Menu
Support
Free Trial

The Biggest Problems with Ecommerce Attribution

by Kathleen Booth, on Nov 25, 2020 9:00:00 AM

Ecommerce platforms rely heavily on their digital marketing efforts to drive sales. As such, knowledge of what is and isn’t working is essential to justifying ad spend or effort, as well as choosing which strategies to use long term. The process of monitoring and attributing revenue to specific activities isn’t always as straightforward as you’d hope.

For example, your team might use something like Facebook ads and weekly emails to attract buyers. The question that marketing teams typically ask is: which of these two methods can we attribute to the most order placements or the highest revenue amount? 

-- Article Continues Below --

ecommerce resource center

Read Now: The Complete Ecommerce Resource Center

This is called revenue attribution.

When you can answer this question, you can strategically choose when marketing activities should be eliminated, improved or maintained. More importantly, you can be sure you're making the best use of your marketing dollars.

Many ecommerce companies struggle with revenue attribution. Let’s talk about the ways you can overcome common challenges and improve your attribution modeling. 

Ecommerce Revenue Attribution Models

Revenue attribution is incredibly important to not only your marketing strategy, but also your marketing budget.

Modeling revenue attribution is a great way to visualize this data. There are many different kinds of models, each with their own focus:

  • First click: This model attributes a final purchase to the first interaction (click) with your website. If a buyer found your website through a Google search and then visited three more times before purchasing, the final sale would be attributed to the Google search.
  • Last click: Basically the opposite of first click attribution, the last click model attributes the sale to the buyer's last interaction. If a buyer visited your company’s website and then made their final purchase after tapping on an Instagram link, that would be the attributed channel. 
  • Position based: This model attributes each channel to a certain percentage of the final sale based on influence. Generally, the first and last click will be at 40% and any channels in the middle will be at 20% attribution.
  • Time decay: This model attributes the first click with the lowest percentage and the last click with the highest attribution. Maybe a buyer visited your website first through an email, then through a Facebook ad, and then through a Google search. The email may get 15%, the Facebook ad 30% and the Google search 55% using a time delay model. 
  • Linear: Each channel gets an equal percentage of attribution towards the final sale. 

Each of these models has advantages and disadvantages. In fact, one of the biggest challenges with revenue attribution is simply choosing the attribution model you want to follow.


Read The Ultimate Guide to Ecommerce Marketing

Discover a solution built to help you block the negative effect coupon extensions have on your bottom line.

Read Now


Mistakes in Ecommerce Marketing Attribution

Even with the help of established attribution models such as those described above, many companies struggle to get started with attribution or don’t know how to use the data they receive.

Even if you already use revenue attribution, do you feel confident that you’re using it correctly? 

Something as simple as incorrect UTM campaign tags can generate inaccurate data. Put simply, if you are inputting the wrong data into your models, you won’t get accurate results (garbage in = garbage out). 

Some of the biggest mistakes, or pitfalls, that companies may run into when building or monitoring revenue attribution include:

    • Not using revenue attribution at all: The biggest mistake you can make is not monitoring or measuring your attribution at all. Remember, done is better than perfect. If you don't currently have revenue attribution modeling in place, any model is better than no model at all.
    • Under or overvaluing a specific channel: This is really something that you’ll have to iterate on and get right over time, but one problem commonly made is forgetting a channel when assigning attribution levels.
    • Inaccurate tracking: This is the most common problem you’ll run into. It is easy to build a model, but continually setting up the methods you need for tracking, and making sure they are correctly tracking, is the hardest part. 
      • UTM tracking: The most common offense in this area is forgetting to use simple things like UTM tags to track the source of traffic back to a specific activity.
      • Coupon codes: Another common problem that most ecommerce merchants aren’t even aware of is the effect of coupon browser extensions (ex. Honey, Wikibuy and the like). Coupon codes for specific audiences (such as affiliates, VIP customers, or new newsletter subscribers) are a common method used by merchants to attribute revenue to specific campaigns. Because they make limited use coupon codes available to anyone with the extension, coupon browser extensions make it possible for buyers who weren’t influenced by the campaign associated with that code to use it, completely messing with your attribution reporting.
      • Influencer fraud: This is another issue related to coupon extensions where influencers or affiliates who you’ve created a specific coupon code for can submit that code to coupon browser extensions like Honey (opening that coupon code up for any Honey user), making it appear as though more sales have been completed based on their efforts than actually have.
      • Using the wrong model or not enough models: This issue arises mostly because companies have chosen an attribution model without knowing about the other options out there, or not having the tools in place to build the attribution model that fits them best.
      • Not adjusting marketing strategy to attribution data: It's great to collect data, but it only becomes valuable when you use it to continually update and inform your efforts on the front end.

