Attribution Modeling

Marketers spend the majority of their time optimizing communications programs for success, fine-tuning budgets and maximizing revenue contributions based on the ever-changing, and often competing, list of promotion channels.

As your audience moves through the different stages of the buyer lifecycle, they are exposed to a mix of these channels and interact across multiple devices.

Because of this, it can be difficult to determine which channels have been the most pivotal in affecting sales.

The big question becomes: How do we measure success?

In a multi-channel, multi-touchpoint world, what’s the most efficient way to properly credit conversions to channels so that brands know how to allocate budget, increase projected revenue and maximize long-term profitability?

That’s where Attribution Modeling comes in.

Attribution Modeling is the process of answering the following question for marketers and executives: Which channel fetches the most leads and, subsequently, revenue?

It can be difficult to measure with full precision the effectiveness of a brand’s conversion and channel attribution process; however, here are five ways to do it with a reasonable amount of accuracy.

1. First Touch Attribution: Which channel brought a buyer to you for the first time?

In this methodology, the first channel your audience provides trackable information at takes all the credit. The argument for this is: If a channel can cause a lead to share information, that channel has the most influence on the customer.

The process for first touch attribution is: Define channels as A, B, and C in your campaign, and any success you achieve within a buyer’s lifetime is attributed to the first channel the buyer interacts with. If Buyer X comes through Channel B first, then all sales for Buyer X are attributed to Channel B.

2. Last Touch Attribution: Which channel turned a lead into a buyer?

The last touch method revolves around the final-straw argument. The channel that gets credit is the one that caused the customer to “pull the trigger” and make a purchase. It’s a culmination of the entire customer experience at the time closest to sale.

The process for last touch attribution is: Define channels as A, B, and C and attribute conversion to the last channel the buyer interacts with. If Buyer X comes through Channel C just before the sale, then all sales for Buyer X are attributed to Channel C.

3. Linear Attribution: Which channels did the buyer interact with before point of sale?

In this methodology, all channels the buyer interacts with receive equal credit for each sale. Linear attribution embraces the “formative buying process through experiences” logic. Because every customer interaction with a channel builds upon the last interaction, linear attribution considers these interactions too complicated and connected to define within a hierarchy.

The process for linear attribution is: Identify all the channels the customer interacted with until point of sale and attribute the credit equally to all channels. If Buyer X comes through Channels A, B, and C, then all sales for Buyer X are attributed equally to Channels A, B, and C.

4. Weighted Attribution: Which channel leads to more valuable interactions?

The weighted attribution model believes that the interaction a customer has with certain channels creates a more meaningful experience and compelling impression on the final sale than interactions with other channels. The channels that lead to more valuable interactions command more credit. The weighting process is ”business/product dependent,” requires some experience and expertise to give meaningful weights, and must remain in effect for some time.

The process for weighted attribution is: Assign weighted attribution to different channels. For each customer sale, multiply the sale by channel weight to determine each channel’s credit. If Buyer X comes through Channels A, B, and C and you weighted Channel A as 40%, Channel B as 10%, and Channel C as 5%, then all sales for Buyer X are calculated according to those percentages.

5. Time Decay Attribution: Which channels did the buyer most recently interact with?

The time decay model gives greater weight to the channels with more recent customer interaction. The weighting process is “customer dependent,” requires some experience and expertise to give meaningful weights to each channel’s age, and must remain in effect for some time. Audience interaction with a channel closer to point of sale makes it a more valuable channel and that channel’s age gets a higher percentage of weight.

The process for time decay attribution is: Give weighted attribution to different channel ages, e.g. last touch channel gets 40%, second to last touch gets 10%, third to last touch gets 5%, etc. For each customer sale, multiply the sale by each channel’s age weight. If Buyer X comes through Channel A last, Channel C second to last, and Channel B third to last, then all sales for Buyer X are attributed as 40% to Channel A, 5% to Channel B, and 10% to Channel C.

This list of attribution models is not comprehensive and due to the breakneck speed of innovation in the field, it will only continue to evolve. It’s a marketer’s job to harness their brand’s image and “pull” prospects through the buyer lifecycle to the purchase stage and beyond.

No set of channels can rule all of the time, and marketing and communications professionals need to continuously monitor, track and shuffle the mix depending on what their buyers find useful.

Learn more about using an intelligent mix of communications channels with our latest white paper Maximize the Reach of Your Message with a Strategic, Multichannel Plan.