Not having much customer data can be challenging for the PM to know what the users like and what they don’t. There are several ways to gain possible insights on our product such as dogfooding where we, the employees use the product ourselves (assuming we are part of the target audience) or user research where we can get a qualitative idea as to what the customers will like, or try and take inspiration from what other big companies do. One such instance where I had to make a decision without much data available was when I had to work on an information architecture, I wanted to create 2 pathways to find an expert mental health professional on our app. One was directly by clicking on the category from the footer menu labeled therapy, which in turn had sections for different needs of the customer with their suffering from ASD including articles, video help and expert consultations. The other was by displaying a carousel of all kind of experts on the home screen with a CTA button of ‘view all’ clicking on which would take us to a page showing all the experts available and then we could use filters according to our needs to look at the relevant therapists. Here I took inspiration from various e-commerce websites like flipkart and amazon where they have an option to look for thing starting from the ‘categories’ section where you select a category of your choice and move ahead, and the other on being from a number of carousels on the homescreen showing phones for the premium, mid-range and the budget categories.

The overall approach when I don’t have data is to do: Experimentation and iterate with the data that I get from small experiments.

First I’ll share a bit about the project and then deep dive into it. The project was in the initial launch phase – where we wanted to validate our hypothesis / Validate our idea

Goal: Validate hypothesis – if it increases the productivity of users.

The situation was that: We want to create an app where the goal of the app is to increase the self-service rate and thereby increase productivity.

While we had figured out the features needed for the initial launch we had no clue – how many users will use it and did our MVP served the purpose.

Actions :

1. To mitigate the risks and pivot fast – I launched a small experiment where the feature was only available to a few thousand users – (a mix, of new, old, mid-size, and large sizes)
2. I set up a survey for these users
3. And had also initiated a few perks for being our first X users – to share us quick feedback and help us monitor their usage.

With this experimentation, we got the validation of our hypothesis and we were able to do company-wide release which has a very high impact on productivity