I spent years supporting the best commute in America.

Almost nobody can use it.

I’ve lived in Oakland and worked in Silicon Valley for the past 15 years, contributing to commuter shuttle programs at some of the biggest names in tech. I know these services inside and out — how great they are, what they cost, and why you can’t use them.

I started CommuteSource to change that.

Being driven to work is life-changing

At Apple, I fell into a routine: work on the way in, nap on the way home. Either way, I arrived ready for what came next. My kids were young then, and I can’t imagine how my parenting would have suffered if I’d had to white-knuckle my way through rush-hour traffic on 880 twice a day. That free ride to work kept me sane and made keeping that job affordable.

Commuters who’ve convinced themselves that driving is their “down time” literally don’t know what they’re missing.

What I didn’t know until later was how much Apple was paying for my “free” ride.

The empty coach

Tech giants spend hundreds of millions of dollars a year shuttling their employees to and from work. They pay at least double what they need to, because they don’t play well with others.

A coach carrying one company’s badge holders can only ever be full in one direction. Every bus that arrives full of employees in the morning turns right around and heads back out to make another run — empty! In the evening it does the same thing in reverse. In an exclusive service, every ‘revenue’ trip requires a ‘deadhead’ trip, and the employer pays for both. That empty coach is stuck in the same traffic as everyone else traveling through that corridor.

Nothing separates this excess supply from existing demand but a badge and history. These programs were created inside each company’s facilities department to solve their own parking shortage, and not much has changed since. Service providers are paid by the hour, riders or not. Economies of scale are there to be realized; the incentives just aren’t aligned.

Transportation is more than a system

One way to view transportation is as a market, with supply and demand like any other. The circumstances we’ve inherited have left most of us providing our own supply at great expense, for the illusion of unrestricted mobility: you can go anywhere, any time, if you buy a car, store it, insure it, maintain it, fuel it, and drive it.

So you pay thousands of dollars a year for an asset that sits idle 95% of its life — and takes your full attention the other 5%. You pay with after-tax salary, from an employer who provides twice as much real estate for your car as for your desk. And you arrive exhausted and stressed. You can go anywhere you want, whenever you want; so can everybody else.

Trouble is, cars don’t scale. The average U.S. driver lost about 49 hours to congestion in 2025 — more than a full work week, and that’s the national average, not ours. A conservative estimate of 5-8% puts the land dedicated to car storage within the developed Bay Area at 150-240 square miles, or the equivalent of 2 to 3 San Franciscos. Collectively, inadvertently, we’ve chosen to store cars rather than house people.

None of which means public transit doesn’t matter. BART, Caltrain, Muni, and the ferries are genuinely vital to the region, and the more they thrive the better off we’ll all be. But there are 27 transit agencies in the Bay Area, many of them competing rather than coordinating. They’re stretched thin, facing yet another ‘fiscal cliff,’ and lack the resources to address problems that outgrew their antiquated service models decades ago. Heavily subsidized, yet wholly inadequate, public transit distorts the market — thwarting innovation, and keeping most of us in our cars.

I hate to say it, but that leaves the market as the best venue for a fix. Not as a competitor to public transit — as a partner in innovation.

And no, robo-taxis aren’t that fix. Roughly 40% of ride-hail miles are driven without a passenger; eliminating the driver doesn’t change that math.

The trip that doesn’t exist on any map

My last job was in Sunnyvale. Google Maps told me transit from Oakland would take over two hours. It was wrong by an hour in each direction — two hours a day, ten hours a week — because it didn’t know two things: my employer ran a shuttle between campus and the nearby train station; and I could reach the station near my house by driving, biking, or getting dropped off. Maps only has public transit schedules, and makes you pick one mode for directions. That’s fine for some trips, but useless for trips where getting out of a car would do the most good.

I only found that itinerary because I’m a transit nerd, and I knew the system from the inside. Running it took several apps and a standing coordination burden morning and night. Show most people a two-hour transit estimate against a fifty-minute drive and they’ll choose to drive every time. They never find out the better trip was there all along.

What we’re building

Conventional transit is designed around guesses — some better informed than others, but often historical and self-perpetuating — about who’s going from where to where, and when. CommuteSource is designed to fulfill actual demand, for the trips you’re already taking.

Here’s how it works — four complementary innovations:

• “Hub & bespoke” service. Flexible, demand-responsive shuttles feed BART, Caltrain, and high-capacity luxury coaches on the corridors that need them — solving once and for all the “first-mile/last-mile” problem that keeps so many people driving. Where demand doesn’t warrant a shuttle (yet!), app-enabled coordination across car- and vanpools — a pool of pools — covers the gap.

• Comprehensive multimodal trip planning. Public and commercial schedules in one place, including the services you’re personally eligible for, with integrated booking where required. An early version is live today.

• Pre-booking. Tell the network where you’re going before you go, and the ride costs less. Where advanced booking is available, Uber charges a premium; we’d rather discount it, because knowing tomorrow’s demand tonight is exactly what lets us put the right vehicle in the right place instead of guessing. A calendar integration, for added convenience, is coming soon.

• Sponsorship. Any employer, landlord, hospital, school, or venue can subsidize your ride, fully or partially — no contracts, no overhead, and no funding the entire program. Many sponsors, one network of services.

Here’s what that gets you:

• Affordable fares. Drop the exclusivity, and the cost per ride gets cut in half. Routing through hubs does the same for connections: the shuttle that just dropped you at the station picks up the people getting off the train, so the deadhead miles that make ride-hail so expensive — roughly 40% — get minimized. Book ahead and the price drops again. Fares are eligible for pre-tax transit benefits, per ride or by discounted monthly subscription — whether sponsorships cover none, some, or all of your fare.

• Safety in numbers. For a lot of riders, safety on mass transit is a real concern. Feeder shuttles deliver neighbors and coworkers in groups, so door-to-door, your fellow riders are familiar. If that’s not enough, our pool matching includes filters for gender preference.

• A guaranteed ride home. An actual service, not a reimbursement form. You won’t be stranded at the office, so you’re free to leave your car at home.

What we need from you

This works because you share where you’re planning to be. Not where you’ve been — your phone has been handing that over for years. Where you intend to be tomorrow morning at 8:15 is the one detail no transportation system has ever had.

We don’t ask for it lightly. A network that can see demand before it arrives can move people better than any system built so far — but only if whoever holds that data proves worth trusting, on the first day and every day after. We’d rather earn that slowly than take it cheaply, so we’ve built our system that way: the trip planner is free, open to anyone, and asks you for nothing. Only if you find a trip worth keeping and save it do we ask about your schedule, and your home address is generalized for privacy either way.

Try the trip planner. It’s the one I needed back when I was commuting to Sunnyvale.

Our promise is a better commute for all. Our goal is the end of rush hour.