Multi-touch attribution is a martech function of an advanced attribution solution. It’s a sure-fire method used to prove the value of a B2B marketing strategy in the form of down-funnel metrics that matter to the business -- opportunities and revenue. But multi-touch attribution is broad topic that spans a plethora of models, integrations, and strategies. Understanding the breadth and depth of this topic is no mean feat.
This is why we’ve written this full debrief that explains the details of multi-touch attribution to the letter. The models, methods, data, reporting, strategy, and results are all part of the process.
And if you’re looking for even more information, we constantly write about this topic, so you can subscribe to our blog (on the right) at your leisure.
What is multi-touch attribution?
Multi-touch attribution is best described in contrast to its counterpart, single-touch attribution. Single-touch marketing attribution ascribes 100% of the revenue credit for a customer to a single marketing touchpoint, which in turn ascribes the full revenue amount to the responsible channel.
For instance, if a prospect came to a website via an AdWords click, converted on a blog article, became an opp at a sales dinner, and verbally closed during an outbound call -- only one of those touchpoints would receive 100% of the revenue credit for that customer’s revenue contribution. This hardly seems fair. It would be akin to awarding only one relay runner a gold medal, even though their entire team played a part in winning the race.
Therefore, in order to divide the revenue credit for a sale properly, multi-touch attribution uses weighted modeling in order to allocate credit to the plethora of influential channels, campaigns, keywords, and touchpoints. Weighted touchpoint modeling assigns a percentage of the revenue credit for a customer to an array touchpoints, as defined by the respective multi-touch attribution model chosen by the organization.
What are the different multi-touch attribution models?
The term “multi-touch attribution” refers to the martech attribution solution that tracks a series of touchpoints through the funnel and assigns revenue credit to those touchpoints. And there isn’t just one type of multi-touch attribution model -- it’s a category to which a wide variety of different models belong. The common denominator is that all multi-touch attribution models attribute revenue to more than one touchpoint.
These weighted models come in several different shapes and sizes, and we’ve explained the main types of multi-touch models below. For a comparison of single-touch and multi-touch models, take a look at another one of our articles that covers all eleven marketing attribution models.
But for our purposes, we’ll discuss the following 6 multi-touch attribution models:
- Time decay
- Full path
 Linear multi-touch attribution model
Linear is by far the simplest model for multi-touch attribution. There is no difference in the assigned weights. Instead, linear attribution gives equal credit to each individual touchpoint along a buying journey. Each touchpoint is assigned the same amount of revenue credit, a percentage calculated by dividing the whole value of 1 by the total number of touchpoints along a buying journey.
However, this non-weighted system could also be considered a disadvantage. Linear multi-touch attribution doesn’t take into account the comparative importance of key touchpoints that mark a stage transition (FT, LC, Opp, Close). These touchpoints are often considered more influential than other interactions that didn’t result in a stage change. Compare this model to both U-shaped and W-shaped multi-touch attribution models to see the difference.
 Time decay multi-touch attribution model
A time decay model gives credit to more recent marketing interactions. Why would a marketer choose this model? Because B2B sales cycles are so long (ranging between 3-9 months), the early touchpoints may not necessarily be the most important touchpoints. Once the prospect is in the pipeline, more concern is often given to nurturing initiatives that incite a down-funnel conversion to an opportunity or to revenue. The shape of the time decay multi-touch attribution model would look similar to the following diagram.
Another form of the time decay model stops at the opportunity-stage, rather than tracking touchpoints post-opp through to close. With this modeling system, most credit would be given to the touchpoints between the lead and opportunity creation stages, which focus mostly on nurturing strategies. In a long sales cycle, this is often where the prospect spends most of their time. So, rather than disproportionately allocating revenue credit to quick stage transitions (FT and LC), the time decay model gives more credit to touchpoints that serve to move the prospect further toward a sale.
However, some marketers prefer a model that gives more credit to key stage transitions. These stage transitions would include:
- first touch (a user’s first-ever interaction)
- lead (user submits contact information)
- opportunity (user intentionally moves toward a purchase decision)
- close (user makes a purchase)
One of the models that affords these key touchpoint positions more credit than other middle touches is the U-shaped multi-touch attribution model.
 U-shaped multi-touch attribution model
The U-shaped multi-touch attribution model focuses on two key touchpoints, but it also tracks middle touchpoints between the two. The first touch, which represents a new prospect’s initial interaction with any tracked marketing activity (online or offline) is given 40% of the credit, while the lead creation touchpoint is given another 40%. The remaining 20% of the credit is divided across the middle touchpoints that occur between those two stage transitions.
One limitation of the U-shaped multi-touch attribution model is that it stops reporting after the lead-create stage, so marketers won’t have data insights into the prospect’s post-lead interactions. And, they also won’t know which touchpoint instigated the opportunity conversion (demo request, contact form, phone call, etc). If you’re reporting on leads, this model is great. But for multi-touch attribution users who want to report on down-funnel metrics and include post-lead touchpoints, the W-shaped attribution model is the way to go.
