What is multi-touch attribution? (Plus how to implement it)

By Indeed Editorial Team

Published 20 September 2022

The Indeed Editorial Team comprises a diverse and talented team of writers, researchers and subject matter experts equipped with Indeed's data and insights to deliver useful tips to help guide your career journey.

A buyer's journey encounters many touchpoints before becoming a paying customer. This journey offers valuable data for marketing teams, allowing them to promote new ways of attracting more leads and prospects. Learning about multi-touch, or multi-channel, attribution can allow you to help your employer personalise consumer experiences and optimise campaigns. In this article, we explain what multi-touch attribution is, outline how to use it and list the different multi-channel attribution types.

What is multi-touch attribution?

Multi-touch attribution is a marketing process that involves determining the value of each customer touchpoint that leads to a conversion. It evaluates all web pages a customer visits before purchasing from the company and assigns fractional credit to each one. This allows marketing teams to determine which channels have the most influence on a sale and adjust their campaigns accordingly. Multi-channel attribution works for multiple conversion events, such as purchases, free trial sign-ups and subscriptions. By figuring out which channels or campaigns are responsible for these conversions, companies can better allocate funds to acquire new customers.

Multi-channel attribution provides deeper insight when compared to traditional attribution approaches, such as first- and last-touch attribution. These models assign credit to only the first or last marketing touchpoint before a consumer converts into a customer. This means they fail to measure the contribution of other touchpoints throughout the buyer's journey. Multi-channel attribution aims to eliminate these biases by allocating credit algorithmically to every touchpoint.

Related: A guide to branding vs. marketing (plus differences)

How to implement multi-channel attribution

Here's how to implement multi-channel attribution:

1. Collect data on the customers

Start by researching who's visiting the company website, how they got there and whether they convert into a customer. Also, determine what pages the customers view and what actions they take upon arriving on these landing pages. Then, consider what characteristics and behaviours the customers have in common based on these insights. This information helps you better understand a business's customer base and ensures you develop campaigns that align with their values and traits.

Related: How to map a customer journey (with definition and example)

2. Determine the appropriate model and KPIs

Next, determine which attribution model is most appropriate to use based on the company's goals. Some of the factors to consider when determining which model to use include the types of campaigns and the length of the company's sales cycle. Once you've selected a model, select your key performance indicators (KPIs). These are quantifiable metrics that allow the company to gauge their long-term performance and success in achieving specific objectives. For example, most companies use return on investment (ROI) and conversion rates for their KPIs.

Related: What KPI stands for and how to use it in your career

3. Foster team collaboration

The next step involves gathering all relevant team members who participate in the process, including marketing teams, analysts, analytics managers and other key stakeholders. This group of people is responsible for collecting and analysing data from all channels, so good communication between them is essential. The marketing teams optimise touchpoints based on the insights gained from analysts and researchers. Also, including stakeholders in this process helps them understand company needs and set budgets to cover these requirements.

Related: Teamwork skills: definition, types and tips for improvement

4. Deploy marketing analytics software

Next, deploy marketing analytics software with multi-channel attribution modelling capabilities. These systems allow you to collect all attribution data within a centralised platform, allowing for better collaboration and analysis. They help sort and collate data into clear, digestible metrics from which you can derive insights, including the motivations behind conversions.

5. Discover insights and implement improvements

After using marketing analytics software, mine the data to discover key insights. This involves determining how effective the various marketing channels are at generating prospects and helping to change these into conversions. Additionally, try to ascertain whether any touchpoints deter conversions or make it unnecessarily challenging to take place. Then, gather insights from all the analysts to consider whether any channels work together in some way to facilitate conversions. Once you've made these insights, decide on what to do to improve the individual web pages and enhance the user experience.

6. Carry out frequent monitoring and testing

The final step involves continuously evaluating multi-channel attribution data to track, test and optimise campaigns. Some companies use the A/B testing method to optimise touchpoints. This involves creating two or more versions of a website and randomly allocating consumers to one or the other. Companies then compare both sites to determine which one performs better and leads to more conversions.

Related: How to improve online customer service (with steps)

Multi-channel attribution types

Here's a list of the different multi-channel attribution types:

Linear

The linear attribution model values every touchpoint evenly by giving all the touchpoints across the buyer's journey the same amount of credit for driving conversions. With this model, all the touchpoints are equally crucial to the buyer's journey. For instance, consider that a consumer views an advertisement via email and social media. According to this model, both touchpoints earn 50% of attribution credit. If the buyer's journey includes two or more touchpoints, such as a native advertisement and a television ad, the model now attributes 25% credit to all the touchpoints.

Related: A detailed guide to customer success vs. customer service

Time decay

Time decay attribution is similar to linear attribution, but it decreases each touchpoint's value over time. It assigns more value to the touchpoints the consumer interacts with that are closer to becoming conversions. This means it primarily focuses on touchpoints towards the end of the sales funnel, rather than those in the beginning and middle.

For example, consider that a customer views a sponsored social media post about a company's product in the morning. They click on the sponsored post to access the company website but then wait a few hours until the afternoon before actually purchasing the item. In this instance, the time decay model assigns more credit to the website over the social media post because the time between the customer's interaction with the website and purchasing is less.

Lead creation touch

Lead creation touch attribution assigns all the conversion credits to the touchpoint that sees prospects become leads. A prospect is someone who visits a company website but doesn't purchase anything. Prospects become leads once they provide the company with details about themselves, such as by subscribing to a newsletter, or if they've purchased something before.

Custom multi-touch

A custom multi-touch model refers to those created by an organisation to compare the success of different touchpoints according to their own criteria. Custom models can be complex to create and require advanced marketing attribution software, but they allow companies to credit touchpoints according to what they value most. For example, a company may value the detailed nature of the full-path attribution model but may want to improve how it assigns little value to certain touchpoints. Due to this, it creates a custom multi-touch model that combines these elements.

Related: What are customer journey touchpoints? (Including 16 examples)

U-shaped

This is a position-based model that assigns credit to touchpoints based on their position in the buyer's journey. It usually attributes 40% to the first touchpoint and the lead conversion touchpoint while dividing the remaining 20% among the additional touchpoints. U-shaped attribution values the touchpoints that start the buyer's journey and those that lead to conversions.

For example, consider that a customer reviews a television advert, a social media ad and then a native ad on the company website. Then, the customer purchases the item after reviewing the company website. According to this model, both the television ad and company website receive 40% each, while the social media ad receives 20% credit.

W-shaped

W-shaped attribution is a position-based model that prioritises the first, third and fifth touchpoints in a buyer's journey and assigns the rest of the touchpoints equally low values. The reason for this is that the first, third and fifth touchpoints denote the start of the journey, lead creation and opportunity creation touchpoints. It usually assigns each of these touchpoints 30% and divides the remaining 10% among the other touchpoints.

For example, consider that a consumer's first touchpoint for a clothing item is a TV advertisement. Then, they come across several social media ads throughout the day, which prompts them to browse the website. On the website, the consumer finds the desired item and signs up for the company newsletter without buying the product. This represents the lead creation touchpoint. The consumer then receives a promotional email about the product, motivating them to buy the item. This email is the opportunity creation touchpoint. The TV advertisement, website and email all receive 30%, while the social media ads receive 10%.

Z-shaped

Z-shaped attribution is another position-based model that values the first, third, fifth and last touchpoint on the buyer's journey. This leaves the second, fourth and second-to-last points with equally lesser values. Marketers call this the Z-shaped model because, when viewing it linearly, it resembles a Z shape.

Explore more articles