A Leaked Memo from Google CEO Sundar Pichai Comes Amidst Employee Discontent | Hacker News Reader

A Leaked Memo from Google CEO Sundar Pichai Comes Amidst Employee Discontent

inc.com

60 pointstlogan81 comments

It's kind of astounding just how absolutely useless top-level communications are in publicly traded companies. My company is far smaller and less important than Google, but I can still recognize the exact same brand of vacuous crap being passed off as leadership. How is it that as a people we've come to just accept this as table stakes?

Those 7 goals that he gave the company are... nothing. It's so painfully inclusive that there's nothing left to not be a goal. "Build the most helpful personal computing platforms and devices"? What the fuck is that? What activity could possibly be ruled as conclusively in or out of conformance with that goal?

It reminds me of the brain-dead training modules the whole world is required to click-through as fast as possible. Everybody knows it's pointless, nobody complains. I'm left wondering who the hell is driving all of this corporate theater? What is this reality distortion field that makes us feel like it is an inevitable fact of life that these leaders can only communicate in banal, unactionable platitudes? How is it that the entire corporate world has agreed that the liability of slipping up or saying something too opinionated outweighs _actually communicating_ with the damn employees?

> It's so painfully inclusive that there's nothing left to not be a goal.

To be fair... it's Google.

There's nothing computing-related that isn't Google.

ISP, cloud, web, personal hardware, IT security, OS, autonomous vehicles, social media, thermostats, mapping...

There is plenty of computing related stuff that Google doesn't do, has stopped doing or outsources nearly all of, whether it be ceasing to be a Domain Registrar, providing voice & texting services (Google Voice being just another Bandwidth.com frontend, Google Fi being an alternative onramp to the T-Mobile network, with identical roaming).

You mentioned social media, what social media happens on Google's platforms? Google Wave is long dead, Google Talk was broken so badly everyone I know stopped using it. Google Groups is a wasteland that just spams up Usenet, so everyone ignores it. Where is Google in action when it comes to social media?

> what social media happens on Google's platforms?

youtube/youtube shorts have a lot "social media" going on

Youtube is currently full of attention-bait, rage-bait, spam, scams, and AI noise.

It takes work over months of time to find youtube channels with value and a few more months to vet their worth.

While it IS a social media network, the majority of user content on the website is low-value spam comments.

> Youtube is currently full of attention-bait, rage-bait, spam, scams, and AI noise.

there is a chance youtube ML gives you recommendations based on your interests.. I find a lot of good quality content on youtube.

The problem is that one wrong video leads to a cascade of bad suggestions.

Opening one video that looks legit but is really AI noise leads to your recommendations filling with other AI noise.

Opening one video with a completely misrepresented title leads to your recommendations filling with attention-bait, rage-bait, etc...

If I would be youtube ML engineer, I would use some engagement metric as a label in loss function: for example how many % of video you actually watched, not just fact that you clicked on it.

> what social media

Little thing called "YouTube"

"Plenty of computing related stuff that Google doesn't do, for example all of this stuff that Google does or is involved in"

I work in a very large organization with 50,000 people. It's painful and discouraging listening to the person at the top, or their close deputies, because the documents they publish for mission goals are so abstract and boundless. Nothing I want ever seems directly addressed in these documents that get passed to all of us.

However, I've learned that the middle managers below C suite are looking at these objectives and justifying their own plans in alignment with this. Whatever the people who actually "do" want to do in 2025, must be in support of this abstract crap that frustrates me.

I've had more success in the last 2 years aligning my wants and interests with those of my middle managing-leadership, as they orient toward the executives.

Executives are often trying to enable with vapid bullshit, and really it just shows they shouldn't talk to the people who "do". I think a better way of approaching the workforce and inspiring loyalty to the org, is to trust a deputy to read you into their particular program area so the high-level executive can come in and speak "to you" at an unexpected/encouraging moment. I've seen that done well, but not all hands messaging - ever.

(All blanket statements have exceptional circumstances where they should be disregarded.)

Could be worse. The people at the top could be spending their time planning organizational reshuffles.

Ask how I know.

what kind of goals would you set if you be Pichai?

Deploy Google Fiber everywhere?

I think this is something with low reward/risk ratio for google, given where they have moat and expertise.

