Stop Measuring Everything!

Chris Savage Blocked Unblock Follow Following Nov 1, 2016

Too many companies view marketing as an arms race. If they can score their leads, optimize their conversion rates, and predict user behavior better than everyone else, they can win.

The problem is that being data driven no longer gives you the advantage it used to.

Today, everyone scores their leads with Marketo and A/B tests thirty different varieties of their landing page. You can’t get a competitive advantage doing that stuff anymore. You could say that as the percentage of marketers with a certain tech stack or using a certain tool approaches 100%, the competitive advantage you reap from it approaches 0.

Using data to scale your marketing is critical. But when we all have access to the same types of data, it won’t be the data that differentiates us — it’ll be the art.



The only way to stand out and make “art” is to set aside what can be tracked in the short-term and make decisions that align with your values.



That’s hard to do. Data, by giving you an “objective” standard of success, has a tendency to make you more risk-averse. But doing things that are impossible to track now is the only way to set yourself up for success in the long-term.

Losing our religion

Many types of great marketing campaigns cannot have their success measured in the immediate present — they work over the long-term. This is a problem for organizations that want to use data and move quickly, because they have to make decisions about what to do next with incomplete information about the future.



When you’re small, taking that leap is easier. Back in 2011, we spent $8,000 making t-shirts without even a plan to track whether or not they made us money.

A few years later, we’d put up a cute ad featuring our dog, Lenny, on a box truck outside the INBOUND ’14 conference. It turned out to be one of our most successful projects ever:

Last year, we decided to evolve from mobile billboards to the real thing.



The idea came in the middle of a period of significant growth for Wistia. We’d started using data more seriously in our marketing — testing conversion rates, more lead scoring — and we’d had a lot of quick wins. It was a departure from how we normally did things, but we kept growing, so we kept doing it.



Our initial plan was to measure success with the billboards by brand awareness. It turned out that was really hard to do. Instead, we decided to measure lifts in accounts across the geographic regions where the billboards were placed.



When we looked at our campaign numbers, we saw a lift in accounts — but a very slight one. The billboards were deemed a failure.

It’s hard to measure art

Then we went to visit some of the cities where we’d put billboards. And in one of them, I sat down with someone I’d never met before, who immediately perked up when I told him I worked at Wistia.

“Wistia? You guys did all those billboards for like two months! That was awesome! I have a picture on my phone!”

I asked him why he took a picture of a billboard, and he said oh, he’d heard of Wistia before. He thought it was remarkable that we had put Lenny on a billboard. Our silly joke “Video Hosting for business (not cats)” had made him laugh. Then I had to wonder how many other people had seen the billboards and had the same, scientifically unmeasurable reaction. And that’s when I realized how off-track we’d been.

Word of mouth had always been a major driver of our growth. One look at our passionate, growing Slack community and it’s clear that the connections we’ve formed with people are fueled more by “art” than “science.”



We’d done it right. The billboards had worked. But we’d measured them like they were science, and we’d convinced ourselves they hadn’t.



It took me a while to realize how far our pendulum had swung towards the science side of things, and how much things had changed since we were willing to put Lenny on a box truck with no ROI expectations.



What we were always really good at was being creative and doing things that other companies would never think of doing, with no expectation of a return.



This started because no one was paying attention to us and we had no traction. All we could do was try and create remarkable moments for people that would stick with them. If something seemed like a “Wistia” thing to do, we just did it.

When no one cares, you have to do whatever you can to stand out. What we forgot, and what we’ve had to re-learn, is that you have to measure your art differently than you measure your science.

Focus on people

The only method we’ve found that really works for measuring marketing that might seem immeasurable is to focus on the effect it has on individual people.



We take videos that we make and send them to people. If someone responds with “Holy shit,” then we know that we have something good.



Try asking people to fill out open-ended surveys about where they heard about you. Call into the markets where you advertised and see if people can repeat your message.



I ask new employees how they learned about Wistia, and a surprising number of them first heard about us from our year-end rap-up videos. If we’d measured the success of those posts based on their traffic, we would have been misled — they get just as much, if not less traffic than the average post on our blog.

Remember what makes you different

It’s only by asking the right kinds of questions and talking to people about what we’re doing, not measuring a conversion rate or traffic numbers, that we really get a sense of how we’re doing.



Clayton Christensen has a framework for evaluating why someone buys a product called the Jobs To be Done framework. Instead of looking at numbers to tell you why people buy, you need to talk with customers to understand what job they are hiring your product to do. When you can figure that out, decisions that were hard become easy.



The right data can help you protect a downside, find a local maximum, or provide a strategic insight that might be impossible to get otherwise. But data doesn’t help you stand out. To do that, you have to have the art. You need to put yourself into a position where you can take the risks that while hard to measure can provide the biggest long term impacts. It’s not the data that makes Wistia different, but art.



The fundamentals of building a great business haven’t changed, though: make something that people need or want. Be different. Be human.

Do what you’re the best at

When you first start a company, figuring out what makes it special is about creativity. Because you’re not scaling, you have essentially unlimited time to talk to your customers and experiment. You don’t have data, so you trust your instincts. The most creative ideas, the ones that most differentiate you, are the ones that win.



Then you grow, you start using data, and science takes over. It’s easier to digest for people who are new to the culture. It seems unbiased. When the numbers go up, you’re improving. When they go down, you’re not.



The problem is that some things can’t be measured like that. You can’t run a business today without data. But you also can’t let the numbers drive the car. No matter how big your company is or how far along you are, there’s an art to company-building that won’t fit in any spreadsheet.