All Projects

Streamlined Research Onboarding Experience

How I turned a 2-hour onboarding call into a self-serve, 30-minute experience

Product Designer

8 weeks

Mobile App

The Story

This project came to me from one of our research assistants, who works directly with the research labs using our devices and app for their data collection. She walked me through her actual process: a 2-hour onboarding call with every new lab, PDF guides covering tips and error-fixing sent alongside it, and then an ongoing stream of emails and calls whenever labs ran into questions or issues. It wasn't a one-time cost, it was recurring for every lab we onboarded.

That's the problem I set out to fix.

This project came to me from one of our research assistants, who works directly with the research labs using our devices and app for their data collection. She walked me through her actual process: a 2-hour onboarding call with every new lab, PDF guides covering tips and error-fixing sent alongside it, and then an ongoing stream of emails and calls whenever labs ran into questions or issues. It wasn’t a one-time cost, it was recurring for every lab we onboarded.

That’s the problem I set out to fix.

The Challenge

Our research app has many capabilities, but new users struggled to understand how to take advantage of them, and how to collect the best data possible. Without proper guidance, researchers often collected poor-quality data and couldn't diagnose the issues they were having.

Our research app has many capabilities, but new users struggled to understand how to take advantage of them, and how to collect the best data possible. Without proper guidance, researchers often collected poor-quality data and couldn’t diagnose the issues they were having.

Lengthy training
Lengthy training

Every new lab required a 2+ hour onboarding call.

Every new lab required a 2+ hour onboarding call.

Poor data collection
Poor data collection

Many researchers collected poor-quality data due to wearing the device improperly.

Many researchers collected poor-quality data due to wearing the device improperly.

Hand-holding users
Hand-holding users

Our research assistant fielded questions from labs on a weekly basis, indefinitely, not just during initial setup.

Our research assistant fielded questions from labs on a weekly basis, indefinitely, not just during initial setup.

Low success rate
Low success rate

Three labs discontinued using our data collection tool entirely because they were never able to resolve the data-collection issues they ran into.

Three labs discontinued using our data collection tool entirely because they were never able to resolve the data-collection issues they ran into.

The scale made this more than a minor inconvenience. Over my time at Pison, we'd worked with more than 20 research labs, each collecting data weekly, sometimes for years. With a new lab onboarding roughly every other month on top of ongoing support for existing ones, our research assistant's support load only grew.

The scale made this more than a minor inconvenience. Over my time at Pison, we’d worked with more than 20 research labs, each collecting data weekly, sometimes for years. With a new lab onboarding roughly every other month on top of ongoing support for existing ones, our research assistant’s support load only grew.

Research

I approached this with three parallel tracks: behavioral psychology, competitor research, and iterative design reviewed with stakeholders.

  1. Behavioral Psychology

  1. Behavioral Psychology

This research shaped how I thought about presenting new information, specifically how users best absorb and retain new technology when they're learning it for the first time rather than being told about it in a single long session.

This research shaped how I thought about presenting new information, specifically how users best absorb and retain new technology when they’re learning it for the first time rather than being told about it in a single long session.

  1. Competitor Research

  1. Competitor Research

I looked at how companies like Mudra and Muse, both wearable/gesture-tech products with a similar "teach the user new hardware and software at once" problem, handled onboarding. Three things stood out as worth adapting: a visible progress/checklist structure so users always know where they are in the sequence, the ability to return to any piece of onboarding content later rather than losing it after a single pass, and clear, step-by-step language rather than dense explanation.

I looked at how companies like Mudra and Muse, both wearable/gesture-tech products with a similar “teach the user new hardware and software at once” problem, handled onboarding. Three things stood out as worth adapting: a visible progress/checklist structure so users always know where they are in the sequence, the ability to return to any piece of onboarding content later rather than losing it after a single pass, and clear, step-by-step language rather than dense explanation.

  1. Stakeholder Interviews

  1. Stakeholder Interviews

I talked with our research assistant, two data engineers, and two active researchers to understand the most common complaints, what guides were currently being handed to new labs, and what typically got covered in that first 2-hour call.

  1. Iterative Design

  1. Iterative Design

Design iterations centered on narrowing down exactly which information mattered most to surface to users, and pairing it with the right visual guidance, cutting content that was informative but not essential to getting a researcher collecting good data quickly.

The Solution

The final onboarding experience replaces the single long call with a self-serve, revisitable checklist inside the app itself:

A visible progress tracker ("0% Completed") sits at the top, so researchers always know how far through onboarding they are. This was directly adapted from the progress-indicator pattern in Mudra and Muse.

Five discrete, named steps, Wear your Pison Band, Connect your Pison Band, Use the App, Learn About the Pison Band, and How to Replay Onboarding, break what used to be one continuous call into individually digestible pieces a researcher can complete (and revisit) at their own pace.

"Wear your Pison Band" is surfaced first, directly addressing the most common source of poor data quality identified in research: improper device wear.

A dedicated "How to Replay Onboarding" step solves the "return to this later" need directly, rather than leaving researchers to dig through a PDF they were emailed weeks earlier.

A Skip option is available for researchers who don't need the full walkthrough, avoiding forcing repeat users through content they already know.

A visible progress tracker (“0% Completed”) sits at the top, so researchers always know how far through onboarding they are. This was directly adapted from the progress-indicator pattern in Mudra and Muse.

Five discrete, named steps, Wear your Pison Band, Connect your Pison Band, Use the App, Learn About the Pison Band, and How to Replay Onboarding, break what used to be one continuous call into individually digestible pieces a researcher can complete (and revisit) at their own pace.

“Wear your Pison Band” is surfaced first, directly addressing the most common source of poor data quality identified in research: improper device wear.

A dedicated “How to Replay Onboarding” step solves the “return to this later” need directly, rather than leaving researchers to dig through a PDF they were emailed weeks earlier.

A Skip option is available for researchers who don’t need the full walkthrough, avoiding forcing repeat users through content they already know.

Impact

The 2+ hour onboarding call became a 30-minute call, with researchers able to complete the bulk of onboarding themselves inside the app. On top of that, an in-app FAQ gave researchers a place to resolve common questions without needing to reach out at all, directly reducing the volume of follow-up emails and calls our research assistant had been fielding weekly.

Future Considerations

This project was scoped and interviewed narrowly: five stakeholders, all internal or closely connected to Pison, rather than a broader sample of researchers across all 20+ labs. A wider round of interviews, especially with labs that had struggled enough to discontinue use, could surface issues this version doesn't yet address. I'd also like to build in usage analytics on the onboarding flow itself, tracking which steps get skipped and which get replayed most, to make the next iteration data-driven rather than judgment-driven the way this one was.

This project was scoped and interviewed narrowly: five stakeholders, all internal or closely connected to Pison, rather than a broader sample of researchers across all 20+ labs. A wider round of interviews, especially with labs that had struggled enough to discontinue use, could surface issues this version doesn’t yet address. I’d also like to build in usage analytics on the onboarding flow itself, tracking which steps get skipped and which get replayed most, to make the next iteration data-driven rather than judgment-driven the way this one was.