This is high risk, actually. Google has peering agreements with all major ISPs... the same ones that dislike the existence of cheap, fast fiber.

Make 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.

HN 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.

>You probably don't have any hard data in your hands.

Okay? I bet you don't either.

I don't, but you are supposed to provide evidence based on https://en.wikipedia.org/wiki/Burden_of_proof_(philosophy)

I 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.

https://www.thecgo.org/research/tech-poll/

Looks like that poll checks specific topic: "trust in data handling", not trust in general. Also #2 is Ok position.

Hold my beer while I Google for it...

Well, 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?

You 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.

10-20 years ago it was always the same (before Amit Singhal was kicked out): precision +X% / recall +X% for Google search

now they have many more surfaces including actual search(lots of different panels)

The 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.

Those 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.

I'm left wondering who the hell is driving all of this corporate theater? What is this reality distortion field that makes us feel like it is an inevitable fact of life that these leaders can only communicate in banal, unactionable platitudes?

At risk of shilling a bit, Whatifalthist posted an interesting video today "Understanding Modern Civilization" [1]. It's a longer critique of modernity, but I particularly liked the discussion of why the managerial elite came to power. In modernity, the lack of life or death pressure, social mobility, plus the gameability of metrics / systems used to determine who has power (ie. ivy league and big 4 consultancy pedigree vs inherited power back in the day) has made our leadership subject to the same foibles that affected Chinese mandarins historically: corruption and a focus on appearances over the details of actual execution.

So yeah, Sundar is one of our modern mandarins.

[1] https://www.youtube.com/watch?v=-SjMVJrmhh8

Interesting take, I'll watch the video later.

One important note is that there's a big difference between our culture and historical Chinese culture - and basically all powerful cultures that have existed before now. That is, the flexibility and rapid change of the system itself. It's not just upward mobility of people, it's upward mobility of companies and rapid systemic change, driven by technology and cultural shifts. Think about the difference between our society, and the 80s, the 50s, the 30s. There's never been a period like this before, so making comparisons to previous cultures is of limited utility and that needs to be noted when making the comparisons.

We might think of Google as a never-ending behemoth with incredible power because in our recent lifetimes it's been that way. But the reality is that the company has only existed for ~26 years and if it truly slips up it could be a shadow of itself in another five years. Something else would jump into the power vacuum and our society would barely blink, except for a million articles with titles like "What happened to Google?" being written.

I believe bureaucratic structures at scale are the same irrespective of cultures.

>https://www.youtube.com/watch?v=-SjMVJrmhh8

I've watched the whole thing with great attention.

There's some thoughts about "modernism" drawn from other books that the author cites that are worth exploring if you have an interest in history/anthropology. Somewhere in the middle the author's own theory starts protruding through.

I'll save everyone the 66 minutes: at the end of the video the author claims the CIA conclusively proved you can travel to the spirit world. This is really all you need to know.

I would rate this as a crock of shit, go read the books if you're interested in the topic, but skip the video.

For those downvoting, the video really does cite the CIA’s 1980s research into astral projection and the “Gateway Experience”, asserting it is “all correct”.

https://www.vice.com/en/article/7k9qag/how-to-escape-the-con...

The current ceo is cut from the same cloth of generic Wall Street b-school flunkies. What do you expect?

Their pay is often determined based on the performance of the stonk. An easy way to pump their $$$ and meet whatever OKRs are required to satisfy bonuses, is to layoff staff. Use vague excuses (ie, blaming remote work) for poor performance. Rinse and repeat as needed every quarter until they can sustainably compete or until they are 86’d.

> but I can still recognize the exact same brand of vacuous crap being passed off as leadership

I googled a bit and found he has an MBA from the Wharton School of the University of Pennsylvania. So that's probably the answer.

I agree with your general idea. This particular example looks mixed to me.

> 1. Deliver the world's most advanced, safe, and responsible AI.

For a company with Google's resources, this could be a good mission.

> 2. Improve knowledge, learning, creativity, and productivity.

What?

> 3. Build the most helpful personal computing platforms and devices.

This could be a good mission, or a disposable platitude.

> 4. Enable organizations and developers to innovate on Google Cloud.

If they want to keep a cloud business, I guess they have to mention it in the goals, though this one sounds like they haven't given it any thought, and just cribbed some disposable copy out of sales materials at the last minute.

Maybe it's alluding to some innovation-enabling AI technology push (like focusing on meeting the demand for some kinds of AI compute capacity, through proprietary hardware and software)? Why not say that?

> 5. Provide the world's most trusted products and platforms.

If they wanted to make this a cornerstone, I'd say they have to be more specific. Does it mean making things work better? Respecting people's privacy and security in some way other than what they currently say they do? Start focusing on products and services in which canceling isn't an option, so people can depend on them?

> 6. Build a Google that's extraordinary for Googlers and the world.

Needs elaboration for what this means, and how they're going to do it.

> 7. Improve company velocity, efficiency, and productivity, and deliver durable cost savings.

They sure to have a lot of goals that would be tricky to solve for simultaneously, if these were all real top-level-ish goals.

More interesting would be if they said (off-the-cuff) "Our sole top-level goal is to deliver the world's most advanced, safe, and responsible AI. We actually really, really mean it. To that end, adtech is now a subgoal in support of short-term funding the AI goal, adtech will disappear in 4-8 years, and until then will be staffed at minimal sustaining level. Information services and devices are solely to provide data and user access points for the AI work. Cloud services are refocused on AI-based solutions. We don't need 70% of our current employees, and we need to replace some of that freed headcount with different kinds of thinking. And we need to get the remaining and new people to be motivated by the mission, despite 20 years of hiring and incentives that would naturally lead most to focus on career rather than mission."

Probably neither they nor any other huge company could pull off a serious goal focus like that, and they'd basically throw away a cash cow machine that could've been milked much longer. But it would still be interesting.

It's really an exercise in what it might mean to take stated corporate goals very, very seriously, rather than the default, which is a behemoth lumbering on pretty much as before, while some costs get cut, and people spin their metrics and reports a little differently.

Or try "Our sole top-level goal is to maintain adtech dominance and revenue, through..." That's not as big a mission change, and maybe you can get by with just nudging huge hierarchies of career-directed people with broad platitudes and KPIs&OKRs.

>More interesting would be if they said (off-the-cuff) "Our sole top-level goal is to deliver the world's most advanced, safe, and responsible AI. We actually really, really mean it. To that end, adtech is now a subgoal in support of short-term funding the AI goal, adtech will disappear in 4-8 years, and until then will be staffed at minimal sustaining level. Information services and devices are solely to provide data and user access points for the AI work. Cloud services are refocused on AI-based solutions. We don't need 70% of our current employees, and we need to replace some of that freed headcount with different kinds of thinking. And we need to get the remaining and new people to be motivated by the mission, despite 20 years of hiring and incentives that would naturally lead most to focus on career rather than mission."

>Probably neither they nor any other huge company could pull off a serious goal focus like that, and they'd basically throw away a cash cow machine that could've been milked much longer. But it would still be interesting.

Yeah, that's interesting, I had almost the same exact sequence of thoughts myself. If _I_ were running google, I'd probably be tempted to bet the farm on the AI aspect, just as you say. But at Google scale, there's an almost crippling responsibility to not kill the cash cow, since you're talking about throwing away hundreds of billions of dollars. If we're being honest, that would probably be a unilaterally bad idea for the health of the company.

Still, Google is so disassociated and somnambulant right now. There's got to be a lot of daylight between "complete, laser focus on AI" and what we're getting now. Evidently the current goal is "just keep durdling with whatever we seem to already be doing".

I don’t think those company goals should be looked at face value. What is interesting to watch is the dynamics (what was there last year and isn’t there anymore), read between the lines (what is not mentioned as a goal) and the ordering (what is left at the end, what is more important than what).

You can’t expect such an exercise to be treated with transparency (#5. Lay off 20% more of those guys) as there’s just too much to lose by doing so. On the other hand, those goals are subtle but clear enough for the top managers of the company to inspire themselves and define their own - more concrete - yearly ambitions.

And the lower you go in the org, the more concrete the goals (assuming there are any at all). No yearly goal is worse to me than unclear ones, as it gives red flags of management not giving a shit and cruising.

>Those 7 goals that he gave the company are... nothing

Goal one is something - world's most advanced AI. There's an actual competition there with OpenAI which seems ahead a little.

The rest, yeah... it's all let's keep doing what we were doing.

I don't get what's so hard to understand here. Vision and mission statements are sanity checks for employees to see whether they help move the company towards or away from the desired direction.

For example, if a company aims to develop and deliver affordable and innovative healthcare solutions that enhance the well-being of individuals worldwide, an employee may realize that whatever they're working is too expensive and therefore inaccessible to many people around the globe.

You may correctly point out that a concrete goal for how much a given solution can cost is preferable, but the statements cover that as well: Management derives operational goals from these statements. Since healthcare solutions must be affordable, management can commission research on what the average person can afford in a given field, and use that research to inform how much a solution can cost.

Values dictate the morale boundaries: Sure, our goal is to to develop and deliver affordable and innovative healthcare solutions that enhance the well-being of individuals worldwide, but don't use child labor to achieve that.

I find the whole thing fascinating because it just shows there are one of two things happening:

1. The CEO is being genuine but is completely powerless to drive the direction of the company at google.

2. The CEO sends memos he knows are absolute hogwash because he tells his direct reports something completely contradictory to the memo in question.

If he's being genuine, even as a plebe from the outside I can see that advertising drives everything the company does. Responsible AI? Sure, as long as we can mine data to drive more ads (not responsible). Most helpful personal devices? Sure, as long as we sell ads while "helping" (aka not helpful at all). Innovate Google Cloud? I guess that's a nice side-project as long as it is cost-positive and helps us expand the cloud we run our ads platform on. Build the world's most trusted products? Sounds like we just need to convince people to trust us so we can sell them more ads. Etc. Etc. Etc.

There's one thing that the article didn't mention about 6. "Build a Google that's extraordinary for Googlers and the world" Each of the OKRs are assigned to an exec, they assigned this one to the CFO. So basically the person who's responsible for keeping costs down gets the control over how they treat the employees. You don't have to guess how it goes. All the employees I've talked to have said how Ruthless the cuts have been over the years and how it's really created a lot of employee discontent.

I see the capitalized "R" in that last sentence :)

What I find ironic is, Sundar himself will almost certainly be canned and replaced by her within the next few years.

Meanwhile, Google will continue to be torn "asundar"...

The memo outlines Google's roadmap for 2024.

    Deliver the world’s most advanced, safe, and responsible Al. 
    Improve knowledge, learning, creativity, and productivity. 
    Build the most helpful personal computing platforms and devices. 
    Enable organizations and developers to innovate on Google Cloud. 
    Provide the world’s most trusted products and platforms. 
    Build a Google that’s extraordinary for Googlers and the world. 
    Improve company velocity, efficiency, and productivity, and deliver durable cost savings.

Also, expect more layoffs throughout the year.

Reported originally at: https://www.androidpolice.com/googles-goals-for-2024/

Almost entirely meaningless corporate doublespeak. Shocking advertising isn't in there anywhere given that's how Google actually makes money.

1) Beat MS and FB at AI

2) Content-free

3) Content-free

4) We should probably make GCP better?

5) Content-free

6) Content-free

7) Work harder, more layoffs are coming.

0) Maximize Shareholder Value

that will be $200 million please :)

it reads like complete corporate garbage but I would assume these are created with

1. a rough organization-level investment prioritization

2. ask the leaders of those organizations to come up with vague mission statements

3. put them together and you get nonsense.

my guess would be the interpreted version is something like "Sundar says this is what we are focusing on"

1. Keeping up with recent AI advancements

2. Making youtube, search, gmail, and drive/docs/calendar better

3. Improving android and related devices

4. Expanding google cloud

5. I have no idea

6. Built a better employee experience

7. Improve devex and reduce operational costs

the only really interesting things I see are

1. AI is at the top

2. monetization is not mentioned

3. efficiency/cost savings are directly mentioned

would be a little concerned if I were a googler. basically AI + existing stuff + reducing costs / operational efficiency

(realizing I made alot of numbered lists here)

Those goals lol. We truly have reached some peak levels of idiocy with the ruling class of MBA executives. When it comes down to it they truly have nothing to offer. They extract wealth and deliver to shareholders, end of story. When real problems need solutions they turn into wet rags

Yep and the same in higher education too. New presidents still the most generic goals that don’t really change anything for us because the goals never said anything specific or different anyway. So frustrating these people make all this money and most are just stupid and lucky.

