Turning Data Stories Into Jobs To Be Done
Recently at Ibotta, I got to work on a cross-functional analytics team tasked with discovering everything we could about our user activation funnel: the journey new users take from downloading the app to redeeming their first offer.
Conducting deep-dives into shopper behavior in the app has produced a number of valuable insights, like uncovering bugs and helping us optimize marketing communications. To help turn this data story into meaningful features for our users we used a ‘jobs to be done’ mentality to find out what drives their behavior.
Typically as an analyst, I use quantitative data to find and communicate meaning. In order to learn all about the circumstances that brought new users to the app, we decided to take a leaf from our Product Design team’s book and seek out qualitative data through user interviews.
Using the User’s Job to Look Behind the Data
Some questions about user’s journey are best answered by looking at data. What channels are driving the most users to download? How many times does a user perform a specific action before making a redemption? How many offers does a user see before they engage with content?
The deep-dive squad aggregated vast amounts of data to answer these questions about what was happening. We built models to help illuminate what factors about a user’s behavior in the app increased or decreased their likelihood to activate.
But behind that data there was an individual who was trying to make some kind of progress in their life, trying to get a job done, using Ibotta. Keeping those jobs in mind helped us make recommendations that were both data-driven and user-centric.
Job #1: Make Money
It’s pretty easy to identify the core job our users come to Ibotta to accomplish: make some extra cash. However, not every user who downloads the app winds up earning money. These users never ‘activate’, or take the critical step from download to redemption. This is why activation was a point of focus for our team — without it, our users don’t accomplish their primary job.
Our data told us that users coming to the app through referrals are more likely to activate. However, what we decide to do with that knowledge depends on finding out why referrals are more effective. Is the $5 referral bonus making the difference — should we make it $10? Are referred users getting more instruction on how to use the app than non-referred users — should we make more tutorials?
One of my theories was that new users who don’t already know how Ibotta works will start out with unrealistically high expectations. They might think the app is going to be magic, printing them money like an ATM. I assumed the classic happiness equation would apply.
Happiness = Reality ÷ Expectations
This would mean the lower a users expectations are, the higher their satisfaction will be no matter what they’re real experience in the app is.
In the Ibotta experience however, this equation doesn’t add up. During interviews, we pulled out a thread to explain why referrals are so effective— the power of testimony.
When we asked active users how they heard about Ibotta, they tended to say things like:
“My friend has earned hundreds of dollars using it, so I knew it really worked.”
As it turns out, many referred users have huge expectations for Ibotta! These folks are motivated by the success story of a friend or family member, so they set out to learn how to use the app and make their first redemption.
In contrast, users who heard about us through an ad seem to have pretty low expectations for their earnings. They tended to say things like: “If I can get a few dollars back, why not give it a try?” But as soon as the process feels like too much work, they are more likely to give up.
“It didn’t seem like it would be worth it, so I closed the app and forgot about it.”
Depending on the context in which someone is introduced to our app, they either see Ibotta as:
- A way to earn hundreds of dollars shopping for things they buy everyday, or
- A complex process that will only earn them a few extra dollars.
If you’re looking to “hire” an app to help earn extra cash, it’s no wonder you’d hire app #1 over app #2.
The data revealed that referrals are the most effective acquisition channel, and interviews showed us functional aspect of a referral — a testimony. With this, we can brainstorm ways to provide testimonies for non-referred users.
We can highlight real success stories in our marketing communication, publish blog posts describing how to join the “$100 Club” utilizing bonuses and high-value online retail offers, and try other ways to help convince new users to hire Ibotta as their money-making app.
Job #2: Personal Shopper
Since launching in 2012, Ibotta has steadily increased our inventory — the number of items with a rebate available in the app — to keep up with user demand for more offers on more products.
When digging in however, the data shows that the more offers a user sees before engaging with content, the less likely they are to redeem an offer at all. Which presents a paradox for us: by increasing the amount of content, we decrease activation.
By listening to users describe their first session in the app, we got more information about their emotional experience. New users described feeling overwhelmed.
“There’s so much to look at, I feel lost.”
“There are a lot of products, but not things that I usually buy.”
If it takes longer to find something of interest, they ultimately feel discouraged, like the app isn’t going to work for them.
Next, users described how they hoped they could use the app. Instead of less content they want better ways to filter what is there, like a way to create a gallery with only organic products, or find things that are already on their grocery list.
Users told us that hunting for the best prices on high quality products already feels like a chore. They aren’t looking to add another task to the list, they want a trusted source to help them filter the noise. They want to hire a personal shopping assistant.
This information serves to bolster our already ambitious Recommender and Search & Discover teams, intent on building a system that effectively surfaces relevant content as quickly as possible. Getting every detail about how shoppers use the app, we can generate ideas for how Ibotta can become a better shopping assistant. What if we gave our users the power to:
- Submit shopping lists and get suggested rebates that match their items?
- Get a reward for uploading any receipt, and let us learn what they like to buy before they scroll through content?
- Find out about on-shelf promotions and catch all their savings in one place?
- Set alerts for items they’re interested in, and receive a notification when they become eligible for a rebate?
- Use Ibotta as financial partner to track towards savings goals and milestones?
Not all of these ideas will make the cut when evaluating for effort vs. impact, but asking ourselves “What job do our shoppers need done?” can give us a better chance at deciding how to affect trends the data has revealed. Every time we make a change, we can check in with users: “Is this getting the job done?”
We’re Hiring!
Speaking of jobs to be done, Ibotta is hiring! If these kinds of projects and challenges sound interesting to you, check out our jobs page to learn more about the team.
