Background

This was the last project I worked on at Pipedrive before they go through a mass layoff, and as the project was supposed to be developed a month later, I don't have the full story of this project.

Anyway, the idea behind it was to give certain integration a deeper and more interactive way to connect with Pipedrive users. Looking at the data we saw an opportunity to start with Calling integrations as they are very frequently used and didn't have a great way to sync the calling data.

Interviews

We had the hypothesis and now it was time to test it. From our point-of-view we have 2 sets of clients: the companies/partners and the end-users.

Partners: we gathered the most used integrations and set up calls with them (about 4 or 5). The interviews were a mix of technical, business and design questions, so in most cases I had my team joining the call. Talking with them we started to have an idea of the kinds of data that can be made available, possible pricing models, their experience with other CRMs, their users feedback, etc. So in the end we had some good insights on how this integration could work and accommodate different partners.

End-users: with the end-users I mostly did it myself, but sometimes had a PM around. I did about 10-12 interviews with users that matched our strategy. I had a set of questions that I wanted answers for, so I tried to drive the conversation in a way that would lead them to answer the questions without my need to ask them. Standard qualitative method in my opinion.
From them I notice the need from a quick and "permanent" way to access the integration, and at the same time a way to have it out of the way when needed. They wanted to be alerted about it and also be able to navigate to part of Pipedrive from the integration (opening a Person page from the caller contact for example).

Design rounds

Based on all that feedback I started to sketch some drafts. I was constantly sharing and getting feedback from the engineers, PM, and other designers.

This process took quite a long time. This project would mean that some changes in the core product would have to be made, so I basically had to have the whole design team on-board. Many rounds of design were done for this to happen.

Prototype and testing

In the end I had 4 prototypes ready for testing. The reason for this number was because I wanted to test different hypothesis, so I created different variations that, when given a problem, they would need to be solved in different ways - for example, I had different entry-points and I wanted to see how did they feel about it.

They way I made it possible was by using 2 prototypes per interview. I would give them a task to perform in one, get their feedback, and then try to do the same action on the second one. The order of the prototypes were different for each interview.

Shown below is the "winner prototype" so to speak. In this prototype you can:

• answer and deny the call
• minimize and re-open
• move the iframe around freely
• open the calling integrations panel
• start a new session
• start a new call

Conclusion

Sadly that's the point were the layoffs happened, so I'm not quite sure what happened to this project. I think it went forward but at this point I don't have any concrete information about the results of this project.