All Projects

Redesigning Team Collaboration

How I helped teams add 150+ users without the busywork

Product Designer

8 weeks

Web Design

The Story

Adding a new team member to a dashboard sounds like it should take seconds. At Pison, it didn't. Our customer audience grew from a handful of users per dashboard to 50, even 150, and they were stuck filling out one profile form at a time. The process was frustrating enough that our own CTO started doing it for customers just to spare them the pain.

That's the problem I set out to fix.

This problem was presented to me when the CTO had meetings all day and asked me to fill out the profiles for one of our customers. And that is when I learned that we were not getting the feedback from customers because our internal team members were doing this step for them.

Adding a new team member to a dashboard sounds like it should take seconds. At Pison, it didn’t. Our customer audience grew from a handful of users per dashboard to 50, even 150, and they were stuck filling out one profile form at a time. The process was frustrating enough that our own CTO started doing it for customers just to spare them the pain.

That’s the problem I set out to fix.

This problem was presented to me when the CTO had meetings all day and asked me to fill out the profiles for one of our customers. And that is when I learned that we were not getting the feedback from customers because our internal team members were doing this step for them.

The Challenge

The existing invite flow had four compounding problems:

Time-consuming process
Time-consuming process

Every profile had to be created individually, which didn’t scale for larger teams.

Every profile had to be created individually, which didn’t scale for larger teams.

Error-prone
Error-prone

Mistyped emails sent invitations to the wrong address, with no way to correct them after the fact.

Mistyped emails sent invitations to the wrong address, with no way to correct them after the fact.

Poor usage
Poor usage

Pison employees frequently added team members on behalf of customers to avoid making them suffer through the flow.

Pison employees frequently added team members on behalf of customers to avoid making them suffer through the flow.

Low satisfaction
Low satisfaction

Both internal and external users complained about the feature regularly.

Both internal and external users complained about the feature regularly.

Of all the organizations we hosted on our dashboard, I learned that 85% of them had been set up by internal members and the other 15% were set up with the assistance of an internal Pison member.

This was a big problem. If people couldn’t set up their organizations, they couldn’t even use the product we worked so hard on.

Research

I approached this with three parallel tracks: user research, competitive research, and iterative design reviewed with stakeholders.

  1. User Research

  1. User Research

I talked with five current users to understand how they used the feature, how long it took them, and where the friction was worst. I conducted one moderated interview, three unmoderated, and one interview reflection.

What came back was clear and consistent:

Current User
Current User

“Adding users one-by-one is infeasible at the scale of 50–150 users.”

“Adding users one-by-one is infeasible at the scale of 50–150 users.”

Current User
Current User

“Completing individual forms for each team member takes a long time and most of the fields are optional.”

“Completing individual forms for each team member takes a long time and most of the fields are optional.”

Pison CTO
Pison CTO

“I often create teams and add team members myself in order to avoid making potential customers go through the frustrating process.”

“I often create teams and add team members myself in order to avoid making potential customers go through the frustrating process.”

  1. Competitive Research

  1. Competitive Research

I looked at how other collaboration tools solved the same problem. Google Drive and Figma's bulk-invite patterns stood out as the clearest, most recognizable models to build from, both let you invite someone to a file with just two pieces of information: an email and a permissions role. That simplicity removed an exponential number of clicks compared to our original per-user profile forms.

I looked at how other collaboration tools solved the same problem. Google Drive and Figma’s bulk-invite patterns stood out as the clearest, most recognizable models to build from, both let you invite someone to a file with just two pieces of information: an email and a permissions role. That simplicity removed an exponential number of clicks compared to our original per-user profile forms.

  1. Iterative Design

  1. Iterative Design

I designed three rounds of solutions and reviewed each with the Chief Product Officer, a Product Manager, and a Software Engineer, refining until we landed on an approach the group was confident in.

Round 1
Round 1

Reduced clicks and introduced error messaging, but still asked for information that wasn’t actually necessary.

Reduced clicks and introduced error messaging, but still asked for information that wasn’t actually necessary.

Round 2
Round 2

Stripped things down further, but introduced a modal that blocked the user’s view of the list of already-added teammates.

Stripped things down further, but introduced a modal that blocked the user’s view of the list of already-added teammates.

Round 3
Round 3

Kept the minimalism of round 2 but moved away from the modal, keeping users on the current page instead.

Kept the minimalism of round 2 but moved away from the modal, keeping users on the current page instead.

The Solution

I designed a bulk-add feature deliberately consistent with the patterns users already knew from Google Drive and Figma, prioritizing minimal clicks and time on task.

Before

Before

Complex Multi-Step Process

Complex Multi-Step Process

Minimum 5 clicks per added member; unnecessary profile fields; a long and confusing form; and a low task success rate.

Minimum 5 clicks per added member; unnecessary profile fields; a long and confusing form; and a low task success rate.

After

After

Streamlined Flow

Streamlined Flow

Maximum 3 clicks to bulk-add members; only required fields; simple, recognizable inputs; and improved success rate and time on task.

Maximum 3 clicks to bulk-add members; only required fields; simple, recognizable inputs; and improved success rate and time on task.

Key Improvements Shipped

Simplified invite flow

Reduced the amount of information customers had to enter before inviting teammates.

Bulk-add with better feedback

Added helpful, correctable error messages before users sent invitations to the wrong address.

Group role assignment

Made it possible to assign roles to multiple invited teammates at once.

Role permission explanations
Role permission explanations

Added real-time explanations so users understood what each role allowed before sending invites.

Added real-time explanations so users understood what each role allowed before sending invites.

Impact

Improvements to time on task, human error, and proper product usage.

Time on task improved dramatically: what used to take 35 to 60 seconds per individual profile became bulk-adding around 50 users in under six minutes.

The redesign also gave us room to design for human error, surfacing feedback to users before mistakes happened, rather than after.

The clearest signal of success, though, was behavioral: Pison team members, including the CTO, no longer needed to circumvent the setup process for customers. After launch, no new organizations required internal assistance to get set up, a complete reversal of the 85%/15% split that motivated the project in the first place. It gave both internal team members and customers far more confidence in the product.