I’m not even kidding when I say ChatGPT could produce a better list of goals for Google.

That's because they used Bard to write it!

Could be! I tried asking ChatGPT to rewrite them but it came out even worse.

I evaluate the quality of a goal using this criterion: if its opposite is nonsensical or foolish, it's not worthwhile. For example:

"Deliver the world's most advanced, safe, and responsible AI."

This is not a very aspiring goal that will distinguish you from your competition because why would anyone want to deliver the world's most outdated, unsafe and irresponsible AI?

The executive equivalent of a rice cake.

1) Make Google Search Great Again

Cut everything else.

I would like to see "Don't be evil" come back. But I guess that ship has sailed so far away that no one at Google even knows what it means anymore.

Build better AI so that your job is automated and there can be even fewer employees.

Careful with leaking memos, sometimes what is sent is not identical copies. Each person could get a slightly different version that singles out who leaked it.

The jobs of the few hundred data scientists that actually work on improving AI at Google/DeepMind are very secure. But in a few years (as LLMs continue improving and people figure how to use them effectively) a significant part of the other employees will very likely become redundant.

It’s not clear what Google is really about any more.

Wins:

YouTube

Maps

Chromium

Apart from that stuff, everything else seems to be cancelled or pending cancellation.

Google makes so much money from search there seems no point in doing anything else.

You forgot Android.

Is Android a win for Google? Does it make any money, lock in anyone to their proprietary hardware? No.

Android makes money because Google Play is effectively mandatory, even if it isn’t absolutely so.

That’s a huge staying power in multiple areas that let’s them keep a lead

Google Play store isn't mandatory in most places where Android is the most popular OS.

You forgot Ads.

(Fixed that for you…)

It's about ads

Chromium isn't a win per se, Chrome is. Chrome users matter, Chromium users don't really do anything meaningful for them.

A nitpick if I ever saw one.

I didn't mean to nitpick. I couldn't tell from the comment if OP was referring to the technical achievements of the Chromium project or was referring to the end-user application (Chrome).

The technical achievements don't really matter as much as the value of having whatever 70/80% market share Chrome has. I'm not saying it's not impressive technically, just that they could've skinned Firefox and called it Chrome and put it on their google.com homepage and gotten a (multi)billion users as well in theory. As long as they could use it for the purposes of driving revenue, that would be a win as well, again in theory.

Same argument could be made for Android Open Source Project (AOSP) vs Android.

Google would not be revoking API keys for Chromium to nerf its functionality if this were not a real issue: https://www.reddit.com/r/linux/comments/mo60q5/psa_you_can_e...

I disagree. Chromium reinforces their position on the web.

They get to push whatever tech they want as a de facto standard because now the web is Chromium or WebKit (and maybe Gecko)

But by having Edge and all the other chromium variants, including electron, they get to define what web technology is.

30% on every transaction on the play store.

Maps is losing it's appeal as it reduces its readability and fills up with ads.

See:

- https://www.fastcompany.com/90930810/google-maps-eyesore-5-e...

- https://twitter.com/elizlaraki/status/1727351922254852182

> Build a Google that's extraordinary for Googlers, the world, and Googlers returning to the world.

There I fixed it.

Buried on page 5. We need to re-learn to click the upvote button, not just comment. (I've also noticed this in the comment replies recently, in which very often someone will reply while seeming to agree or be interested, but seem not to upvote.)

> 144. A Leaked Memo from Google CEO Sundar Pichai Comes Amidst Employee Discontent (inc.com) 46 points by tlogan 3 hours ago | flag | hide | 54 comments

Honestly wish I could downvote this post. So there’s that.

Hm. Hitting an HN post with lots of comment noise (without upvotes) early could be a good way to effectively downvote it off the front page before a lot of people see it. Maybe that also interferes with legitimate comments resonating and giving the post legs from upvotes.

"Provide the world's most trusted products and platforms."

Here's an easy one for them: buy Google Domains back from Squarespace again.

And maybe start considering the impact those kind of decisions have on being "trusted".

the world in paper cup held by fools dancing our way to a global warming ending Bravo Google for leading the way forward

Reply on news.ycombinator.com