Justifying Social Media: Position-Based Attribution and How to Prove Your Social Value

By on May 9, 2019

Digital marketing attribution is an increasingly critical topic, one which agencies and in-house teams alike are working on full tilt in an effort to find tools that can help them better justify their own efforts to internal stakeholders.

Attribution measurement is a rapidly-evolving and incredibly competitive space (Facebook and Google already have tools available or in the works), but it’s dogged by several major problems. Mainly that it’s incredibly complicated from a macro level, often convoluted at a micro level, and overall just plain difficult to understand if you lack extensive experience in digital marketing.

This lack of understanding means that many marketers aren’t even getting to the point where they’re thinking through attribution windows for their respective marketing channels, much less putting in the work needed to understand how their channels impact one another.

I want to share a secret with a lot of social media marketers out there: You’re not doing enough to sell your own work, and attribution modeling is the silver bullet you’ve been waiting for. Social media work runs the greatest risk of having its results marginalized when alternative attribution models are ignored, given the indirect (but powerful) impact social has on things like consideration and customer nurture.

If you’re a social media marketer and you don’t have a solid grasp of attribution modeling, step one is firming up your understanding of the concept. Whether it’s defending your social media budget or getting proper credit for the number of leads coming in the door, attribution is something you need to understand well enough to discuss with your internal stakeholders.  Then, you need to decide on the attribution model that best shows the impact your work is having.

Below I’ve broken out the most common ways marketers look at attribution modeling, and its impact.

ATTRIBUTION MODELING AND SOCIAL MEDIA

While the conversation around attribution is constantly growing and evolving, the topic itself has been around for a long time now.

Marketers have worked hard to understand the best ways to attribute conversions since they first had the tools to track traffic on their websites. Asking how much credit social platforms deserve for driving sales of a product might be relatively new, but marketers have been refining the process of understanding digital attribution across other channels for years. What’s interesting now is the idea that

Partnering web analytics platforms like Google Analytics or Adobe Omniture with something like Facebook’s internal reporting tools means that even day-one practitioners can take a stab at better understanding how the various channels contribute toward conversions in their marketing efforts. But there is a troubling issue inherent to web analytics platforms currently: they’re only offering up first touch or last touch for default attribution, and offer no simple way to shift to any other model that might better help you understand the ways conversions are occurring on your digital property. Facebook is, arguably, even worse—it looks at a default 28-day window that exaggerates credit in the broadest strokes possible.

As an example, here’s a look at one of my agency’s clients in the eCommerce space who run paid ads on Facebook and Instagram. In the first image, we’ll see what Google Analytics credits Facebook from a first-click and last-click model:

Now look at what Facebook self-reports on a 7-Day Click, 1-Day View model:

Wild stuff, right? There’s already an aggressive swing in the two Google Analytics models, and Facebook’s is outright comical in terms of how much credit they assign themselves in the purchase journey.

This stops being comical when you, as a marketer, have to go to your stakeholders and justify your expenditure in a channel. In the above example, Google barely credits Facebook enough to break even, in either model. And Facebook is so grossly over-crediting themselves as to be useless in broader discussions.

So, now you’re forced to think through the purchase path of an average consumer. They may see your product for the very first time as an advertisement on Facebook or maybe as a post on your Instagram feed. But customers rarely convert from a first touch standpoint. Days (or even weeks) can go by before those same customers—prompted by the awareness you drove in social — actually end up on your website to make a purchase. And 90% of the time, they’ll have landed there through a Google search or an advertisement you ran on AdWords.

So, if you’re using last-touch attribution models, your social channels get zero credit for that conversion (as in the example above). Social wasn’t the last digital touchpoint for the customer (remember, they landed on the site through an organic search or advertisement), and your paid search team gets to celebrate how great of job they’re doing at landing revenue.

But you also can’t assume that social media channels are contributing to every decision made on your site, as much as platforms such as Facebook would like you to. You can see the useless type of data produced when that happens.

What you need is a middle ground that assigns credit for actions fairly and doesn’t constantly leave you in a position where you have to defend your social media efforts.

SOCIAL MEDIA’S FAIR SHARE

In a perfect world, you’d have access to a data scientist who could build a custom attribution model that best fits your organization. Or, the time needed to invest in Google Analytics and build out custom attribution models that fit your business perfectly. More realistically, you’ll need to take your very first steps towards identifying an ideal model on your own and begin pushing for recognition of this model internally using data to back you up.

My personal recommendation is what Google Analytics defines as Position Based Modeling.

Position Based Modeling isn’t perfect, but it assigns 40% of credit to the first and last touch, while splitting 20% of the credit to all steps in between. These percentages can be adjusted by your analytics team to better reflect what you think is the fairest split between touch points. For example, you increase budget towards Facebook (first touch), change nothing about your budget in paid search (last touch), and see revenue numbers in paid spike dramatically in your initial Position Based Modeling. That would be a clear signal that paid search is taking more credit than is fair for the conversions taking place, based on the hard fact that you made no changes in that channel. Rather than let the model over credit paid search, you would adjust more credit in the model towards Facebook, the first touch, and make an anecdotal judgement on the results that then come out of that adjustment. That last part is key – a lot of this process is determined by critical thinking and anecdotal observation. Your team can adjust attribution allocation based on the data you track on a monthly basis, but you need to be making informed decisions along the way that data can’t do on its own.

This attribution model allows for you and your team to build a more distinct narrative around the work you’re doing in social media. Maybe paid social is the first touch, or maybe it is organic—either way, you get credit on a monthly basis for the primary KPIs you’re driving for your organization.

It also allows you to argue for your fair share of marketing resources and gives you the data points to back those asks up. We all know that social media is an important field—helping your team understand why is half the battle when it comes to building it up the right way.

WHAT’S NEXT?

As you leverage attribution modeling to help facilitate understanding of the value of social media marketing, you’ll almost certainly find yourself in spaces where data attribution becomes messy or difficult to parse out (most especially in eCommerce or verticals with numerous touch points).

As your understanding of attribution grows, so too will your desire for a toolset that allows you to better track and measure attribution across your marketing stack.

Social media marketers will want to look to Facebook’s Attribution Tool or similar soon-to-be-released tools from Google. These platforms allow you to easily move around different attribution models to assess the efficacy of your channel work and will allow you to better understand the role social media plays in your primary KPIs.

About Alex DeLeon

Alex DeLeon is the Director of Search and Social at Portent, a Clearlink full-service digital marketing agency, where he oversees SEO, paid search, and social. Alex has a decade of experience in marketing communications and digital strategy encompassing multiple verticals, including technology, sports entertainment, health tech, eCommerce, non-profit education, and the gaming. Follow him on Twitter.
Close

Like what you're reading?

Like us on Twitter, Facebook or Google+ for more!