Attribution Models—Tips for Deciding Which Is Right For You
Does your business have attribution under control? Or, is your marketing ROI in a “leaky bucket,” full of holes that show a lack of understanding about your customer behavior?
Having a solid attribution model ensures the money invested in marketing is being spent wisely and brands aren’t hemorrhaging budget where the ROI is least effective.
For example, if your company spends $500 on paid search and acquires 30 leads, this means your cost per acquisition (CPA) is approx. $17 per user. Contrast this with a $2,000 spend on a direct mail campaign resulting in only 10 leads generated and a $200 CPA, and it’s clear paid search is more efficient and generates a better ROI.
But do you really know the complete picture of what happened? For instance, did any of the direct mail recipients then became organic search conversions? Only having half the information means only being able to fix half of the problem.
Attribution is crucial. Knowing where your leads and sales come from makes it possible to pinpoint when, where and how your audience is most likely to convert, whether it’s through pay per click advertising (PPC), email, social media or another channel.
What Attribution Model is Right For You?
Here are six different types of attribution models currently used by marketers and we also list the pros and cons of each model.
This model applies 100% of the credit for a conversion to a consumer’s first touchpoint with your business (filling out a contact form, subscribing to emails or making a purchase).
Pros: If you’re focused on demand marketing, you’ll see which effort drove the first interaction with your business.
Cons: This model offers limited insight into the rest of the customer journey. Without knowing which marketing channels have been the most influential, you will struggle to justify marketing’s full impact on your business’ bottom line.
You May Also Enjoy:
- Attribution Still a Huge Challenge for B2B Marketers
- 2019 B2B Marketing Outlook Survey
- B2B Marketing Automation All-Stars
The opposite of first click attribution, last click attribution assigns all credit to the final touchpoint that leads to a purchase.
Pros: If you’re solely focused on closed leads and sales this model will be an effective strategy for you.
Cons: This model ignores the multi-touch path-to-purchase. And, considering it takes six to eight touchpoints, to secure a sale, you be left with zero visibility over influential actions along the way.
Here, we split the weighting evenly across every touchpoint in the customer journey. For example, if there are five touchpoints, each would get 20% of the credit. This model lets marketers make the best of the buying journey as a whole, rather than just focusing on one touchpoint.
Pros: By assigning equal credit to each interaction, you’re able to analyze your campaigns with multi-touch attribution. You can then optimize the entire customer journey instead of just one interaction.
Cons: If each interaction is given equal credit, you lose the ability to optimize for specific channels. It also levels out the value of all interactions, placing seemingly low value interaction such as an email click, on the same level as a high value conversion, for instance a demo request or call back enquiry, making it difficult to optimise for the most valuable interaction.
This model assigns 40% of the credit to the first and last touchpoints, then shares the remaining 20% amongst each of the touchpoints in between.
Pros: This ensures every touchpoint in the customer journey receives a portion of the credit whilst allowing optimization for the first and last touches. This means you can assign significant credit to the marketing channels that drove and converted the customer. You can adjust the weighting in your position-based model to reflect your business’ focus, e.g. lead generation through PDF downloads.
Cons: Assigning so much value to first and last touchpoints can be dangerous. This could result in two very low value touches being given too much credit. For example, should a first touch email click be given the same amount of credit as a paid search ad that resulted in a call to your business which converted over the phone?
Custom or algorithmic attribution is much more technical and requires a data specialist to create a tailor-made model that matches the customer journeys within an individual business. It gives marketers the best model to optimise every single touchpoint in their customers’ journey since it’s completely customisable.
Pros: Allows you to make amendments to the credit and weight assigned to each interaction and optimise the outcomes that matter to your business.
Cons: You’ll need to effectively build the model yourself, and with changing business outcomes your model weighting will likely change with them.
This multi-touch model also acknowledges that different touchpoints have different values. The idea behind this is that each touchpoint takes the prospect closer to the point of conversion, with each touch slightly more valuable than the last.
Pros: Puts greater emphasis on down-funnel conversions and can be advantageous to companies that have a sales cycle between 3–9 months. It can also help to guide strategy relating to the middle and bottom funnel marketing efforts.
Cons: Isn’t accurate in weighting importance. For example, say someone watches a demo the week before they convert, but the day after watching the demo they click on a link to read a blog post—is that blog post a more important interaction than the demo?
It’s no secret that attribution isn’t easy which is why so many marketers fail to get it right. But to thoroughly attribute your marketing activities and sales back to spend, it’s something you need to do.
Five Tips to Improve Your Attribution Strategy
1-Build a user experience (UX) map.
Consider the different interactions, and channels within the buying journey, where do potential customers enter? Are they bouncing? Why? Where? How long does it take to convert?
2-Don’t be afraid of radical change.
Test your data, test your methods and check your results repeatedly. If something isn’t working change it. There is no point wasting time, money and effort uselessly chasing something that isn’t producing results.
3-Set campaign objectives.
What is the objective? What story do you want to tell? This information should form the data you want to collect and that will sign post which attribution model will best suit your company.
4-Don’t forget the phone.
Even in the digital age, phone calls are still an important source of leads for businesses. Call tracking allows you to track each unique call back to the marketing source. By applying a small piece of code to your website and setting up click-to-call’for PPC ads, you can see where the telephone call journey started, and follow a customer’s journey right up until they called you.
5-Focus on quality, not quantity.
This could relate to almost any metric your company values its clients or customers by, whether it’s the amount they spend, the speed at which they convert, etc. And remember to look at lifetime value and not just the single pay-out at the start of a relationship.
Natalia Selby is the marketing coordinator for Mediahawk.