 W-shaped multi-touch attribution model
W-shaped multi-touch attribution modeling is similar to U-shaped, but the model includes an additional key touchpoint. The first touch, lead-create touch, and opportunity-create touch collectively receive 90% of the credit (30% to each touchpoint). The remaining middle touchpoints (between FT and LC, and between LC and OPP) are given the remaining 10% of the credit.
A fair number of organizations find the W-shaped model ideal for their needs. After a nurtured lead becomes an opportunity, the sales team takes over, and there are fewer marketing interactions post-opp. It’s not uncommon for an organization to decide that W-shaped multi-touch attribution is a good fit for their needs.
At the same time, more data never hurt anyone. And even those marketing orgs satisfied with the W-shaped model want the freedom change the percentages of revenue credit to meet their own reporting criteria. Perhaps they’d prefer to use a 25%, 40%, 25% distribution. This would require a custom attribution model. But before a custom model can be considered, we should first cover the full-path attribution model, as this highly extensive model paves the way for custom modeling.
 Full-path multi-touch attribution model
The full-path marketing attribution model is one of the most sophisticated attribution models available. In addition to the three key touchpoints that W-shaped attribution modeling tracks, a full-path model also includes the customer-close touchpoint. This includes post-opportunity stage marketing initiatives in the attribution model, and it incorporates sales activities into the mix as well. This way, the sales team’s follow up interactions can be measured in sync with touchpoints from marketing activities as well.
Across all funnel stages, 22.5% of the revenue credit for a sale is allocated to each of the key touchpoints. The remaining 10% of the credit is distributed to all of the additional touchpoints that assisted in moving the account down the funnel toward a closed sale (scroll down to see a screenshot of this type of reporting in real time).
 Custom multi-touch attribution model
A custom multi-touch attribution model is built upon the basis of the full-path attribution model. If an organization is already investing in comprehensive touchpoint tracking required for full-path, then they also have the capability to adjust the weighting of their attribution model to fit their individual reporting needs.
Rather than being locked into a static model, a custom model allows B2B marketers to allocate varying amounts of credit to touchpoints across their funnel stages. The method used to determine these weights should be dependent upon the company’s industry, the buying behavior of prospects, and the marketing org’s channel mix. Because no two companies are the same, it stands to reason that each company would be best served by forming their own attribution model, which is why custom modeling is arguably the most sophisticated model available.
Does multi-touch modeling (vs. single-touch modeling) really make a difference?
Sure, all of those multi-touch attribution models seem more intricate than single-touch models, but do they actually make a difference in the report? In our Bizible marketing org, we found that it does. Single-touch attribution reporting misrepresents ROI for the different funnel stages when compared to a multi-touch representation of the same marketing data. Take a look below.
Single-touch attribution (first touch modeling above) overestimates the number of leads generated by the organic search channel. It underestimates leads generated from the offline channel, and the multi-touch report indicates that the lead-to-opp ratio for that channel is much lower. Also, single-touch modeling underestimates the opportunities and revenue generated from the referral/direct channel. While paid media is a significant piece of the pipeline in both reports, the multi-touch graph depicts an opposite shape compared to the paid media graph in the single-touch report.
Suffice it to say, multi-touch attribution makes a difference. Looking at the single-touch report, marketers might undervalue their referral channels and/or overscale their focus on the offline channel.
How multi-touch attribution tracks marketing touchpoints
The multi-touch attribution models explained above might seem like a fantastic concepts in theory, but how do they work themselves out in practice? How can a single piece of martech track both online and offline touchpoint data so hygienically? There are three main ways that an advanced, multi-touch attribution solution can track marketing activities across all channels, platforms, and mediums. Here’s how it works.
This is a critical element of the tracking mix because a company’s website acts as the hub for most other marketing activities (paid media, paid search, organic, display, content, webinars, referrals, WOM, events, direct mail, etc). Each of those channels funnels into your website in some way or another, so integrated tracking in this area is foundational to other forms of tracking, such as UTM parameters and cookies.
Multi-touch attribution tracking with UTM parameters & cookies
For example, marketers can use UTM tracking when configuring social media ad campaigns. The UTM tracking code following the URL would include the platform (twitter), the campaign (persona, offer, funnel stage, etc.), and any other desired parameter. An advanced attribution solution can decipher a UTM medium and correlate that information with the down-funnels results generated by that ad campaign.
Cookies are another form of tracking, and one important function that they perform has to do with anonymous first-touch tracking. When a user initially arrives on your website, or clicks an ad for the first time, your marketing org doesn’t know who they are as of yet. In order to capture true first touch data, marketers need a way to track anonymous users. Long-term cookie tracking can do that.
