Beyond the golden age: reflecting on 12 years at Google

📝 The golden age of Google as I’ll remember it; and what may lie beyond it

I joined Google in 2012, in the middle of what turned out to be its long golden age of post-IPO growth. The company was doing great, steadily expanding both its revenue and the engineering footprint.

In November 2012 I got to become part of that expansion. Fresh out of university, I moved to France to join Google at its recently inaugurated office in the 9th arrondissement of Paris. The sign at the office entrance read “I’m feeling lucky” and that’s exactly how I felt. A newly minted Software Engineer with a ticket to the chocolate factory.

The next 12 years were an exhilarating ride. I got to contribute to ambitious R&D projects, work with excellent colleagues, and run a programming competition as a side project. Among all that heads-down work, I didn’t spend much time thinking about the big picture: what was it that made Google work so well?

As I take a break from Google to pursue a new adventure, with this post I’ll try to fill that gap.

The golden age formula

So, what made Google successful?

First, there were the engineering strengths. Google Search and the underlying site ranking method were brilliant innovations. The company’s early engineering team pioneered the design of reliable distributed systems composed of unreliable off-the-shelf machines. The resulting services were remarkably performant and cost-effective.

Then, there was the business model. The attention of people who turn to Search with commercial intent is very valuable. Imagine marketing a vacuum cleaner. How much would you be willing to pay for an ad impression (single person seeing the ad once) for an ad on a city billboard, versus showing the ad directly to people who had just typed “best vacuum cleaner under 200 USD” into a search engine?

Google’s auction-based pricing model for keyword ads allowed the company to monetize these intents efficiently, automatically scaling ad prices to their commercial value. (Bids for “William Blake poetry readings” tend to be, sadly, cheaper than “Vacuum cleaner shops near me”.)

Finally, there was the continued growth of the digital advertising market. Throughout the 2010s, Google and other ad-supported tech companies were being carried by the headwinds of advertising budgets gradually shifting to the Internet.

Here's a relevant chart from Ben Evans (source):

That was the formula for Google's long golden age, as I understand it today: strong engineering + great business model + favorable market conditions. Google’s revenue kept growing 20% year-to-year pretty much all the way from the IPO in 2004 to the 2020 pandemic. What a run!

The perks of the golden age

Revenue solves all known problems.
- Eric Schmidt

When a company grows 20% year-to-year, all sorts of nice things follow.

Google of the golden age was investing across the board in ambitious, future-looking R&D. The company pulled off fantastic things: it built a new operating system from scratch (a project I got to work on and dear to my heart), experimented with cloud streaming of games, made cars that drive themselves.

To staff these expanding R&D projects, Google was hiring big. Hiring is not only important for the people currently being hired. It's important internally: the existing employees have a natural path of career growth, as they end up leading the junior colleagues. And with abundant job openings, people ready for a new challenge can easily find it within the company.

(When I was joining Google, the dominant narrative was that people are expected to change teams/jobs every 1.5 to 3 years. During placement interviews I told my future manager I can only commit to staying for 1.5 years. He said it’s fiiiine. He left 3 years later, I stayed for 9 more.)

My favorite perk of the golden age were ample travel budgets. For most of the time in the 2012-2020 Google, I would not hear the phrase “travel budget” used in a functional way. It was more of a figurative idea, we assumed it exists, we assumed it means we should be frugal within reason, but we never saw it do anything in practice.

In my peak travel time, I’d visit Mountain View every 3 months. As far as I could tell, it was a good use of the company's resources. In addition to the information exchange and relationship building, I would come back from each trip with renewed energy and excitement for my project. There’s something about a generous travel budget that communicates to satellite teams in smaller offices: your success is important to the company. We want you to have the tools needed to be effective, so travel as you need.

Or maybe it was just the California sun boosting my depleted vitamin D levels.

The tide goes out

Market downturns have the pesky tendency to eventually arrive.

Spending on digital advertising in the US is forecasted to settle into a period of moderate growth. From 2023 to 2028, growth will — for the first time — fall to single digits due to the maturity of the market.
- Forrester Research

The pandemic was an economic wildcard. Lockdowns initially supercharged digital advertising, as everyone took to online shopping out of necessity. Then, when this short-term burst receded, it exposed signs of a deeper slowdown.

I was asked to limit business travel in mid-2022.

A few months later, Ben Evans published this chart, subtitled "The end of 20% annual revenue growth?” (source):

The clearest mark of the turning tide were the historic layoffs of January 2023.

Google laid off 12k full-time employees, cutting across all units and functions. The process was quick and abrupt for people based in the US, and dragged out for many more months for those based in countries without at-will layoff provisions.

After the layoffs, the vibe in the company shifted to unprecedented levels of pessimism. People would say that Google lost its way, that we strayed from the engineering culture of the founders. My former colleague Ian wrote in his goodbye note: “Google's culture eroded. Morale is at an all-time low. If you talk to therapists in the bay area, they will tell you all their Google clients are unhappy with Google.”

Why so much discontent? I think we were all mourning the golden age. Loss aversion makes us hurt when we lose something of value. And we did: layoffs meant seeing valued colleagues being let go, and losing our own sense of job security. Reduced hiring meant little internal mobility; reduced travel budgets meant my team in Paris practically stopped travelling to California.

This change of circumstances hit all the harder that it followed such a long period of prosperity.

Beyond the golden age

What’s on the horizon? Nobody knows for sure, but I’m optimistic about Google’s future.

The 20% growth of the digital advertising market is (probably) not coming back, but the company has been working hard on diversifying the business. The public cloud is growing ~30% year-to-year and has a long way to go. Google is positioned to do great in the AI race, Waymo is leading the way to scalable autonomous driving. YouTube Premium is my personal favorite subscription on the Internet. Amazing what you can learn from YouTube videos.

The biggest challenge ahead may be cultural. How to overcome the latent mood of discontent? For better or for worse, regression to the mean is a powerful force. Just like the long golden age boosted the current morale crisis, these few years of choppy waters will help rebound the optimism when the fortunes inevitably turn for the better.

Harder times are galvanizing, and the grief for the first golden age won’t last forever.

Thank you to everyone I shared these 12 years with. Good luck to all of us 🍀!

Credits

Thank you to Amélie, Dragoș, Gabriel and Mesch for review and feedback 🙏.

Postcard from Guatapé

I’m taking some time to travel in South America before taking on the next challenge. I started with a week of Spanish classes in Medellín, Colombia. Everyone is very nice and patient with my attempts at Spanish 🥹.

Have a good week,
Przemek