Steffen Wachenfeld

@StWachenfeld

Co-Founder & Managing Director @savedo_de @FinLeap @HitFoxGroup. Formerly @BCG @Penn @Wharton @WWU_Muenster. Interest: #bigdata #digitalization #underwaterrugby

Berlin, Germany

Joined October 2015

29 Following

50 Followers

10 Posts

StWachenfeld retweeted

Wes

@wmorrill3

almost 2 years ago

Have you heard of Conway's Law? It's an interesting observation about the root cause of why large organizations usually make products worse. In 1967 Melvin Conway wrote "Organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations." When you think about any product carefully, if you see an obvious lack of integration it's usually due to an organizational social boundary. If you are working on a product you should consider if organizational boundaries are getting in the way of optimal design. A well known example - early days of Tesla there was a battery team and separately a vehicle structures team. Structures team designed their vehicle body to meet given requirements of strength, crashworthyness, torsional stiffness, etc. likewise the battery team designed their part to be self contained, it could survive durability, accidentally being dropped, being hit in a crash, etc. As a result, we ended up with was a super dense battery in a strong box like structure, which was then Installed into the vehicle which had a nice space for it to mate into. There were no issues with integration, everything fit together perfectly and met all product goals. It achieved one of the highest crash safety ratings measured at the time. But we had a box full of battery cells that was installed into another empty box shaped receptacle on the body. A box in a box. When you simplify it down that far it sounds obviously wrong. The two organizations had achieved their goals, worked together without friction, and the product met its overall goals. Yet the product ended up with a clear lack of optimization as a result of the organizational boundaries of the two teams working in isolation. Nothing was wrong, but it wasn't optimal. So before the next product was designed, the battery team gave responsibility of the battery structures also to the vehicle structures team. On this iteration, we ended up with the structural battery, which is an integral part of the body and crash structure. Without it, the vehicle body will not work. It's the literal floor for the vehicle. But the redundancy is gone and the design is more efficient as a result. This vehicle also achieved one of the highest crash safety ratings measured at the time. This is a super obvious example (in retrospect) and solved with a fairly large organizational change but you can also see this happen in small technical decisions and doesn't require structural change to fix. Someone just needs to question if there is a better solution in a team open to criticism. This mindset to work together to make the best product regardless of ego is where you end up with the most innovative products. Some smaller examples have been seen when inspecting Cybertruck design. The chassis air suspension which is used to pressurize the battery pack to prevent water ingress. The subwoofer which utilizes the air volume of the body side instead of making the enclosure larger. Centralized zonal vehicle controllers instead of many small distributed controllers. Doors which use the exterior surface as a crash intrusion beam. The pedestrian warning system used as a horn. The list goes on. The excitement and motivation by everyone involved to work across boundaries and actively break down Conway's Law is one of the many reasons I love working at Tesla.

wmorrill3's tweet photo. Have you heard of Conway's Law? It's an interesting observation about the root cause of why large organizations usually make products worse.

In 1967 Melvin Conway wrote "Organizations which design systems are constrained to produce designs which are copies of the communication structures of these organizations."

When you think about any product carefully, if you see an obvious lack of integration it's usually due to an organizational social boundary. If you are working on a product you should consider if organizational boundaries are getting in the way of optimal design.

A well known example - early days of Tesla there was a battery team and separately a vehicle structures team. Structures team designed their vehicle body to meet given requirements of strength, crashworthyness, torsional stiffness, etc. likewise the battery team designed their part to be self contained, it could survive durability, accidentally being dropped, being hit in a crash, etc. As a result, we ended up with was a super dense battery in a strong box like structure, which was then Installed into the vehicle which had a nice space for it to mate into. There were no issues with integration, everything fit together perfectly and met all product goals. It achieved one of the highest crash safety ratings measured at the time.

But we had a box full of battery cells that was installed into another empty box shaped receptacle on the body. A box in a box. When you simplify it down that far it sounds obviously wrong. The two organizations had achieved their goals, worked together without friction, and the product met its overall goals. Yet the product ended up with a clear lack of optimization as a result of the organizational boundaries of the two teams working in isolation. Nothing was wrong, but it wasn't optimal.

So before the next product was designed, the battery team gave responsibility of the battery structures also to the vehicle structures team. On this iteration, we ended up with the structural battery, which is an integral part of the body and crash structure. Without it, the vehicle body will not work. It's the literal floor for the vehicle. But the redundancy is gone and the design is more efficient as a result. This vehicle also achieved one of the highest crash safety ratings measured at the time.

This is a super obvious example (in retrospect) and solved with a fairly large organizational change but you can also see this happen in small technical decisions and doesn't require structural change to fix. Someone just needs to question if there is a better solution in a team open to criticism. This mindset to work together to make the best product regardless of ego is where you end up with the most innovative products.

Some smaller examples have been seen when inspecting Cybertruck design. The chassis air suspension which is used to pressurize the battery pack to prevent water ingress. The subwoofer which utilizes the air volume of the body side instead of making the enclosure larger. Centralized zonal vehicle controllers instead of many small distributed controllers. Doors which use the exterior surface as a crash intrusion beam. The pedestrian warning system used as a horn. The list goes on. The excitement and motivation by everyone involved to work across boundaries and actively break down Conway's Law is one of the many reasons I love working at Tesla.

122

409

749

211K

Steffen Wachenfeld @StWachenfeld

almost 2 years ago

@vueling Vueling just left three passengers on the ground due to overbooking. One of us was sold his ticket 1h ago!! For an overbooked flight. We need to go to BCN from BER and ask if Vueling could book us onto EasyJet this afternoon? We hang out at BER airport now

236

StWachenfeld retweeted

Borriss

@_Borriss_

almost 3 years ago

Less than 31 hours since OpenAI started dropping the ChatGPT vision feature on pro users... People are scratching their heads in disbelief. 10 wild examples:

602

41K

24K

12M