Product Management

Redesigning the ATM Experience (Product Tournament)

This past Sunday, on January 21st, 2018, I participated in my first Product Tournament. It was organized by Headlight Labs, a company that sells take-home assignments to tech startups searching for talent. Take-home assignments are becoming a standard part of the product interview process, and they are also my favorite part of any job search. It’s a great way to show off my ideas right from the comfort of my own home. I have found that if a company has a take-home assignment I almost always get an invitation for an interview.

What’s a product tournament?

A product tournament is an event similar to hackathons but without coding. You’re given a situation, i.e. a failing company or product, and within three hours you have to come up with a product plan to turn the situation around. The proposals are then examined by a panel of product managers, engineers, designers, and marketers. Top candidates (only 13 participants) win prizes AND get to interview with top NYC tech firms hiring product managers.

My favorite part of the challenge is the blind review process. Basically what that means is your work has no name on it, and because the tournament is done online there are no indicators of the participants’ gender, age, experience or education. It’s the fairest shot you’ll ever get at a job. Your great ideas are the only thing representing you.

What was the assignment?

The first step of the process was filling out a 15-minute emotional intelligence questionnaire. It was a short personality test meant to assess traits such as assertiveness, self-discipline, and deliberation. Because the employers don’t meet the participants in person this test is a proxy for assessing the candidates’ cultural fit with the company.  

As soon as I completed the assessment, I pressed “GO” on the product challenge. The assignment was to develop a product plan for a company with a new technology called “SecureSmile”, designed as a FaceID competitor. The first pilot program of the technology used biometric identification- your smile- to authenticate access to ATM machines instead of a PIN, just like your fingerprint ID unlocks your phone. Unfortunately, the initial pilot program fell short of the 20% adoption estimate. People just weren’t using the technology. The issues that users had with the product were widely split between usability and security.

My task was to first assess whether the company should stay in the ATM/consumer banking market, or, pivot and tackle a new market opportunity. Once that decision was made I’d have to come up with a product plan for how we’d execute on the new strategy.

So, how did it go?

Overall, I did okay! The strongest points of my presentation included defining the problems, and coming up with a creative solution to the problems. I’m really proud of those parts. Unfortunately, I did not have time to do any research to support my ideas or to explain the implementation details of my solutions. I also did not specifically address any metrics of success in the plan. Lastly, my presentation just wasn’t very pretty and probably included a fair number of typos/grammatical errors.

That said, creating a product plan can take weeks in the real world, so this was a pretty good result for 3 hours of work.

How did I do it?!

I spent the first half an hour defining the problems with the technology itself and with the execution of the pilot. I wanted to understand why the pilot failed. This is what I was able to come up with:


Then, I spent another half an hour defining the user value, partner value and business value of using the technology within the banking industry.

At the first hour, I knew that my big decision was going to be to “Iterate on the ATM/consumer banking concept using lessons learned from the pilot program.”

Then, it was time for the fun part – coming up with solutions.

My FAVORITE part of the assignment was, by far, imagining the new experience of the product – one that solves the usability and security concerns. I imagined a two-step process:

While walking to the ATM

  1. Install your banking app, i.e. Capital One
  2. Log in & authenticate for usual services, i.e. fingerprint, passcode, or (in the future) smile
  3. Initiate an ATM usage session within the banking app-click button ‘USE ATM’
  4. Based on GPS location, the phone suggests closest ATM locations and the user selects the correct one s/he’s at
  5. Session countdown is initiated (i.e. 5 minutes)
  6. User selects the services needed on the app, i.e. ‘withdraw from checking $60 with 2 $20 bills and 4 $5 bills.’

While at the ATM

  1. Smile at the webcam
  2. The ATM recognizes your smile among the list of valid sessions for the ATM
  3. The software sends an update to the phone
  4. The phone presents a code valid for 30 seconds
  5. User types the code in the number pad of the ATM
  6. Access granted

In the real world, I’d actually sketch out all of those steps to help my audience visualize the workflow. I’d want them to imagine every step of the process. I’d compare this new experience to the current inconvenience of using an ATM.

If I had more time this section would have included at least 5 slides of a presentation with drawings and mockups for each slide. But…. right around this time, at the height of my excitement about the solution, I realized that I only had 35 minutes left to put together an implementation plan for this idea and to create a presentation! I made a quick decision that the MVP of the product will consist of expanding the existing API to include a secure session token – essentially, everything you’d get walking to the ATM and at the ATM in the above workflow. Version two of the product would consist of a process for registering the initial smile and updating the smile. I thought that these two things were doable by 4 engineers within 6 months.

The sad news is that the last part- the implementation plan- is actually the most important part of the assignment, and I simply did not have enough time to do it well. I did not explain how each feature would be implemented, what assumptions I had made about the implementation, or why I prioritized the order of development the way I did. These explanations are extremely important components of any good product plan.

And with that, I’m not sure that I will win the competition, but, I’m still proud of what I was able to put together in 3 hours. I loved thinking about the problem and I would love to do a tournament like this again in the future! All in all, it was a super productive Sunday 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *