Amazon, let’s face the facts: You got it wrong on your HQ2 decision.

You turned a corporate expansion — usually a routine, boring business decision — into a 14-month-long North American beauty contest that brought glory to a staffing model that is a relic.

You pitted 238 cities and regions against each other, earning hundreds of millions in free publicity and garnered billions in state and local tax concessions.

Your executives in charge of finding HQ2 ultimately realized that “no one city had enough technical talent to accommodate the company’s projected growth,” as reported by The Wall Street Journal.

You decided to evenly split your second headquarters into 25,000 jobs each between New York’s Long Island City and Crystal City (sorry, “National Landing”), a neighborhood of Arlington in Northern Virginia, and also create another 5,000-person operations facility in Nashville.

But let’s be honest, when it comes to actually building a distributed team that will be effective for your business in the long term, you completely missed the mark.

Pursuing talent is obviously a key consideration for any company, let alone a giant one like Amazon. It’s just that, as one of the most influential technology companies in the world, I wish you would have gone about it in a more responsible way. Despite leaps in the technology of mobility and collaboration, it’s shocking in 2018 to see Amazon is still approaching how it accesses talent like it’s 1958. Building a few massive campuses in hopes that talented professionals want to live within commuting distance is not realistic. Instead, it’s time for U.S. businesses and cities to think smarter and smaller, leveraging the technology we already have and bringing work to workers.

We live in an age defined by knowledge work and mobility technology, giving companies the freedom to hire talent with the flexibility to live anywhere.

We not only have cheaper communication but more ways to work collaboratively outside of the traditional offices. Coworking spaces like WeWork have proliferated, and so have collaborative software tools like Microsoft’s Office 365, Slack, Google Hangouts, and Atlassian.

To have effective teamwork, Amazon didn’t have to create large campuses. It didn’t need to limit the potential cities for HQ2 down to all but the largest 70 across the U.S. and Canada. All Amazon needed to do was think smaller and smarter and a little more creatively.

The American dream is being threatened by our obsession with luring large corporate offices toward increasingly dense central business districts, and you just reinforced this path towards assured crisis.

From the very start, Amazon required applicants have populations greater than one million people. That requirement alone eliminated all but 70 U.S. cities, overwhelmingly concentrated along the already-prospering coasts.

Amazon could have sparked a boom in remote work options, allowing talent living in the Heartland, for example, to make Amazon wages in Des Moines — or a small farm town 100 miles outside of it. It could have created co-working space in hundreds of cities and towns across North America for those projects that would benefit most from face-to-face collaboration. Instead, Amazon decided to reinforce the glorification of in-person, mega-corporate work.

And then you gave the prize to the most illogical winners possible, if you look at this through the lens of actual need. If any U.S. cities outside of San Francisco and Seattle needed more jobs less it would probably be the New York and Washington, D.C. areas. Housing costs and wages there are already among the highest in the nation. And few major U.S. cities have tighter housing and labor markets, meaning that Amazon’s giant presence will likely lead to even higher cost of living and zero-sum wars for talent among their existing tech employers.

To add insult to injury, even though the city and state of New York combined have given Amazon nearly $3 billion of incentives, Amazon isn’t obligated to assist with many of the problems it will likely create, including an increased strain on infrastructure. In the aforementioned “National Landing” outside Washington, Amazon committed to build new infrastructure with Metro and National Airport to connect itself to the community. The deal with New York, on the other hand, doesn’t require Amazon to commit to helping the city’s deteriorating subway system, a major pain point that an influx of thousands of commuters won’t help.

What Amazon has committed to also doesn’t seem impressive. The executive director of New York City’s Employment and Housing Coalition called Amazon’s $5 million donation toward workforce training (combined with the city’s added $10 million) a “drop in the bucket.”

Setting aside what this backwards approach means to others, it’s not the right call even for Amazon. Bigger offices aren’t better, except maybe for the company’s own ego. There’s evidence to show that the benefits of colocation and “office community” break down once you reach about 150 people, as I’ve written about before. British psychologist Robin Dunbar studied communities and found that 150 is the maximum number of people with whom any one person can maintain a real sense of community. The so-called “Dunbar Number” represents the upper limit of a network with which you can “have a personalized relationship, one that is reciprocal — I’d be willing to help you out and I know you’d help me,” wrote Dunbar.

Amazon could have redefined what the office looks like for an entire generation — and brought prosperity and options to places that need it most. Instead, it just threw its own beauty pageant and we all come out losers.

Stephane Kasriel is the chief executive of Upwork, a global freelancing website, where he built and led a distributed team of more than 300 engineers located around the world as SVP of engineering before becoming CEO. He holds an MBA from INSEAD, an MSc in computer science from Stanford, and a BS from École Polytechnique in France.