How LinkedIn measures what matters to drive results
Developing a measurement strategy that provides actionable insights isn't always easy. Here are key tips to getting there.
Tracking the right metrics can have dramatic impact on campaign and business outcomes. That’s no secret. Yet as any seasoned marketer knows, measurement is often easier said than done. Internal hurdles can hinder efforts to achieve more meaningful measurement. How do you get over these kinds of hurdles to achieve measurement success?
There are four key steps to adopting a new approach to measurement that is oriented around the customer journey and business goals, said Jennifer Brett, Ph.D., who leads North American Insights at LinkedIn. Brett will be speaking in-depth on this topic at our MarTech Conference in San Jose this week.
Don’t under-rate mental preparation
The first step, said Brett, may sound obvious but will save you all along the way. Get into the right mindset. Changing the way and what you measure is often a long, difficult process. The bigger the shift or the longer the sales cycle, the longer the journey will take and the more important it is for the entire team to be mentally prepared.
“Measurement can be complex and hard to do. You’re not necessarily doing anything wrong by the fact that it is difficult,” said Brett by phone. “Marketers can think they are missing the quick fix, but accepting that it’s a journey that the marketing team is going on will help you mentally prepare up front.”
Research and be realistic about feasibility
Second, do your research. “There is a huge amount of information available — whether on media mix modeling, attribution models — so self-education is a big part of it,” advised Brett. Talk to other companies about what they’ve done, go to conferences, read white papers , watch webinars to learn as much as you can before diving into a measurement shift.
If you don’t have access to large amounts of data, what you want to move to might not be a good solution. Researching what’s feasible ahead of time will save you a lot of agony.
Determine whether you’ll need to train or hire
Finding measurement skill talent is a particular challenge, as we’ve seen from studies. Once you know what type of measurement you’d like to move toward or model you’d like to build, you need to ask if you have the right talent to do that, said Brett.
The level of investment needed in technical training can be intensive. Understanding how to leverage marketing technology is something teams can do, while moving from basic to more sophisticated marketing metrics with model building needed, hiring is often more efficient, said Brett.
That said, “You don’t always need to hire a data scientist for a lot of analytics and attribution work,” she said. You may just need someone with basic SQL skills, for example. “You don’t always need very technical people.”
Perhaps most important to implementing measurement change, said Brett, “You have to have a change management plan in place.” The plan should capture who and what the changes will impact and a timeline for hiring, vendor vetting, building, implementing and data collection.
Start stakeholder discussions earlier than you think
“Start socializing early with marketing teams and stakeholders who hear about measurement,” she advised. “Making the ‘why’ of the change really clear is incredibly important.” Be able to articulate the additional value of the change. “A multi-touch attribution model, for example, might allow you to optimize the channels you’re using more effectively and drive more revenue.”
Particularly with stakeholders, she said, start showing what the new metrics will be or what you plan to show as an output before you have data. When you have results they will be ready to receive and interpret the new reporting. “That can reinforce the ‘why’,” said Brett.
To learn more about LinkedIn’s measurement strategies, register for the MarTech Conference taking place April 3-5 in San Jose. Jennifer’s talk, “Measure What Matters: Data Strategy to Drive Results,” will be held at 10:55 on April 4.
Opinions expressed in this article are those of the guest author and not necessarily MarTech. Staff authors are listed here.