Live UX Case Study: Fixing Onboarding Drop-Off and a Weak Dashboard with Bar Rachamim - Product Builder Podcast episode with Eyal David
00:42:33

Live UX Case Study: Fixing Onboarding Drop-Off and a Weak Dashboard with Bar Rachamim

Host: · Guest: Bar Rachamim · About the host

In this episode

  • When users drop off during onboarding because of a complex technical step (connecting integrations), the solution is to simplify the connection to a one-click "authorize connection" level, and to prioritize integrations along two axes: how easy it is to connect, and how much value the data brings.
  • To bridge the empty gap at the start of usage, you can show dummy data or an industry benchmark ("this is the average in your industry until you upload your own data") — so the user understands the value even before connecting their own data.
  • During a long wait (loader), give the user a sense of control with an estimation — exactly like Waze, which doesn't solve the traffic jam but tells you how long it'll last — and in parallel you can use the time to collect more information or provide value.
  • User research should ask about past behavior, not future wishes: not "what are you missing" (you'll get "faster horses"), but "when did you run into a problem and what did you do" — in order to reach a real need.
  • A good dashboard prioritizes items by importance (the most important at the top left), guides the user toward action rather than just displaying information, and starts with a sketch on paper before Figma.
  • Deep, interactive onboarding that gets the wheels turning and asks the user questions increases engagement and a sense of belonging — but only up to a certain limit, without overloading.

Episode Description

Bar Rachamim, CEO of the design studio ProgressBar and a UX designer with experience in complex systems and cybersecurity, joins an episode unlike the usual ones: instead of a classic interview, she and Eyal dive together, in real time, into a UX case built with AI. The two analyze a fictional analytics company called Data Insight, build a persona ("Sarah from Nashville"), and break down two real product problems — high onboarding drop-off caused by complicated data integrations, and a dashboard that doesn't deliver enough value. Along the way, UX principles, research methods, KPIs, and real-world examples come up.

Read the full transcript
Full transcript with speakers and timestamps

Selected quotes

“One of the big, well-known rules in UX — Don't make me think.”
— Bar Rachamim
“A lot of times people ask what are you missing or what don't you have, and that's Ford — that's really asking what you want, faster horses.”
— Bar Rachamim
“Any UX puzzle whatsoever, I can stew over it — the more complex it is, over a few days, and it'll stay with me.”
— Bar Rachamim

Questions & answers

Who is Bar Rachamim and what does she do?

Bar Rachamim is the CEO of the design studio ProgressBar, a UX designer with about five years of experience, who over the past two years has specialized in complex systems including cybersecurity. She is also a UX lecturer at Shenkar, at Product for Product, and at Reichman University.

Why is this episode different from regular episodes?

Instead of a classic interview, Eyal and Bar take a UX case built with AI — a fictional analytics company — in real time and analyze it together so there's real "meat" to work on.

What is Data Insight's main problem in the case?

High onboarding drop-off: about 40% of users fall off at the data-integration connection stage, which is technical and complicated and usually requires an IT person. On the other hand, those who complete the process and see value — about 50% stay on the platform.

How do you prioritize integrations during onboarding?

Along two axes: how easy the integration is to connect, and how much value/data it brings. You start with the connections that provide the most value for the least effort, and not necessarily with the technical integrations that provide no initial value.

What is the right way to ask users in a research interview?

Ask about concrete past behavior, not about future wishes. Instead of "what are you missing" (which leads to "faster horses" in the Ford sense), you ask, for example, "when did you run into a problem and what did you do" — and that way you get a precise answer that justifies developing a feature.

UX User Experience Onboarding SaaS Analytics Product Design Dashboards AI