Attribution Modelling for Beginners

An effective digital marketing strategy combined with the right tools and technologies allows you to trace all of your sales back to a customer’s first digital touchpoint with your business. This is called attribution modeling, and it allows you to identify trends in the way people research and buy your product, helping you to make more informed decisions about what parts of your marketing strategy deserve more attention, and what parts of your sales cycle need refining. The marketing attribution is the science of determining what media are driving purchases and it is a way in which marketers assess the value or ROI of the channels that connect them to potential customers. In other words, it’s the means by which the customer came to know and buy your product or service.

Connecting the dots between marketing and sales is hugely important — according to Aberdeen Group, companies with strong sales and marketing alignment achieve a 20% annual growth rate, compared to a 4% decline in revenue for companies with poor alignment. As shown in this HubSpot video if you can improve your customer’s’ journey through the buying cycle by using digital technologies, then it’s likely to reflect positively on your business’s bottom line.

It seems simple enough when you think about the final sale. But does a customer ever go straight to a website and make a purchase? Rarely. Multiple channels and messages were responsible for the final buying decision, including the Facebook ad they initially clicked on or the email they received when they signed up for the newsletter. In an ideal world, you’d be able to track the entire customer journey from start to finish with personal anecdotes from each customer about why they made the decisions they did along the way. But that’s not realistic, or scalable.

With so many touchpoints to consider, operational marketing roles are becoming more and more complex. Luckily, there are a number of marketing attribution models that have been introduced and evolved since the digital boom to account for multi-channel selling.

Before someone purchases a product or service, they are exposed to numerous marketing “touchpoints”. These touchpoints cover a wide range of interactions, from seeing a television commercial to conducting online price comparisons on a comparison shopping engine (CSE) site. Attribution is the science of assigning credit or allocating dollars from a sale to the marketing touchpoints that a customer was exposed to prior to their purchase. When a paid product or service isn’t directly involved, so-called “conversion events” such as a signup or registration for a website can be used instead of a sale, and credit for such a conversion can be assigned to marketing touchpoints in the same way.

In proper attribution modeling, that credit is assigned proportionally to each touchpoint according to its influence on the customer’s purchase or conversion decision. The goal of attribution is to determine which touchpoints are producing a positive result, and, by using the cost of each touchpoint, an advanced attribution system can then show which touchpoints are profitable. This allows unprecedented optimization of marketing expenditures.

Early attribution models fell well short of the goal of understanding each touchpoint. For example, the “Last Click” or “Last Touch” model assigns all credit to the last touchpoint and ignored all earlier activity. The First Click model does the opposite, giving all credit to the first touchpoint and ignoring all others. Despite the extreme inaccuracies in these models, they are still used today by many advertisers to determine the value of touchpoints. For example, many affiliate publishers are still paid by advertisers today based on a Last Click model.

In advanced attribution modelling, relationships between various touchpoints are well understood and modeled accordingly. For example, customers that search for your product online after seeing an ad on broadcast TV will behave much differently than customers that come upon your product after doing blind internet searches for a type of product that they know they need. Understanding these interactions across marketing channels (broadcast TV to search) and across devices (tv to a smartphone, tablet, or PC) is the key to any attribution system. Failure to account for these types of interactions will make any attribution system inaccurate and flawed.

It is only recently, with technological advancements that allow user behavior to be followed anonymously and in a truly privacy compliant way, that advanced attribution systems have become possible. For the first time, we are able to observe all touchpoints leading to a conversion, and make highly accurate predictions of what particular marketing expendi-tures and interactions will produce what results. We know this is true based on the results obtained by advertisers using these advanced, and scientific, attribution systems. This has enabled a revolution in marketing measurement that many companies are already taking advantage of to dramatically improve their results.

Every sale has a journey and a story. Marketing attribution reveals that story. It takes you beyond vanity metrics, and gives you real answers: Where did the lead first meet your brand? What prompted them to buy, and what happened in between?

