4 ms·
what kind of goals would you set if you be Pichai?
by riku_iki 3y ago
what kind of goals would you set if you be Pichai?
- shmerl 3y agoDeploy Google Fiber everywhere?
- operatingthetan 3y agoMake the public trust Google more than any other FAANG again. People like Microsoft more than any of those companies now, which is a hilarious flip from how people felt about them in the late 90s early 2000s.
- riku_iki 3y agoHN crowd may be not trusting google, but I am not so sure about general public. You probably don't have any hard data in your hands.
- operatingthetan 3y ago>You probably don't have any hard data in your hands. Okay? I bet you don't either.
- riku_iki 3y agoI don't, but you are supposed to provide evidence based on https://en.wikipedia.org/wiki/Burden_of_proof_(philosophy) https://en.wikipedia.org/wiki/Burden_of_proof_(philosophy)
- operatingthetan 3y agoI didn't say something that goes against the status quo. I'm not interested in a pedantic tangent thanks. Just as we can see from OP, the goals don't need to be based on data, they are goals, not arguments.
- toomuchtodo 3y agohttps://www.thecgo.org/research/tech-poll/ https://www.thecgo.org/research/tech-poll/
- riku_iki 3y agoLooks like that poll checks specific topic: "trust in data handling", not trust in general. Also #2 is Ok position.
- readthenotes1 3y agoHold my beer while I Google for it...
- sebastos 3y agoWell, I'm no CEO of Google or anything, but if the company is so broad, then shouldn't he be giving an audacious but concrete goal to each of his segments separately? It's just supposed to be scoped for the year 2024, after all. What's the point of having your 2024 goal be "make the world a great place for Googlers"? What about that starts or ends in 2024? In his shoes, I'd feel pressure to devise specific high-level strategies that compliment the different segments and then throw down the gauntlet. So you know, you'd have high level strategies, like "AI all the things = synergy" and "Regain public trust on our bread and butter products" and then you'd break it down more: - Self-driving car people: Make me a giant dataset of multimodal data for the AI people - AI people: Show me a demo of Gemini driving the self-driving car people's car by end of year - Nest and assistant people: Real-time Gemini assistant And then over on the pure software side you'd be saying: - Youtube people: Expand premium membership by foo, reduce negative ad sentiment by bar - Google Drive people: Close gap with M$ Office features in GSuite - Search people: Radical crackdown on dark SEO patterns - Gchat people: pointlessly re-imagine the entire look and feel of the web app for the fifteenth time etc. etc. No?
- riku_iki 3y agoYou can read "most helpful personal platforms" as personal platforms(chrome, android, bard, search, drive) being more helpful than competitors. I guess defining specific benchmarks and metrics should be included into the goal. Then it looks very specific without 10 bullet points like in your proposal.
- xiphias2 3y ago10-20 years ago it was always the same (before Amit Singhal was kicked out): precision +X% / recall +X% for Google search
- riku_iki 3y agonow they have many more surfaces including actual search(lots of different panels)
- xiphias2 3y agoThe point is that OKRs are good only if anybody from the company can track if / how much they are contributing to the company goals. It only works if the key results have a clear metric.
- riku_iki 3y agoThose from Pichai are some global "goals" not "OKRs". Also, that's if you have some quality improvement you can use some metric. For launching new products you just have: launch new product, and then next year you can claim user growth, features adaptations, track engagement metrics, and this works for specific surface. For multiple complicated surfaces you likely can't come with single integrated metric. I even don't know how they could measure precision/recall for search, there are likely so many slices/skews/verticals/corner cases.