more details on their workings, pros, and cons:
First-touch attribution: The first touchpoint of the customer journey gets all the credit. (Also known as first-click attribution.)
Last-touch attribution: The last touchpoint takes the honor. (Also known as last-click attribution.)
Linear attribution: You spread recognition for the sale across all touchpoints evenly.
Time-decay attribution: All touchpoints get credit, but the closer one is to the sale, the more it gets.
U-shaped attribution: The first and last touchpoints get most of the recognition.
W-shaped attribution: The first, middle, and last touchpoints get most of the honor.
Custom attribution: Do-It-Yourself marketing attribution modeling and create your own calculation.
Imagine you're accepting an Oscar, and you thank the people who helped you throughout your life. If you used U-shaped attribution, then you'd give most thanks to your parents and the Director of your Oscar-winning movie.
How to choose an attribution model
There are several ways to choose the right attribution model for your business. Here are some ways to find the right one.
Use incrementality analysis
Suppose you want to understand precisely how each touchpoint and marketing activity contributes to a sale. In that case, you can use incrementality analysis, in which you expose two groups of potential customers to two variations of the buyer journey:
One version with the touchpoint you want to test.
One version excluding that touchpoint.
These groups typically need at least thousands of users for meaningful results. You'd also need to run these tests with variations for every channel in your customer journey.
If you can pull this off, you'll know exactly how much each channel contributes to a sale and can pick the right attribution model.
In their Marketing Strategy course, Reforge suggests such marketing efforts make sense if you:
Work with a large marketing budget containing online and offline channels.
Leverage view-based and click-based channels.
Have a product with a long sales cycle or a complex value proposition.
Data-driven with a little bit of intuition
In other cases, mixing some marketing data with your instincts can work. Say you hear from most customers that they first heard about your brand through top-of-the-funnel channels like organic search, social media posts, or a Facebook ad.
You also see from your data that there's not much engagement for a few days, but then people often make their first purchase after you start an email marketing campaign. You could then choose the U-model, where you give most credit to the first and last touchpoint.
Choosing the best attribution tool
The days when you could figure out attribution by asking people, "How did you hear about us?" are long behind us. Even spreadsheets aren't up for the job anymore. In an omnichannel world, you need marketing attribution software.
Such tools take care of the tracking, modeling, and analysis of customer data across all the different marketing touchpoints and platforms involved in your customer journey.
There are many such tools available. Here's what you should pay attention to when picking one:
How many integrations the attribution tool has. Generally, the more integrations, the better. Ensure the software supports more integrations than you currently need to accommodate future growth.
How many devices the tool covers. Some solutions specialize in certain devices, like mobile, while others cover a broad spectrum. Depending on your business needs, you might prioritize specialized attribution tools over more general ones.
How secure the platform is. Your attribution tool houses your end users' data, so pick one with best-in-class security features.
How they handle over-attribution and fraud. Make sure you choose a solution that detects fraud and blocks fraudulent attribution.
A visual overview of attribution tools we evaluated and their characteristics.
Despite the availability of so many attribution solutions, businesses typically run into these challenges with these tools:
They need to use multiple attribution tools for different channels.
Many tools can’t sync attribution data with their offline customer data.
Individual attribution tools often can't give a complete view of the customer journey.
You can solve these challenges with Twilio Segment. Our customer data platform collects and consolidates data from the various channels and platforms within your tech stack.
Streamline cross-channel marketing attribution with one tool
Twilio Segment simplifies marketing attribution by taking care of collecting and combining customer data from all your different channels and tools. You can then:
Effortlessly integrate attribution tools with your stack:Try out and combine different attribution solutions without requiring data migrations and other heroic engineering feats.
Use identity resolution to get a single customer view across all channels: Segment's built-in identity resolution determines which devices and data points belong to which users, saving you many hours of detective work.
Create a customized data pipeline: With Segment's Connections feature, you control which data from which tools you feed into your attribution models and what information you leave out.
To see these benefits in action, take StackCommerce as an example. Their account managers used to update attribution reports for their partners manually. After implementing Segment, this data automatically feeds into their partner dashboards, saving them hundreds of hours of work.
Read the full case study: How StackCommerce Made Paid Ad Attribution More Accurate With Twilio Segment.
Frequently asked questions
Segment + Wilco: The Quest for better learning options
Introducing Wilco - a platform for hands-on tech learning with real-world simulations. It encourages readers to try Wilco for free.
Extensibility at Segment: Use Mobile Plugins to enable Location-aware Marketing
Follow OwlEats, a food delivery startup, that uses Segment's mobile plugins to seamlessly integrate location data, enhancing customer targeting and incentivizing in-store pickups.
3 things we learned at CDP Week 2023: AI Edition
At CDP Week 2023, we looked at how AI trends are shaping customer experiences. Here's the recap.