Skip to content
Lenny's PodcastLenny's Podcast

How to scrappily hire for, measure, and unlock growth | Crystal Widjaja, Gojek and Kumu

You don't need lots of employees to achieve impressive growth, but you do need a unique approach to hiring and measuring what matters most. Crystal Widjaja has used scrappy tactics to unlock massive success for Gojek (a wildly successful ride-share app in South East Asia) and is currently the Chief Product Officer at Kumu. In this episode, she shares the exact strategies she’s used as a product leader to hack growth, hire the best, and perfect data collection. Join us. Find the full transcript here: https://www.lennysnewsletter.com/p/how-to-hire-for-measure-and-unlock — Thank you to our wonderful sponsors for making this episode possible: • Amplitude: https://amplitude.com/ • Flatfile: https://www.flatfile.com/lenny • Eppo: https://www.geteppo.com/ — Where to find Crystal: • Website: https://www.crissyw.com/ • Twitter: https://twitter.com/crystalwidjaja • LinkedIn: https://www.linkedin.com/in/crystalwidjaja/ — Where to find Lenny: • Newsletter: https://www.lennysnewsletter.com • Twitter: https://twitter.com/lennysan • LinkedIn: https://www.linkedin.com/in/lennyrachitsky/ — Referenced: • Generation Girl: https://www.generationgirl.org/ • Experiment Design Question Example: https://docs.google.com/spreadsheets/d/1tSrEGpvg19OMKrRg45HGvsZBGz8BkeVc/edit#gid=1841057518 — In this episode, we cover: [00:00] Crystal's path to product [09:40] How Crystal decided to join a risky startup [11:31] Why haven't super apps emerged in the U.S.? [13:15] What startups in the U.S. can learn from companies in Asia [16:35] How to get intentionally scrappier in your organization [18:28] How to get and utilize scrappy and small data sets [21:26] How to increase retention [22:28] What does and doesn't work in growth? [25:01] The conversion rates Crytsal looks for to determine viability and success [28:03] How founders should think about approaching growth [32:44] Figuring out how to grow and the resources you have at your disposal [34:53] Optimizing the funnel to make growth happen faster [37:40] Crystal's biggest lessons on unlocking growth [40:07] Why most analytics efforts fail and how to avoid failure [44:24] Signs your organization is getting analytics wrong [46:17] The best resources for figuring out how to do analytics right [47:05] Crystal's recommendations for metric tracking stacks. [48:37] How you should set up your growth team originally [51:38] Integrating growth teams or separating them? [52:43] Who should be the first growth hire? [53:55] How to hire a great growth person for your organization [56:55] How to help women and young girls get into STEM (a non-profit that you started) [1:02:07] Where to reach Crystal

Crystal WidjajaguestLenny Rachitskyhost
Jul 30, 20221h 3mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Scrappy growth tactics, smart analytics, and hiring to unlock scale

  1. Crystal Widjaja, former Head of Growth at Gojek and now CPO at Kumu, shares how she’s driven massive consumer growth across Southeast Asia using scrappy, high‑leverage experiments and rigorous data thinking.
  2. She explains why American startups can learn from Asian companies’ bias toward doing things that don’t scale, like renting stadiums to recruit drivers or running Wizard-of-Oz tests over WhatsApp.
  3. A major focus is on building growth models, improving retention through pre‑conversion steps, and setting up analytics that generate true insights rather than vanity metrics.
  4. Crystal also covers how to structure and hire growth teams, and she highlights her nonprofit Generation Girl, which helps young women in Indonesia pursue STEM.

IDEAS WORTH REMEMBERING

5 ideas

Do scrappy, non-scalable experiments to validate ideas before building.

Crystal describes tactics like renting a stadium to onboard 60,000 drivers, running subscription tests via WhatsApp and interns, and faking in-app flows with screenshots—using Wizard-of-Oz setups to prove demand before investing in engineering.

Even with tiny user bases, you should still run experiments.

She argues that 30 users are vastly better than zero for learning; trends usually stay consistent as you scale, only their precision changes, so early-stage teams should still formulate hypotheses and test them rigorously.

Retention problems are usually solved one step before conversion, not at retention itself.

Instead of vaguely “improving retention,” she focuses on the specific preceding friction (e.g., trust in a new restaurant, confusing flows, slow search) and removes that barrier, which then naturally lifts retention.

Clarify your growth ‘physics’ and pull existing levers before inventing new ones.

She models the market, product, business model, and channels (including offline levers like visible drivers) to find underused assets—such as turning drivers into salespeople for Gojek’s e-wallet—rather than overhauling everything at once.

Most analytics efforts fail because they track KPIs, not user journeys and context.

Teams often log bare events with no properties and end up with dashboards that are “interesting” but not actionable; instead, events must carry rich contextual properties so you can answer “why” behaviors occur and change real-world decisions.

WORDS WORTH SAVING

5 quotes

The only thing better than having data on what your customers are doing is having data that you can actually act on.

— Crystal Widjaja

If you don’t have a tested hypothesis and can’t think of a way to run an experiment, that idea is pretty useless.

— Crystal Widjaja

Do not treat metric gathering as entertainment. Real news is information that changes what you do in the real world.

— Crystal Widjaja

We’re not wizards—you can’t change the physics of your universe. Start with what already works and pull the right levers there.

— Crystal Widjaja

Listeners, please do instrumentation correctly. Please don’t track your KPIs. Please track your user journeys and experiences.

— Crystal Widjaja

Crystal Widjaja’s unconventional career path and Southeast Asia startup contextSuper apps, market physics, and why they thrive in Asia but not the U.S.Scrappy, non-scalable growth tactics and early-stage experimentationRetention, growth models, and identifying the right levers and constraintsAnalytics done right: instrumentation, insights vs. measurements, and toolsHow to structure and hire effective growth teamsGeneration Girl and increasing women’s participation in STEM

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome