Underpromise and overdeliver. That’s the official line of advice in many business books. But what does it even mean?

Is underpromising about lowering the client’s expectations before the work even begins? Or delivering less than you’re capable of? Is it maybe about shipping exactly what’s asked for but on a slower timeline? Overdelivering comes with equally few instructions. Should you do what’s needed but faster? Do 10% more? 20%? 100%?

I still remember reading James Altucher’s take on this concept years ago: “You have to overpromise and overdeliver.” James suggested that overpromising “sets you apart from the people who underpromise” and overdelivering “sets you apart from people who just delivered.” “It’s not that hard to do both,” he wrote. “It’s easy to slightly overpromise and slightly overdeliver because nobody else is doing it.”

The truth is setting expectations and meeting them is highly context-depended. On Reddit, developers report great results with asking for more time they think is needed for a task to account for unexpected hiccups. If they turn in the work faster, they look like a hero. If bugs do turn up, at least nothing will catch fire until their original estimate is up. Meanwhile, a writer working against tight deadlines in quick succession can’t afford to take twice as long on every article. It’ll make them look unprofessional.

When I was self-employed, I didn’t have to promise anyone anything. Sometimes, I did. And while I often moved deadlines, I always ended up overdelivering—even if it was just to myself. I wrote posts tallying 5,000, 7,000, even 10,000 words or more in hopes of completely covering a topic. Sometimes that worked; sometimes it didn’t.

Now that I work for a company, overpromising and overdelivering seems like a good recipe to burn out within a few months and little else. My firm is basically a startup. Naturally, everyone asks for more all the time. Our visions are simply bigger than anything we can ever build with the people we have. In this context, it doesn’t matter how perfect each delivered project is. What matters is getting more work out the door. Underpromising keeps you alive, and then how much you can deliver will change for each next venture.

Samsung, on the other hand, can’t afford to ship even 2% of their phones with cracked screens. When you sell expensive devices to consumers with high expectations, overdelivering is tablestakes.

There are times to underpromise and overdeliver, times to overpromise and overdeliver, and perhaps even times to underpromise and underdeliver—for example to lose a bad client or show how hard a goal is to actually pull off.

Rules of thumb can be good starting points, but remember: You have nine more fingers, and it’s okay to use them differently each you get back to work.

Nik

Niklas Göke writes for dreamers, doers, and unbroken optimists. A self-taught writer with more than a decade of experience, Nik has published over 2,000 articles. His work has attracted tens of millions of readers and been featured in places like Business Insider, CNBC, Lifehacker, and many others. Nik has self-published 2 books thus far, most recently 2-Minute Pep Talks. Outside of his day job and daily blog, Nik loves reading, video games, and pizza, which he eats plenty a slice of in Munich, Germany, where he resides.