Remote Site Growth Is Not A Conquest Game
...not exactly...
First published March 2017. I was addressing the problem of growing remote development sites (this is when work from home wasn’t common). You open a new engineering site in, say, London. How do you structure your projects? The same dynamics apply when building a single software product, though.
Way back when, before there were interwebs or MacBooks, we used to load all of our Macs into the back of our Volvo station wagon, pick up a whole rum cake (still warm) at the Boulder Creek Deli, and drive up to my cousin Herb Derby’s house on a Saturday. There we would wire up a temporary LAN and start playing Spaceward Ho!
Ho! was a game of exploration and conquest set in space [ed: what would now be called a 4X game]. It was full of tradeoffs: invest in better technology or more colonization, invest in better range or better firepower, long-term or short-term? Each tradeoff created risks and opportunities, depending on what the other players chose.
These game evenings went on far into the night (and sometimes the morning). At the end there was one gloating winner and a bunch of losers, but all full of cake and eager for the next battle.
I played many hours of Ho! against the fiendishly clever computer opponent, trying to refine my game play so that when the time came I could earn the coveted Last Slice.
In the end I found two reasonable strategies. I was reminded of Ho! when discussing growth strategies recently.
Thank you to my sponsor Tuple.
Remote pairing for fun & profit
Collaboration enhances focus. But what if you get distracted on the way to collaborating? Tuple's 2 engineering pillars are:
Make it easy to start a session
Make pairing fun
Screen sharing, fun pairing in 2 clicks with Tuple. Use code KB1 to get 33% off your team's first year.
The Problem
One tradeoff facing each remote site: concentrate on a few (or even just one) project or to have several projects? Succession plays a part as well. When you have one successful project, how quickly do you start new projects? How do you staff them? New hires, average folks from the successful project, the best folks from the successful project?
Once a site is well established the question is moot. There are always going to be a variety of projects. Early growth, though is tricky. Diversify too soon and you risk crippling the foundational projects. Diversify too late and you slow the growth of the site.
The Strategies
Ho! contains an analogous tradeoff. Do you grow conservatively? Do you wait for each colony to become profitable before starting the next one? This lets you maintain your investment in technology for the inevitable clash with enemies. Or do you grow quickly, expecting that the eventual revenue from many colonies funding rapid technology development will make up for lack of early technology investment?
The Analogy
A development site and a game of Spaceward Ho! are not perfect analogies. Site building is not a zero sum game. The rest of the company is not an opponent (at least needn’t be an opponent). The part of the analogy that spoke to me, though, is that the longer you are willing to wait for the big payoff, the bigger the payoff in the end. The proviso is that you can’t wait so long for a payoff that you get shut down. (Mumble mumble Kelly Criteria mumble mumble.)
Applying this insight, I would err on the side of aggressive growth in number of projects, even at the expense of some short-term progress on existing successful projects. I would err on the side of encouraging the best engineers to seed those new projects. It’s not like the new people who come in are idiots. They are going to quickly ramp up and provide their own perspective.
I would hedge my bets on new projects by applying explore/expand/extract to keep new projects small, diverse, and rapidly iterating based on concrete feedback.
tl;dr early in the life of a development site, grow the portfolio of projects as fast as possible consistent with giving each project a chance of success
I think I was employee #58 at Google Cambridge, at a moment where I think overall Google was still under 10,000. Our site director (miss you, Steve!) pursued a strategy of staffing out many small, local projects that were spread throughout as many different Google business units as possible. It worked exceptionally well, in my memory. Not only was the site de-risked from company pivots, in many cases, we were able to get things done much more quickly than the home site in Mountain View, because even though we hailed from the furthest corners of the organization, we all knew each other by name.
At least once a week, someone in Mountain View would get on a VC with me and complain that they were having problems negotiating with some other business unit also headquartered in Mountain View. At the end of the call, I would just walk over to someone from that unit I'd played ping-pong with the previous day, ask how her kids were, and say if they'd heard of the problem. Often times, the best communication link between two people at headquarters was a 6,000 mile roundtrip to the East Coast.
So I think that diverse remote sites can be actually superpowers for the entire organization, if managed well.
I really liked your quote "The longer you are willing to wait for the big payoff, the bigger the payoff in the end. The proviso is that you can’t wait so long for a payoff that you get shut down".
So I ran it through ChatGPT as an experiment trying to think of people with different perspectives. Thought the results were worth sharing!
Professional POV:
Strategic patience leads to greater long-term rewards; however, it is essential to balance vision with momentum—waiting too long risks missing critical opportunities or project viability.
- maintains the essence of the message while making it sound more structured and aligned with corporate communication
Professional but friendly:
Big rewards come with patience, but timing is key—wait too long, and you risk losing the opportunity altogether.
How might Elon Musk say it?:
The longer you can hold out, the bigger the reward—but wait too long, and you’re dead.
- It’s direct, bold, and in line with Musk’s style of communication
How might Steve Jobs say it?:
Great things take time, but if you wait too long, the moment passes—and so does the opportunity.
- It captures his visionary yet pragmatic approach
How might Albert Einstein say it?:
Patience can lead to great rewards, but time is relative—wait too long, and the opportunity may cease to exist.
- This keeps his scientific perspective while maintaining the essence of the message
How might Simon Sinek say it?:
The best rewards come to those who play the long game. But if you wait too long without action, the game might end before you win.
- This aligns with his leadership and purpose-driven philosophy