Social Attribution: Because Web Analytics Is the Wrong Tool for the Job

Social Attribution: Because Web Analytics Is the Wrong Tool for the Job Colin Zalewski Blogger Extraordinaire Simply Measured

Before I start, you should know that this story starts out sounding like a brag, but stick with me: justice is served.

Every year since 1973, Bellingham, Washington holds the Ski to Sea race on Memorial Day weekend. Hundreds of teams and thousands of participants race from the glacier on Mt. Baker to the frigid waters of Puget Sound. I was a year out of college and a friend from Bellingham invited me and some other semi-in-shape buddies to form a team. The race has seven legs covering over 90 miles. I’m bad at the outdoors, so I asked what my options were:

  1. Downhill skiing – We already had a guy who was way better than me at this.
  2. Cross country skiing – I don’t know how, and this sounds like an activity where you would be simultaneously uncomfortably cold and uncomfortably sweaty.
  3. Running – Pass.
  4. Road biking – See “downhill skiing.” 
  5. Cyclocross biking – Still unclear what this entails.
  6. Sea kayaking – Sounds dangerous.
  7. Canoeing – I’ve been in a canoe before, and this was the only leg done with a teammate.

So, hooray canoeing! In hindsight, I may have committed to this without asking enough questions. Key factors were not taken into account, such as: What body of water? How long is the canoeing leg? Where do I get a canoe? That last one proved to be the most critical element and what led me to writing this blog.

Image courtesy of Ski to Sea.
Image courtesy of Ski to Sea.

Neither my canoe partner nor I had a canoe, so we borrowed one from a nice local couple. On the morning of the race, we carried our canoe, oars in hand, to the waiting area near the river. We must have passed a couple hundred teams, and we noticed two things:

  1. We were younger and in better shape than most of them, and
  2. Their boats and oars looked different than ours.

Young and brash, we nodded at each other, thinking the same thing: “We’re gonna smoke these people.” Of the ~250 teams in the race that year, we expected our road biking teammate to show up somewhere in the middle. After the first three legs, we were in ~175th place. Turns out this guy was an animal on the bike, and we heard our team name called in ~40th place as he arrived.

We sprinted to the water, threw our canoe in and started paddling, honestly thinking, “We can compete to win this thing.” Then, something strange happened. We started getting passed.

First it was just a boat or two. We didn’t think much of it. They clearly had rowed a lot, and they had these sleek boats with carbon fiber paddles. “Whatever, we’ll catch ‘em later.” Then it kept happening. At one point, we got passed by a guy and his 14-year-old daughter, who then decided to take a water break 50 yards in front of us (this would be a taunting penalty in the NFL). We weren’t having any of that, so we started rowing as hard as we could to pass them and, even though it was just the current pushing them along in their carbon fiber canoe, we couldn’t catch them.

This happened for 18.5 miles.

After our road cyclist passed over 130 bikes, we were then passed by over 130 boats. We passed one other boat, which had previously passed us and then got stuck in a tree.

You see, my partner and I, despite our obvious physical superiority to any other athlete on the planet (see photo below), were using plastic oars and an old plastic canoe like you would’ve used at summer camp as a kid. We didn’t know any better. We didn’t know competing in this race required different equipment. We didn’t know we had the wrong tool for the job.

Canoe
My teammate and I, not going fast.

Have you ever wondered if you’re using the wrong tool? Perhaps you don’t know. Perhaps you don’t know that the tool you’re using to attribute the impact of social on your business wasn’t built for you or for social, just like the plastic canoe wasn’t built to be used in that race (which we totally would have won with the right equipment, and I stand by that!). Enough abstract metaphors, let me just lay it out there:

If you’re using web analytics to attribute the onsite activity driven by social, you’re using the wrong tool.

Here are three reasons why:

1. The numbers you’re seeing for social are just plain wrong. Whatever your web analytics is telling you–for total visits, pageviews, goal completions, and everything else from social–is simply incorrect. How can I make such a brash claim? 

a. Web analytics tools rely on referrers and user agents to indicate where someone came from to get to your website. When a visit happens and these indicators are not passed, your web analytics tool shrugs and plops it into the “Direct” traffic bucket. Yes, you do have direct traffic to your site, but much of that bucket is actually coming from social. This effect is called “Dark Social.” 

Clip5
Simply Measured’s Social Attribution solution gives you unprecedented access to Dark Social tracking.

b. This is a problem the industry has known about for years. Alexis Madrigal started talking about this back in 2012. We’ve been talking about it for a while, too. Why? Because it’s a big deal. Much of dark social traffic comes as a result of private messaging (text, email, WhatsApp, Messenger, etc.), and the volume of this activity, when properly attributed, may more than double what you’re currently seeing as the total contribution of social to your website.

Messaging Apps2. UTMs don’t solve this problem. Social marketers seem polarized on the topic of UTM parameters. Some have told me they hate them and are willing to do anything to make them go away. Others have invested so much time, energy, and money into fine-tuning their UTMs that they are in denial of the idea of them not solving for every possible scenario. The truth is, while UTMs may do a reasonable job at direct response attribution for your owned social content, there’s one thing that messes them up: people. If the only way people used social was to find brands’ content, click on it, and convert, UTMs would work pretty well.

The Quick Guide to Social Media Attribution Models

But, as we’ve discussed previously, what makes social different from other digital channels is the fact that people and brands are using the same channels, independent of one another, to share your brand content which leads to conversions. People don’t buy display ads for one another, but they do post, tweet, text, and chat with each other constantly. When someone goes to your website, finds a product for a friend, and copies the URL to text him or her, you can’t ask that person to append a UTM to their URL before sharing. In other digital channels, UTMs work because you are constantly following people and can see all their actions. In social, it’s people who choose to follow you, and they are constantly popping in and out of your line of sight by using dark social channels. When people go where you cannot follow, UTMs become ineffective.

3. Web analytics weren’t built for you. Frankly, even if the first two points weren’t the case, this should be enough. Why do we do attribution at all? No, it’s not so the boss will give us more budget. No, it’s not because we need to show ROI. We do attribution because we need to increase ROI. What we learn from attribution is what allows us to deliver the right content, to the right people, in the right places, at the right time, to ultimately increase return. I’ve yet to find a marketer who combs through their web analytics tool to help them decide what they should post next or how to strategize around their next campaign.

Even if the social numbers were right in web analytics, they were built for web marketing managers, not you, the person responsible for increasing the impact of social. You need a solution that doesn’t just tell you numbers. You need a solution that allows you to explore your data, find answers to the tough questions, learn from your audience, and elevate the insights that will make next month’s complete social ROI even higher than this month’s.

 

Social Attribution 101: The Beginner’s Guide

Don’t be like me out there on the river. If you’re trying to attribute social using a plastic canoe, it’s time to reconsider if you have the right tool for the job. I know you’ve been using that old canoe for a long time, and I’m not saying to throw it away. In fact, for the waters the rest of your digital team navigates, they should keep on rowing. But for you, whether you just started rowing or you’ve been fighting the current for years, the next 18.5 miles of river are going to be a lot easier–and yield much greater rewards–if you’ve got the right boat. Take it from me.

Interested in rowing the boat built for you? Request a demonstration of Simply Measured Social Analytics, including Social Attribution.

Learn about the complete social analytics solution

Request a Demo of Simply Measured

Find Out More
Colin Zalewski

Colin Zalewski

I'm a marketer with a passion for winning and keeping customers through strategically crafted messaging across the buyer's journey. I'm passionate about product innovation -- and, especially, Simply Measured Social Attribution and our Social Analytics API!