Let’s say you released an eBook last month. You promoted it with some social PPC, and sent it to your blog subscribers. You see that your overall conversion rate has gone up, which increased revenue, but which part of the campaign drove the best results?

What if you’re a business investing a lot in top funnel social media and display ads designed to drive brand awareness and direct traffic to your site? After being exposed to Google display ads, Facebook video ads, a buyer types your domain into search bar, navigating to the site and converting. Does all the credit go to search traffic? How much credit goes to the display and social ads? Marketing attribution answers that question.

It helps you map the journey from lead to customer, taking into account every touch point and interaction along the way. This detailed insight shows you exactly what works and what doesn’t — maybe your social campaigns fell flat while your email marketing is responsible for 80% of the results. In that case, it’d make sense to go all in on email marketing and stop spending on social. Without attribution you’d never know that.

An attribution model is the rule, or set of rules, that determines how credit for sales and conversions is assigned to touchpoints in conversion paths. For example, the Last Interaction model in Google Analytics assigns 100% credit to the final touchpoints (i.e., clicks) that immediately precede sales or conversions. In contrast, the First Interaction model assigns 100% credit to touchpoints that initiate conversion paths.

You can use the Model Comparison Tool to compare how different attribution models impact the valuation of your marketing channels. In the tool, the calculated Conversion Value (and the number of conversions) for each of your marketing channels will vary according to the attribution model used. A channel that predominantly initiates conversion paths will have a higher Conversion Value according to the First Interaction attribution model than it would according to the Last Interaction attribution model.

Attribution modelling example

Attribution is the catch-all term to describe how we assign credit for a conversion to different touchpoints in a user’s path to conversion. The standard attribution used by AdWords (and most other systems) is last-click attribution. A conversion registers as being caused by whatever touchpoint was the final one, and you don’t see the influential effect of earlier touches. For example a customer finds your site by clicking one of your AdWords ads. She returns one week later by clicking over from a social network. That same day, she comes back a third time via one of your email campaigns, and a few hours later, she returns again directly and makes a purchase.

• In the Last Interaction attribution model, the last touchpoint—in this case, the Direct channel—would receive 100% of the credit for the sale.
• In the Last Non-Direct Click attribution model, all direct traffic is ignored, and 100% of the credit for the sale goes to the last channel that the customer clicked through from before converting—in this case, the Email channel.
• In the Last AdWords Click attribution model, the last AdWords click—in this case, the first and only click to the Paid Search channel —would receive 100% of the credit for the sale.
• In the First Interaction attribution model, the first touchpoint—in this case, the Paid Search channel—would receive 100% of the credit for the sale.
• In the Linear Attribution model, each touchpoint in the conversion path—in this case the Paid Search, Social Network, Email, and Direct channels—would share equal credit (25% each) for the sale.
• In the Time Decay attribution model, the touchpoints closest in time to the sale or conversion get most of the credit. In this particular sale, the Direct and Email channels would receive the most credit because the customer interacted with them within a few hours of conversion. The Social Network channel would receive less credit than either the Direct or Email channels. Since the Paid Search interaction occurred one week earlier, this channel would receive significantly less credit.
• In the Position Based attribution model, 40% credit is assigned to each the first and last interaction, and the remaining 20% credit is distributed evenly to the middle interactions. In this example, the Paid Search and Direct channels would each receive 40% credit, while the Social Network and Email channels would each receive 10% credit.
• In the Data-Driven attribution model, the credit is distributed for the conversion based on past data for this conversion action.

A company’s customer journey can be complex, not only across channels but also devices. If you are investing in paid advertising as part of your marketing strategy, then should look for tools that analyze your conversion funnel.

Attribution tools help to identify which ad campaigns and marketing channels actually influenced someone’s buying decision, although there can be a learning curve involved. It’s very important to grasp and have a good understanding of which marketing channels are working and which ones are not assisting with conversions.

