Comment by mji | Hacker News Reader

Comment by mji | Hacker News Reader

Parent Full thread

Having worked at both, 20% time is the only part of the article that is somewhat off-base.

In particular the following string of paragraphs match exactly what I’ve seen at both:

> With all that up-front planning, the actual product decisions and working code were reviewed much less. Only the other PMs on my team reviewed my specs, and only me and the tester (also new grads usually) reviewed the working changes before they were added to the branch. For what it’s worth, I think this is why the quality of the details on Microsoft products is often lacking—a random PM made a decision and no one bothered to push back on it.

> Contrast that with Google.

> At Google, I never saw a strategy document. I went to every Friday all-hands, but I didn’t hear the leaders talk about a broad vision of what we needed to build. The people above me didn’t set direction and ask me to follow it.

> Instead, leaders encouraged teams to generate their own ideas. The founders had ideas of their top priorities and worked with the relevant teams on those, but the list of the top company OKRs was a subset of all the OKRs, not a roll-up. Some people were not working on anything the company cared about strategically!

Again, I can only speak for Cloud. I find product changes are reviewed by a dozen people at least and take months to finalize. It can be slow, but usually we're sure why we're making the decisions, and where risks lie. A large weight is given to customers asking for features, so they're in the loop.

It's absurd that there's no Google strategy documents. There's company-wide and division-wide OKRs that get major events to publicize them, with large Q&As and all. It's easy to learn what the priorities are and why. But, usually you just pay attention to what your local group is working on. At a smaller level, every feature has a business rationale backed by customers requesting them or hard dollar figures.

I really don't know what's going on here. Do people think Google runs the same way as it did in its early years? It's a fairly mature company with a good track record of revenue growth, especially for such a massive company. These things don't happen haphazardly.

We seem to be looking at the same text and drawing wildly different conclusions from it.

When the blog post says "the actual product decisions and working code were reviewed much less. Only the other PMs on my team reviewed my specs" referring to Microsoft, to me that lines up with you saying "I find product changes are reviewed by a dozen people at least and take months to finalize." referring to Google. The distinction the author is drawing is that Microsoft doesn't review day-to-day engineering work like this, because fundamental product decisions are made at a higher level, while smaller design decisions might be made by individual PMs or engineers and make it into the product without ever being reviewed.

When the author says "The founders had ideas of their top priorities and worked with the relevant teams on those, but the list of the top company OKRs was a subset of all the OKRs, not a roll-up. Some people were not working on anything the company cared about strategically!" referring to Google, that sounds very aligned with you saying "There's company-wide and division-wide OKRs that get major events to publicize them, with large Q&As and all. It's easy to learn what the priorities are and why. But, usually you just pay attention to what your local group is working on."

Reply on news.ycombinator.com