Skip to Content

Are you using the right attribution method?.

James Holding, Head of Analytics

The author

James Holding

Head of Analytics

It’s never been more important to ensure that your brand is using the right attribution method. Making the correct decision can completely transform your customer's experience, and make it easier to understand exactly where to focus your marketing spend.

Using a simple attribution method can skew the results of marketing performance, falsely displaying the impact of a campaign. On top of this, information concerning a customer's journey may remain hidden, making it increasingly difficult to pinpoint where your brand should focus marketing spend.

Attribution challenges

Today, there are three major challenges facing attribution. From measuring data in isolation to choosing the wrong method for your brand, you can view the common issues marketers face below.

Isolated attribution

So many businesses continue to measure a channel's performance in isolation, without considering the true effect on performance. This is a dangerous route to take, as you fail to look at the complete customer journey and make decisions that appear good in isolation, however harm the overall business.

Customers are likely to engage with your brand from multiple channels, whether they've been directed to your website from a display ad or first hear of your products from a TV ad. However long their journey, more than one channel will certainly play a role in driving a sale.

Choosing subjectively

Every choice you make will have an impact on the results you receive. From choosing an attribution model, selecting weightings or prioritising certain channels, your opinions will sway the output, providing an inaccurate view of a campaign.

Selecting the wrong method

Taking the time to assess and compare various attribution models, rather than choosing the simplest approach, is a must for any business. Every method has its pros and cons - the most important thing is picking the perfect approach for you.

Let's say your business uses the last-click model; marketing channels that are regularly last touchpoint are valued highly while others (that increase awareness and consideration) are undervalued.

Your results may benefit your brand in the short-term, but will not serve you well in the future for other campaigns. The last-click method offers a hazy view of the customer journey, highlighting channels that perform well at a single stage of the user journey and failing to acknowledge the rest.

Comparing attribution methods

One size certainly doesn't fit all when it comes to attribution methods. Think about your own goals, while reading about the most commonly used attribution methods below:

Single touch (first/last click)

Single touch attribution covers two of the simplest attribution models, where 100% of the revenue is credited to a single interaction (and therefore a single channel).

First click

First click attribution assigns 100% of the credit to the customer's first interaction.

Pros

  • First click helps marketers to invest spend at the start of a customer's journey
  • Easy to put into place, thanks to low level complexity and low-level investment

Cons

  • Focuses largely on brand awareness
  • Ignores efforts further down the marketing funnel
  • Prevents informed strategic decisions from being made further along the customer's journey

Last click

Last click assigns 100% of the revenue credit to the customer's final interaction with a brand.

Pros

  • Offers insight on the channels that convert the most leads
  • Easy and affordable to implement

Cons

  • Marketers only focus on the final touchpoint a customer makes, and fail to acknowledge the rest of the journey
  • Engagement further down the marketing funnel are disregarded, despite being important

Linear

A linear model considers all touchpoints of a customer their journey. A linear model may be an improvement when minimal research is required before a conversion, such as buying a card for a loved one. 

Pros

  • Offers more information about a customer's journey than single touch attribution
  • Assigns credit evenly across multiple channel

Cons

  • Remains a simple model, by assuming every interaction is weighted the same
  • However, it is still more complex - and more expensive to implement - than single touch methods
  • Increasing marketing activity will increase the value of a channel, resulting in larger investments for that channel. For example, more display ads will increase the value of that channel, resulting in more spend for display overall.

U-shaped

Every touchpoint in a customer's journey is credited, but not equally. The u-shaped model splits the journey up into three sections:

  1. Beginning (first)
  2. Middle
  3. End (last)

Different levels of priority can be given to the three sections by choosing different weightings (for example, a u-shaped model could be turned into a first or last click model with the correct choice of weightings).

A u-shaped model allows businesses to overlay knowledge about their sales cycle, rather than relying on a static model.

Pros

  • A much more sophisticated model, considering every touchpoint of a journey
  • Assigns credit to multiple touchpoints
  • Assigned credit varies rather than fixed

Cons

  • Subjective
  • The number of touchpoints in the overall customer journey could affect the value assigned
  • Complex, using assumed understanding to assign weightings

Channel weighted

As the name suggests, the channel weighted attribution model varies credit based on a marketing channel. Channel weighted models vary in complexity and can include custom rules in a similar way to the u-shaped model, bespoke for each channel.

Models like this are much more advanced than single touch attribution methods, however they are very subjective and can often provide skewed results based on existing bias. 

Pros

  • An advanced model, tracking every individual touchpoint
  • Assesses and applies weight to every touchpoint

Cons

  • Subjective
  • Doesn't include interactions that take place throughout a customer's entire lifetime

Cross-channel, data-driven approach

The above attribution models suffer from subjectivity, and the only way to measure the effectiveness of a marketing campaign accurately is with a data-driven model. This way, opinion bias in choosing a model does not have an effect whilst marketing activity designed to game the results, is no longer effective.

With the cross-channel, data-driven approach, every channel is valued as it should be. Accurate and clear, it's easier to see where marketing efforts are making the most impact.

It's best to use individual level, online customer data as this offers the chance for marketers to do further analysis and tailor customer communications and channel preferences.

Cross-channel, data-driven attribution is best for products with a long buying cycle, or when multiple purchases are likely.

Pros

  • Non-subjective
  • Weight given to every touchpoint in a customer's journey
  • Quantifies the impact of each channel
  • Combines data sources, including click and impression data

Cons

  • The most complex attribution model to implement, but not to use

Establishing best practice

When you choose a sophisticated attribution model, you'll gain more insight on your customers' journeys for future campaigns - putting you, as a marketer, back in control and removing any guess work.

To get the most out of your attribution model, you should:

  • Begin with a large data platform, ensure this has complete and high-quality data
  • Use individual level data; don't forget to look at both online and offline interactions of individual customers to make sure your data platform can offer a single customer viewpoint (tools, such as Almanac by Jaywing, are great for this)
  • Pick the model that is right for your business and its goals
  • Always focus on the outcome for the customer; if you want to improve a customer's experience, aim for a sophisticated attribution method
  • Make sure you are considering the full impact on all channels, and all areas of your business
  • Revisit/review your model choice (every quarter)