There are several different types of attribution models available for business owners to use with paid advertising campaigns. A hotel chain with a large budget might have broader awareness goals of introducing people to their latest building and brand, while a coworking software app may care immensely about every ad interaction on the signup journey. The gist is you will want to pick a model that best fits your business’s marketing efforts and goals.

The time decay, linear and the position based models all divide up one conversion across each visitor’s touch-point. How this can help you, is when you split up a single conversion across the steps in the conversion path…it allows for a better sense of a particular keyword’s value. All of these models allow you to use Adwords automated bidding and will give you more control over how much credit each ad and keyword gets for your conversions.

Regarding for example Google AdWords, the attribution modeling can help your company to:

• Reach customers earlier in the purchase cycle: Find opportunities to influence customers earlier on their path to conversion.
• Match to your business: Use a model that works best for how people search for what you offer.
• Improve your bidding: Optimize your bids based on a better understanding of how your ads perform.

As a marketer, you will want to define the objectives and goals in your AdWords campaign, and then pick a model that best aligns. Some paid acquisition strategies focus on maximizing efficiency why others may be more growth-oriented. So attribution is about trying out different models to understand how it changes the performance of different ads and keywords in regards to conversions. Performance that seemed typical under pre-existing last click models could be dramatically different when you evaluate those keyword campaigns with a new model. As a general rule, Keywords phrases that are clicked earlier in the conversion path (often generic terms ) tend to behave differently that keyword queries later down the conversion path ( like branded terms ). As you evaluate and update the model you end up trying, it also makes sense to re-examine the performance goals you have for your Adwords account. Some business can see better performance with lower costs-per-acquisition for keywords earlier in the click path once they move to a more aggressive model, like first click, shifting performance stats to reward keywords earlier in the click path, while a more conservative model, like last click, will reward keywords that occur later in the click path.

Once you have figured out the search intent of a keyword in the user journey of your website,  then you can start modify your ads and landing pages to start a more mutually beneficial conversion.  A set of keywords that tend to be the early stage(awareness) of the marketing funnel or user’s click path could benefit from a “less salesy” or more educational landing page. You could view these landing pages as inspirational landing pages instead of the typical late stage “buy right now” transactional page.

As an example, a person who conducts a search for “Coworking Space Dublin” might not be ready just yet to purchase a membership on his first click. After verifying that a particular search phrase tends to be earlier in the click path of your buyer’s journey in your Adwords attribution reports, you can start change your content to speak to them a little differently, both on the results page and on your site itself.   Establish your brand as a helpful trustworthy resource as the first in a conversion.

Data-driven attribution (DDA)

If you have enough conversions you might consider using a data-driven attribution (DDA). This can be slightly confusing if you are a beginner with using analytics tools, but using a data-driven approach takes the guesswork out of choosing a model.   It will depend on how much your monthly ad spend is, and if you have enough data (web traffic) but, if so, Google Analytics Premium services DoubleClick, Analytics 360 and Attribution 360 offer the use of data-driven attribution which can tell and show you the clearest possible picture of conversion success in your account.

Google Doubleclick and Analytics 360 use sophisticated algorithms to analyse all of the different paths in your account (both non-converting and converting) to figure out which touchpoints are help the most with conversions. Google’s algorithm will factor in the number of ad interactions, the order of exposure and the creative assets used in each conversion path. It uses a counterfactual approach: the algorithms contrast what actually happened with what could have happened to determine which ad clicks are most critical for a conversion.

Data-driven attribution benefits include:

• Pretty quick and easy to implement
• Values all steps in conversion path
• Works with automated bidding ( as do the other attribution model options )
• Even works on very short conversion paths

If your account does not have enough traffic to use (DDA) then consider a rules-based model.

Conclusion

As consumers explore new channels and companies adopt advancing marketing technologies, attribution becomes crucial to centering your entire marketing organisation around a common goal of revenue generation. In an effort to overcome the current marketing challenges marketers need to understand the combined power of audience insight and attribution modelling in order to optimize budgets and drive higher ROI, while delivering coordinated experiences that maximize business results.