So I have some bad news. There are actually way more than five ways to fumble your marketing attribution. There’s a whole universe of marketing attribution mistakes out there.
Don’t get me wrong. Attribution is a phenomenal tool that can help marketers invest their time and budget more effectively, but it’s also easy to get wrong, especially if a team is just getting started with this methodology.
Not your team, though! Not after you’ve read this list of the most common marketing attribution mistakes and how to avoid them.
Marketing Attribution Mistake 1: Not Laying the Groundwork
Sometimes, when marketers are getting started with attribution, they want to cut straight to the advanced stuff. “Give me some of that marketing attribution you have on the shelf over there, nothing else please. And I need it by next month.”
To build a framework for producing marketing attribution that’s complete and accurate, you need to …
- Implement a standardized tracking strategy. That means coming up with a common process for naming your campaigns on all the channels or ad networks you use, and attaching the correct attributes and budgets so they carry through into your web analytics, CRM or eCommerce platform. This will make it possible to see what marketing activities produced which results.
- Create a process for efficiently aggregating campaign data from alllll the relevant sources, so it’s ready to be analyzed. Gathering all this data is something that may be easier said than done because marketing platforms and ad networks are notoriously finicky. Ideally, you should employ a dedicated platform that can automate your data.
- Put your data in a persistent, scalable storage solution, something like a data warehouse, not a ragtag collection of spreadsheets. A data warehouse will allow for long-term storage, reporting and analysis. It’ll make your attribution models smarter.
These three elements are must-haves. If you don’t have them in place as part of your attribution strategy, you’re going to be spending 80 percent of your time gathering and organizing data and only 20 percent actually making business decisions. (And it doesn’t even get into the staffing that you may need.)
Are there tools that can let you “do attribution” without going through all the steps above? Kinda sorta. Let’s tackle that in the next section.
Marketing Attribution Mistake 2: Thinking Google Analytics is the End-All, Be-All of Marketing Attribution
I don’t want to bash Google Analytics. I’m a huge fan of GA and have been using it for years because it delivers a lot of really useful, powerful information. Google Analytics makes it possible for pretty much any marketer to get started with attribution.
But we also need to be clear about what Google Analytics does and doesn’t do well. It’s great at tracking website activity. It can give you insight into what digital sources are sending traffic your way. You can also get information about customer activity on the other Google services.
But it isn’t always a great fit for data from non-Google networks. Again, with GA, the focus is on what happened on your website. Google Analytics can tell you if your last email sent visitors to your site, but it won’t capture other metrics that you need to know, such as deliveries or opens. Its models also won’t account for other real-world factors like weather or Covid-19.
Just remember all this as you get more experienced with attribution. At some point, you might outgrow Google Analytics. (One exception: If you’re an eCommerce site that only uses the Google Ads network, you’re probably in a pretty good spot.)
Marketing Attribution Mistake 3: Confusing Attribution Modeling with Media Mix Modeling
“Hey, I’ve got three years’ worth of aggregated data. How do I build an attribution model to optimize my future campaigns?”
Technically, an attribution model isn’t what you want here. You need a media mix model, aka a marketing mix model.
Attribution is about the paths to conversion. “This person (or this segment of our customer base) saw a search ad, opened an email and then made a purchase, so now we know to keep using search ads and email.”
Ideally, you should be using attribution to optimize campaigns while they’re still running. It’s more granular, more tactical.
Media mix modeling (MMM) is a larger correlational study. It looks at all of the media or marketing you had running during a given time and compares that activity to the results that were produced. You can even incorporate season, weather, economic conditions and other factors into your MMM analysis.
If attribution is granular, media mix modeling is the top-down big picture. MMM can be used to guide future planning and spend, and it can even help you forecast the results of a campaign. But it won’t show you how specific channels work together to influence particular customers or segments.
The good news is you don’t have to choose one over the other. By using both attribution and media mix modeling, you can make holistic decisions around planning and optimization.
Marketing Attribution Mistake 4: Using Incomplete Data
If you don’t account for all the relevant data in your reporting, you might come to the wrong conclusions and make the wrong optimizations.
You’ll see this happen a lot with online vs. offline retail data. Here’s an example.
Let’s say you notice that your paid search ads are doing really well, generating a ton of sales through your website. So you decide to take the money you’ve been spending on traditional advertising like TV and radio (which can’t easily be tied to specific sales) and put it behind paid search instead.
Unfortunately, that’s a mistake if you’re not including sales data from your brick-and-mortar locations in your analysis.
After all, 2020 was a HUGE year for ecommerce. But it was still just 21 percent of overall retail sales, according to one estimate. For many retailers, brick-and-mortar is still where most sales are made. Before you optimize your spend, you should understand what impact it’s having on those stores.
It isn’t always easy to discern what marketing channels are helping generate these offline sales, but there are some techniques you can use to capture point-of-sale data in the store. You might learn that, actually, traditional media is still pretty useful.
Look at all the data to the best of your ability. Making the wrong optimizations could be even worse than making no optimizations at all.
Marketing Attribution Mistake 5: Not Understanding the Campaign’s Goal
Here’s another case where attribution could lead you to optimize the wrong tactics.
Maybe your reporting shows that your branding campaigns aren’t that impactful. You can’t see it appear in any converting paths that are leading buyers to make a sale.
As a result, you want to shift your strategy and put more money into remarketing ads, which do show you when they’re involved with a conversion.
That’s not really fair to your branding campaign, though. It’s supposed to assist other channels, not directly boost sales. Your model should judge the campaign by its true goal and not penalize it just because it doesn’t show user-level activity.
So for your TV-heavy branding campaign, you might assess its impact by looking at overall lift in conversions.
You Can Avoid Marketing Attribution Mistakes!
The good news in all this? Once you get everything dialed in, marketing attribution is absolutely worth the effort. You’ll achieve clear proof of marketing’s impact. You’ll gain the insight you need to increase your conversions, sales and revenue. And make no mistake, that’s a huge win.
Multi-Channel Attribution Made Simple
Alight’s end-to-end analytics solutions feature multi-channel attribution modeling, built right into Power BI or Tableau, to help you reduce the time and cost to generate a conversion. Schedule a free solution consultation with our team!