Insights/Action/Outcome: People didn’t use previous analytics tools at FINN.no because they were “too hard.” By asking questions first, the data team created a learning loop and built Amplitude into existing processes to increase usage and understanding.
In my decade of experience in data analysis, I’ve learned this truth: from the perspective of non-analysts, analytics tools are often difficult to use. For the analyst community, these solutions come naturally. We have an in-depth understanding of data and often assume that as long as people understand the tool, they can access that data. Time and again, companies invest in tool training—but we neglect the ultimate goal. A product team wants to learn about their product, not learn about data and analytics.
If people decide a tool is too complicated and don’t use it, even the best functionality doesn’t matter. One way to address this issue isn’t with more detailed or frequent training. It’s by answering the questions that matter to them in ways that fit into their daily processes.
This was also the case for us at FINN.no, the country’s leading online marketplace for everything from secondhand goods to cars, travel to real estate, insurance to job listings. It’s one of the country’s biggest tech companies and most recognizable, reputable brands. With 30 million page views a day, our 200 developers face a significant challenge: to continue innovating a platform with more than 800 microservices.
I started at FINN four years ago as a web analytics lead. Eventually, I became the manager of the insight data team, whose mission is to always have data ready for analysis. Connecting the dots between data, analysis, and action wasn’t always easy.
We’re not all data analysts, so why should everyone learn their tools?
FINN was using Adobe Analytics, but decided to change the analytics tool because various teams felt it was too difficult to use. We’d done a lot of tool training with little impact, and it wasn’t a good match for our functionality needs.
Various teams felt Adobe Analytics was too difficult to use. We’d done a lot of tool training with little impact, and it wasn’t a good match for our functionality needs.
Next, our marketing department used Google Analytics, and the prevailing view among the non-analysts at the company was that Google Analytics would be super easy to learn because of its self-serve data capabilities. To accommodate, we ensured that the implementation resulted in high-quality data. Once again, we put a lot of effort into tool training, but as with Adobe Analytics, the new tool wasn’t well adopted. Users didn’t find it as easy to use as they thought they would. It was clear to the analytics team that Google Analytics would be only an intermediary step between Adobe and something else.
We began to research alternative analytics solutions on the market and learned that other brands in our parent company, Schibsted, used Amplitude Analytics. They said great things about the platform, and I got to see the capabilities when I gained access to a project from the Schibsted team.
The most important of these capabilities were funnels and the ability to turn everything into an event, making it easy to understand the combination of page views, click events, and other types of interactions within a funnel. There was also a lot of cool stuff for analysts that would help us get results quickly without having to be data scientists to solve the problems at hand. We decided it would be smart for FINN to adopt Amplitude Analytics for our organization.
A more strategic view of self-service data
The hurdle we had to jump through was teaching our team that Amplitude Analytics wasn’t “too difficult” for our users. We knew we had to take a different approach to training and shift to a more strategic, shared approach to self-service data. As an analytics team, we had to redefine what we meant by self-service.
Self-service data shouldn’t necessarily mean the user performs every step from data extraction to transformation, analysis, and interpretation. It’s not very motivating to tell someone, “after 50 hours of training, you should be able to answer this question yourself.” Self-service data could mean our insights team curates the content and puts the information into context for the user.
If information doesn’t lead to action, it’s not an insight; it’s just noise.
Instead of training people on a tool and letting them loose, we had to start with the end result and work backwards. If a product manager’s goal is to understand their product, our starting point as analysts should be to discover the questions that are most important to them and to help them answer those questions.
We developed an iterative learning loop to help teams prioritize and answer these questions with data. We know that teams have a lot of contextual information about their projects, and eliciting that information out of their heads can be hard. It comes down to being a good listener and asking a lot of questions. In our sessions, we push teams to reflect on each question by asking, “If you had the answer to that question, what decision would that allow you to make?” If information doesn’t lead to action, it’s not an insight; it’s just noise.
Once we know what information the team needs to take action, we start building visualizations and dashboards around the questions most relevant to the team. Connecting these dots helps people learn without becoming a tool wizard because they can look at a data point that once seemed mysterious and see how it relates to the question they asked.
Amplitude enables data learning and integrates with existing workflows
I used to believe dashboards were where numbers went to die. Too often, they sat unused and unnoticed. Using Amplitude Analytics changed my view. Features like tracking plans and Notebooks enable our learning process and help us explain where we can give context to the data and advise on next steps.
Building Amplitude into existing processes makes it easy for teams to adopt this tool and helps everyone align from the start.
The learning loop process makes us less vulnerable to changes in the team, as the entire team is involved as opposed to one single person. Amplitude Analytics has easy-to-understand visualizations so everyone and anyone can look at a dashboard, read the attached Notebook, and get up to speed. Amplitude Analytics also integrates with Miro and Slack, the tools our teams already use daily. Building Amplitude into existing processes makes it easy for teams to adopt this tool and helps everyone align from the start.
Schibsted has marketplaces similar to FINN in other Nordic countries, and all have started implementing Amplitude over the past 12 months. We are now a cross-country team, with people from the Swedish and Finnish brands on my team, and it’s great to see all of us leverage Amplitude Analytics to work toward shared goals. For anyone who becomes especially interested in data as they move throughout the learning process, we plan to create tutorials and how-to guides and provide links to Amplitude Academy content so they can investigate further independently. Our primary goal is to continue getting people into the right place to achieve their team-specific goals, such as improving product, performance, or campaigns.
The right process motivates everyone to develop their analytics skills, regardless of their role in the organization. In one instance, someone wanted to know where people travel at different times of the year. We built a dashboard to answer this question, and I walked him through the content. Not only did he find that useful, but he felt confident enough in Amplitude Analytics to continue exploring. Now, he does almost all the analysis he needs by himself.
After migrating to Amplitude Analytics, a wider range of people see a wider range of dashboards and analyses, and we get more questions from more people than we used to.
We’ve always had widespread access to analytics tools at FINN, but only a small core group of people used the analytics tool. After migrating to Amplitude Analytics, a wider range of people see a wider range of dashboards and analyses, and we get more questions from more people than we used to.
Building user confidence is the best way to grow a data-led culture
I’ve come to think of non-analysts at FINN as my customers and data as my product. My goal is for people to use data to do their jobs better. Data team members have become strategic advisers to help people answer their questions. Through this process, people have become more familiar with the tool, not the other way around.
For FINN, Amplitude is more than just an analytics platform—it’s a learning hub. Amplitude Analytics isn’t a standalone solution distributed in different pockets of the organization. It’s widespread, and its easy adoption means analytics is part of everyone’s process.
Using Amplitude Analytics to invest in processes and people, our data team gives colleagues a good data experience, which builds user confidence to try something on their own.
Encouraging “crawl, walk, run” is great, but a company will always be stuck in the crawl stage if people feel insecure about data. Using Amplitude Analytics to invest in our processes and people, our data team can give our colleagues a good data experience, which builds user confidence to try something on their own and keep having fun with data. Eventually, we will have enough people who can run, and we will have developed a data-led culture. From there, we can all move together into advanced analysis and all the other good stuff that comes from that.
To learn more about how to consolidate your data stack for optimal customer journeys, download our ebook below: