top of page
  • Writer's pictureDavid Avery

A Definitive Introduction to B2B Marketing Attribution Modeling

3 reasons why this is the most helpful B2B marketing attribution guide you’ll ever read. That’s because most resources:

  1. Are written by people trying to sell you software. I don’t have attribution modeling software to sell, so I will cover the pros and cons of building versus buying.

  2. Center around just one use case for marketing attribution, which is marketing performance optimization. That is important, but in most B2B organizations, marketing attribution plays a big role in overall revenue source attribution. This guide includes explanation and guidance for broader business attribution to provide a more comprehensive understanding of marketing attribution.

  3. Focus just on multi-touch attribution, which isn’t always the best option for marketing teams. In this post, we’ll share the full spectrum of models and provide guidance on which one is right for you.

What is marketing attribution, and why is it so important?

Marketing attribution is the ability to assign all or a portion of “credit” to individual marketing touches or campaigns in their relative contribution to converting a lead throughout their lifecycle into revenue. It is an attempt to assign all or part of an eventual deal (e.g., a dollar value), to individual marketing touches along a lead’s buying journey.

So, why do you need it?

By assigning a dollar value to a marketing touch point, you can then calculate the holy grail of marketing metrics: ROI. ROI is simply the value generated by a marketing campaign or channel above and beyond the cost associated to that marketing campaign or channel.

How to create an attribution model.

Firstly, while there are quite literally an infinite amount of attribution models, all models fall into 3 categories: first-touch, last-touch, and multi-touch. The most complex type of marketing attribution is multi-touch, but just as all models fall into 3 buckets, there are essentially only 2 types of multi-touch models: weighted equally and weighted unequally. The figure below summarizes the framework.

Let’s break each category down further...


In first touch attribution modeling, 100% of the revenue credit is given to the first touch of the lead. This can be useful in understanding where the lead first heard of your brand.


In last-touch attribution modeling, 100% of the revenue credit is given to the touch that was the most recent before the deal was created. This model puts an emphasis on the touch that created the conversion point from a lead to pipeline.


Unlike first or last-touch models that put all the weight on one specific touch, multi-touch models attempt to distribute revenue credit across multiple touchpoints along the lead lifecycle, from first touch to deal creation. While there are an infinite amount of ways to accomplish this (literally), there are only 3 approaches.

1. Linear

With a linear model, each touch is given equal credit. You simply take 1 divided by the total number of touches. So, if there were 4 touches (as illustrated in the graphic below), each touch would receive 25% of the revenue amount, as in the graphic below.

The challenge with this model is that it doesn’t put any specific weight on conversion points such as lead creation or deal creation, which are typically seen as important events in a buying journey. But the benefit of this model is that because it doesn’t favor any particular channel, it is the most equitable model.

2. Non-linear

A non-linear model weights each individual touch differently. There are many common forms of this approach, such as the U-shaped model in the graphic below, which weights the first and last touch greater than those in between. However, the reality is that you can provide any custom weighting to each touch that you want based on which events and order in the buying journey your marketing team finds important.

3. Time-decay

A time-decay model is the perfect blend of last-touch and multi-touch modeling. It puts more weight on the most recent touches that occurred prior to deal creation and less weight on those that happened earlier in the lead’s lifecycle, thus putting an emphasis on deal creation while still attributing some value to the touches that came before.

Which method is right for you?

Now that you have a sufficient overview of all the categories of models, how do you know which model is right for your marketing team? Well, the answer is… it depends.

Different models can be used to answer different business questions. It’s a common practice to run multiple models that allow you to answer various questions. For example:

  • If your team’s primary objective is to understand touches that impact pipeline creation, last-touch or time-decay are the best option.

  • For understanding digital creative and which channels attract the most qualified traffic, then first-touch is often the better model to choose.

  • For a gauge of the optimal marketing mix or better understanding your customer buying journey to build a better integrated program, multi-touch attribution is the way to go.

Marketing Attribution vs Business Attribution explained.

It’s common to use attribution modeling not just for marketing optimization such as marketing mix optimization or ROI calculation, but also for other line-of-business leaders to better understand marketing’s impact on the business. This is often referred to as “Marketing Sourced” pipeline or revenue.

Other sources of revenue are Sales sourced, partner or channel sourced, and inside sales or BDR/SDR sourced.

However, Marketing typically still touches leads even on the deals it is not directly responsible for sourcing. For example, in the buyer’s journey, a lead might download content from a website and then attend a webinar during their researching phase. Though it may be several months before they purchase, which was due to effort from another part of the organization, they still helped in the creation of the opportunity. This sort of impact is referred to as “Marketing Influenced,” as opposed to “Marketing Sourced.”

The percentage of Marketing Sourced contribution to revenue or pipeline often depends on the profile of the customer and priorities of the organization. For example, within organizations that sell primarily to large Enterprise accounts, marketing might only be expected to source 10% of total pipeline and be primarily focused on influencing existing deals. This graphic below from Sirius Decisions helps lay out the proper Marketing Sourced contribution targets within most organizations based on their GTM strategy:

Getting started with implementing attribution modeling in your organization.

Having an attribution model is an indispensable tool to have and critical to understanding the success of marketing tactics, as well as marketing’s impact on the wider organization. But how do you go about creating one for your team?

Step one is to align sales and marketing teams. Because marketing will be trying to assign some degree of revenue to its efforts, it’s important that the Sales team agrees. Here are the primary factors on which to agree:

  • What is the cut off date in the past, prior to the creation of a new Opportunity, before touches are considered too old to receive attribution (hint: 6-12 months is standard).

  • Which types of touches will receive attribution? Website visits? Content downloads? Event attendance? Email replies?

  • Does the model align to the way that a salesperson would think about a particular marketing touch “sourcing” an opportunity?

  • What further steps need to be taken to qualify the touch before it can receive credit? Does the lead need to be a good fit? Does it need to have caused the lead to reach a certain lead score? The more qualified the lead, the more likely sales will buy into the definition of Marketing sourced and the better your learnings from the model will be.

Once you’ve aligned across the organization on the terms and logic of the model, the next step is to build it.

Several MarTech vendors have robust platforms for building just about any of the model’s we’ve discussed in this guide. Here are the leaders in the space:

Your marketing automation platform or CRM might also have built-in attribution models. Here are just a few examples; check with your provider to understand if they have native attribution capabilities before you spend money on a separate tool:

Last but not least, you can always build your own attribution model with a business intelligence tool. Some vendors, like VisualIQ that use AI-powered attribution, don’t always expose their logic, which can lead to a black box that isn’t trusted in the organization.

Building your own model gives you the transparency and flexibility if you need to have complex rules or account for constantly changing business logic. At Optimax, we specialize in rapidly building bespoke attribution models for companies that have complex needs or want complete, unfetterd ownership of their models.

On many occasions, we’ve asked vendors if they can support an attribution model that takes into account Sales touches, or two sets of business logic before and after a certain period to account for organizational changes — but most fall down.

We can help!

Whichever option you choose, Optimax helps companies select the best model for their business questions, align teams on model logic, build custom models or procure the right vendor, and help teams gain immediate insight into their marketing performance.

Does your organization need help with Marketing Attribution? We’d love to meet. Contact a specialist today.

47 views0 comments


bottom of page