These issues are pretty common amongst marketing teams. Even the most experienced ecommerce marketing and revenue teams may struggle when it comes to revenue attribution.

The good news is, there are ways to ensure that you are correctly using revenue attribution. 

Effective Revenue Attribution for Ecommerce Marketing

Even if you have made one of the mistakes outlined above, we’ve got some tips and tricks to keep you on the right track.

Revenue attribution can become extremely technical and complex if you dig deep. In keeping it on the surface level, however, we can look at some of the ways that marketers are able to effectively pull off revenue attribution. 

Let’s break this down into a few steps:

1. Make sure you are receiving accurate data

When you input your data into an attribution model, it is important that the data is accurate. This is especially important if you use Google Analytics, as many companies do. Google Analytics relies on UTM campaign tags to sort channel traffic. Anything without a UTM campaign tag gets sorted into direct traffic. 

This means that specific channels could have extremely high numbers compared to other channels. So, before you start any revenue attribution, go through and correctly tag your marketing campaigns. 

Additionally, if you are using coupon codes as a method to track attribution back to specific activities, make sure you are aware of how your users' coupon browser extensions may be affecting your attribution reporting. The best way to get ahead of this is to find a way to block extensions from being able to auto-inject coupon codes at checkout.

2. Choosing and using attribution models

With so many attribution models, you may not know which one to choose.

The key is not choosing just one model –– it is knowing how to interpret all of the models.

Each revenue attribution model can tell you something different about your marketing strategy:

  • First click: How did you reach a new buyer? What attracted them to your company?
  • Last click: What was the deciding factor that pushed them to purchase from your company?
  • Position based: Which channel brought the buyer in? Which channels continue to generate interest? Which channel was the deciding factor?
  • Time decay: What is bringing repeat buyers in?
  • Linear: Which channels are most consistent in influence?

Now, instead of relying on just one attribution model, you can choose specific data to answer a specific question.


Protect Your Ecommerce Margins

Increase average order value by blocking discount codes from coupon extensions like Honey, CapitalOne Shopping, Piggy, and more.

Learn More


3. Adjust your marketing strategy and budget


Last, you should actually use the information that you are learning from this process.

If Facebook ads are not doing much to influence buyers, then you may want to try a different strategy or drop the ads completely. Many companies use their revenue attribution models to save money in marketing or figure out where to place the bulk of their spending to generate the most results. 

When you interpret your attribution models, ask yourself what the data means for your strategy. What can you improve upon? What should you continue doing? What should be eliminated?

The answers to these questions can help you build a strong marketing strategy that will maximize growth and revenue.

Ecommerce Revenue Attribution Tracking Solutions

When it comes to revenue attribution, coupons and other digital savings offers may be a top of funnel entry point for new buyers.

While offering coupon codes to prospective buyers has many potential benefits, the challenge is when that process is muddied by coupon browser extensions. 

Buyers having an all-access pass to your limited use coupon codes will dilute both the impact of those coupons and your ability to understand how well they are serving you.

Bottom line? Coupon extensions can negatively impact your revenue attribution by creating inaccurate data.

New call-to-action

Topics:ecommerce

Our blog

Where businesses come to learn more about protecting the points of digital engagement with their customers, audiences and users.

Subscribe to Updates