Should a product organization optimize its shipping culture for quality or for speed?

Matteo Franceschetti · Nov 20, 2023

Validation timelines collapse if you lower the certainty requirement — 80% certainty in three days beats 100% certainty in a month

Following the Elon algorithm of simplifying and deleting requirements; once the team dropped the demand for full certainty they found a way to validate in days

36:09 20VC: The Ultimate Hiring Playbook: Five Questions to Ask Every New Hire | What Makes Truly Great Leaders and How They Give Feedback | Do VCs Really Add Value; Lessons from Hard Fundraises with Matteo Franceschetti, Co-Founder @ Eight Sleep

Matteo Franceschetti · Nov 20, 2023

The point of velocity is not speed for its own sake but learning as fast as possible so you can adjust as fast as possible

Because you can't know everything up front, the value of moving fast is compressing the learning loop; long rigid roadmaps defeat this

38:04 20VC: The Ultimate Hiring Playbook: Five Questions to Ask Every New Hire | What Makes Truly Great Leaders and How They Give Feedback | Do VCs Really Add Value; Lessons from Hard Fundraises with Matteo Franceschetti, Co-Founder @ Eight Sleep

Jeff Seibert · Nov 22, 2023

Speed of product cadence is critical, and companies fatally default to waiting for more data and statistically significant results

At Twitter stat-sig data always took two to three weeks plus analysis time, which is a disaster for progress

6:58 20VC: Why OpenAI Will Become an Infrastructure Play, Why Apple Will Win in an AI World, Why Google is the Most Vulnerable Incumbent, Will LLMs Be Commoditised, Which Startups Are Thin vs Thick Wrappers on Top of LLMs with Jeff Seibert, Founder @ Digits

Parker Conrad · Oct 3, 2022

There is no real trade-off between speed and quality; speed is more often aligned with quality than opposed to it

It's rare to see slow-moving projects nail it at the finish line — usually they're slow and also bad; urgency about fixing underlying issues is a virtue and being lackadaisical about them doesn't produce higher quality

Scope: acknowledges it is possible to do things fast and bad

11:01 20VC: Rippling's Parker Conrad on The Four Main Benefits From Building a Compound Startup | Why There Should Never Be a Trade-Off Between Speed and Quality | How Zenefits Gave Parker a Chip on the Shoulder and Why That is so Important?

Kevin Niparko · Sep 29, 2023

Speed and quality are not really a trade-off; shipping faster is a key unlock for higher quality

The faster you ship, the faster you learn all the ways you were wrong and how to make it right — reality, customers and procurement teams are unforgiving

6:49 20Product: Why Product Memes Are More Important Than a Product Roadmap, Why Writing is the Essential Skill for Product People, How AI Changes The Role of Product, Big Mistakes Founders Make When Hiring Product Teams with Kevin Niparko, VP Product @ Twilio

Harry Stebbings · Sep 29, 2023

Products that take a long time are rarely perfect because of the extra time; you learn more by shipping

7:26 20Product: Why Product Memes Are More Important Than a Product Roadmap, Why Writing is the Essential Skill for Product People, How AI Changes The Role of Product, Big Mistakes Founders Make When Hiring Product Teams with Kevin Niparko, VP Product @ Twilio

Julio Vasconcellos · Sep 23, 2022

Prefer's core mistake was iterating too fast and not giving each product version enough time; he should have listened more to Scott Belsky's craftsman approach

The move-fast-and-break-things mindset that worked at Facebook and Peixe Urbano landed them too far on the rough side of the spectrum, so tests never got a fair run

18:36 20VC: Why Greed is the #1 Enemy of Venture Returns, Why Not Enough VCs Play to Win and Lessons from Scaling to $100M and 1,200 Employees and Then Cratering with Julio Vasconcellos, Founder @ Atlantico

Matteo Franceschetti · Nov 20, 2023 · hedged

Moving fast on the subscription launch cost them: with more testing beforehand it could have been more sophisticated and launched a month earlier in better shape

Scope: single example

39:00 20VC: The Ultimate Hiring Playbook: Five Questions to Ask Every New Hire | What Makes Truly Great Leaders and How They Give Feedback | Do VCs Really Add Value; Lessons from Hard Fundraises with Matteo Franceschetti, Co-Founder @ Eight Sleep

Harry Stebbings · Oct 3, 2022

In the interviewer's own stated view, speed of execution is the only thing that matters — better to get something 80% done than 99% perfect

10:45 20VC: Rippling's Parker Conrad on The Four Main Benefits From Building a Compound Startup | Why There Should Never Be a Trade-Off Between Speed and Quality | How Zenefits Gave Parker a Chip on the Shoulder and Why That is so Important?

Matteo Franceschetti · Nov 20, 2023

Speed of execution is as important as everyone says, provided quality is not compromised — and he would take an 80% version today over a 90% version in a week

Quality means clearing a threshold, not perfection; ship at 85% and improve daily to 100%, rather than shipping something shitty that damages the brand

Scope: quality must clear a defined threshold; requires simplifying the requirements

33:33 20VC: The Ultimate Hiring Playbook: Five Questions to Ask Every New Hire | What Makes Truly Great Leaders and How They Give Feedback | Do VCs Really Add Value; Lessons from Hard Fundraises with Matteo Franceschetti, Co-Founder @ Eight Sleep

Julio Vasconcellos · Sep 23, 2022

He now weights product strategy roughly 60/40 toward scrappy iteration over craftsmanship, up from about 90/10, because great design and crafted experiences matter for finding the early nut of product-market fit

The Prefer experience raised his appreciation for the craft side that Scott Belsky always advocated

20:15 Craft and design investment early helps find pmf justifying a balanced speed craft mix

Matteo Franceschetti · Nov 20, 2023

Velocity has to be built deliberately into culture by naming and publicly celebrating fast work, with the goal of doing in half the time what any other company does

34:49 Deliberately celebrate fast work to build a velocity culture

Christian Kleinerman · Sep 22, 2023

The right balance between deliberation and fast iteration depends on the technology: core data-integrity subsystems need heavy up-front design, while UI surfaces should ship and iterate

Nobody wants corrupted data or wrong results from a badly built core subsystem, whereas for query-editor suggestions there's no right answer so you learn from users

36:58 Subsystem criticality determines deliberate design versus fast iteration

Christian Kleinerman · Sep 22, 2023

Shipping faster is not necessarily better when simplicity is what's being traded off; taking longer to make the automatic choice for customers is often the right call

You can ship faster by surfacing all the choices to users, but simplifying for the customer takes longer and is worth it

38:50 Prioritizing simplicity for customers justifies slower shipping

Jean-Michel Lemieux · Mar 20, 2024

The core cultural difference is that Shopify ships on quality while Atlassian ships on speed

Atlassian does many things at once — many products, lots of M&A — while at Shopify they spent far more time doing fewer things and Toby would stop shipments on quality grounds

9:22 Quality first and speed first are both viable company cultures

Jean-Michel Lemieux · Mar 20, 2024

Teams should not choose between speed and quality but split work monthly — polish roughly half the things to best-on-the-planet standard and rush the other half

That mix is what produces long-term momentum: the polished things are what customers rave about to friends, and the rest wouldn't be talked about anyway

13:25 Split work into a polished half and a rushed half monthly

Jean-Michel Lemieux · Mar 20, 2024

You get both speed and quality by time-capping work and deciding as a team what must be polished and what must not

Refusing to discuss things that don't matter and forcing either shipping this week or a time-capped direction debate has given his team a huge amount of speed

14:44 Time capping decisions achieves both speed and quality