When a first-timer arrives on a company site, the site “cookies” them by assigning their digital signature with a sort of tracking beacon. That cookie remains on that user’s digital presence until it’s cleared or it expires. Long-term cookies don’t expire for at least ninety days, so it gives the user plenty of time to make an identifying action.
Identifying actions usually involve submitting a name or email address on the company’s website by filling out a form. This matches the touchpoint activity history with the user’s name and email address, and all of that vital information is then pushed into the company’s CRM (which we’ll cover later on).
Multi-touch attribution tracking with martech API integrations
Application Program Interface (API) integration is a technical term for how software programs interact with each other. An API integration secures a seamless data flow between two applications. Advanced, multi-touch attribution solutions are equipped with these types of integrations between multiple types of martech -- ad platforms, marketing automation programs, optimization applications, live chat systems, and others.
Since prospects interact with multiple channels and platforms across their buying journey, a multi-touch attribution solution should be able to track and assimilate data from those same channels and platforms. Without API integrations, a multi-touch attribution solution wouldn’t be able to holistically assemble complete buying paths for users and accounts, and data hygiene would be easily compromised.
Ad platforms (such as AdWords, AdRoll, Bing Ads, Google Display Network, Facebook, Twitter, LinkedIn, etc.) have their own tracking protocols displayed within their user interface. However, isolated from its relationship to other attribution data, it can give a marketer the wrong impression of their results. Multi-touch attribution solutions assimilate this information and associate it with users’ other behavior, resulting in a complete picture of advertising yield.
Marketing automation platforms (MAPs), like Marketo, Pardot, and Eloqua, collect data that is crucial to include in multi-touch attribution data collection. API integrations between a multi-touch attribution solution and MAPs ensure that data can move from one program to the other without compromising its integrity.
Optimizations & A/B testing
Data from A/B tests are also important to include in multi-touch attribution reports. If a marketer can identify which tests generated the most down-funnel results (opportunities and revenue), then they’re able to base the results of their test on metrics that actually impact the business’ bottom line. Hence, martech integrations between A/B testing and optimization programs are critical to informed decision-making.
Live chat programs also integrate with multi-touch attribution solutions, which gives proper attribution credit to the higher-funnel efforts of those individuals (be them marketers or salespeople) who are responding to the chats. Data-level integrations are the only way to ensure that this information is properly assimilated into multi-touch attribution reports and dashboards.
How CRM integration creates accurate multi-touch attribution
A CRM integration is just as important, and arguably more important, than other martech integrations. When a multi-touch attribution program works hand in hand with your company’s CRM, it puts the attribution data where the action is. Both marketers and salespeople can use the same martech tool to track, monitor, and report on their strategies and performance. Salespeople have access to attribution dashboards that show the method behind marketing’s madness -- allowing the sales team to be more effective at their jobs.
Without this pivotal integration, attribution data would be far more difficult to use because it wouldn’t be actionable across the organization. Siloed information is what separates departments from interacting with each other. Creating a hub of mutually beneficial information inside the CRM bridges the gap and creates a greater degree of sales and marketing alignment.
Marketers can see how the salespeople go about picking up where marketing left off. Both teams can also run reports based on opportunities and revenue, which shows how well top and middle funnel marketing efforts converted downstream in the sales pipeline. In fact, all sorts of insights can be gleaned from tracking top and middle funnel activities to down-funnel results.
For instance, how much revenue did a specific landing page generate? How many customers resulted from a given ad campaign? What’s the rate of opps converted on a given email nurturing sequence? These types of questions can easily be answered when you have comprehensive data tracking throughout the entire funnel.
Multi-touch attribution reporting for account-based marketing
Advanced, multi-touch attribution solutions don’t just tell you keyword, channel, and touchpoint -- they track company and employee identifiers in order to form the basis for account-based marketing reports. For example, the report below shows four employees from the same company interacting with marketing touchpoints from the first touch through to close. W-shaped revenue credit is represented in the fifth column while full path revenue credit is shown in the sixth column.
King, Adams, Hill, and Fernandez had their own parts to play in moving this account through to close. This information builds within the CRM (discussed below) as the touchpoints are created by prospects. Before the BDR reached out to Tony Fernandez with an outbound call (touchpoint #7), the salesperson could tell that King, Adams, and Hill had previously interacted with other ads, emails, and content. The account was primed and ready for a stage change to opportunity.
Multi-touch attribution isn’t just another piece of martech
Advanced, multi-touch attribution is more than a piece of martech. Rather, it’s an integrated system that creates a stream of insightful data that informs marketing decisions. These decisions influence how budget is allocated and how strategy is formed, which in turn influence how marketers use their other pieces of martech (marketing automation, optimization tech, ad platforms, and sales outreach).
Multi-touch attribution undergirds a marketing strategy, and it centers the entire B2B marketing org around the ultimate goal of revenue generation.