Set Expectations Right — The Art of Underpromise, Overdeliver

After 25 years of building technology companies, four exits, and more board meetings than I can count, I’ve come to believe that managing expectations is the single most underrated skill in tech leadership. It’s not glamorous. Nobody writes LinkedIn posts about it. But the leaders who master it build extraordinary careers, and the ones who don’t eventually flame out — no matter how talented they are.

The principle is deceptively simple: trust compounds when you consistently deliver more than you promised. It evaporates instantly when you miss a commitment. That asymmetry governs everything.

Why Tech Leaders Overcommit

I’ve done it myself. Early in my career, I’d walk into a board meeting, feel the expectant energy in the room, and find myself nodding along to aggressive timelines I knew in my gut were unrealistic. There’s a cocktail of forces at work.

Optimism bias is the obvious one. Engineers and product leaders are builders by nature — we see the solution before we see the obstacles. When someone asks “can you do this by Q3?” our brain immediately starts architecting the happy path. We forget that the happy path almost never materializes.

Pressure from investors and boards is real but often self-inflicted. In my experience, most sophisticated investors actually prefer conservative forecasts delivered consistently over hockey-stick projections that miss. But in the moment, when a board member raises an eyebrow at your timeline, the temptation to shave a month off is enormous.

The desire to please runs deep in many technical leaders. We got where we are by solving problems and saying yes. Saying “I’m not sure yet” or “that timeline isn’t realistic” feels like admitting weakness. It’s not. It’s the opposite.

Underestimating complexity never goes away, no matter how experienced you get. The difference is that experienced leaders build buffers for the complexity they know they’re underestimating.

The Unfair Math of Perception

Here’s something I wish someone had told me twenty years ago: delivering 110% of a modest promise creates dramatically more trust than delivering 90% of an ambitious one.

Think about it from the other side of the table. A CTO tells the board the platform migration will take nine months. It takes eight. The board thinks: “This person has it under control. They know their domain. I trust their estimates.” That trust carries forward into every subsequent conversation.

Now imagine a different CTO — equally talented, maybe more so — who promises six months and delivers in seven. Objectively, they delivered faster than the first CTO. But the board’s reaction is entirely different: “They missed their commitment. What else are they getting wrong? Should we be more hands-on?” That skepticism also carries forward, and it compounds just as powerfully as trust does, except in the wrong direction.

The math of perception is unfair. A miss is never just a miss — it’s a signal about judgment.

The Board Room

Board reporting is where this principle pays the biggest dividends. Over the years I’ve developed a straightforward approach.

Forecast conservatively and explain your methodology. Don’t just throw out numbers — show the board how you arrived at them. “Based on our velocity over the last three sprints, accounting for the two new integrations and the holiday period, we’re projecting delivery in Q2” is infinitely better than “we’re targeting Q1.” The first invites dialogue. The second invites disappointment.

Build in explicit buffer and be transparent about it. I tell boards: “Our engineering estimate is fourteen weeks. I’m forecasting eighteen to account for unknowns, integration testing, and the fact that estimates are always optimistic.” Nobody has ever criticized me for this. In fact, it signals maturity.

Then beat your forecast. Over multiple quarters, this builds an incredible track record. When you’ve beaten your own projections four quarters in a row, you earn something invaluable: the benefit of the doubt. When you eventually do hit a genuine blocker and need to push a timeline, the board trusts you because you’ve built up a reservoir of credibility.

During M&A due diligence, this discipline becomes even more critical. Every commitment you make is being evaluated not just on its face, but as evidence of your judgment and your organization’s execution capability. I’ve seen deals crater because a technical leader made promises during diligence that didn’t hold up three months later.

Inside the Organization

The same principle applies internally, and arguably matters even more because the cycles are shorter and more frequent.

Product roadmaps are promises. When you tell a sales team that a feature will ship in March and it ships in May, you haven’t just missed a deadline — you’ve undermined their ability to sell, damaged their trust in engineering, and created a cross-functional rift that takes quarters to heal. I’ve seen this pattern tear organizations apart.

Set honest timelines with buffer. If your team thinks something will take six weeks, communicate eight. If it lands in six, your team looks great. If it takes seven because of an unexpected dependency, you’re still ahead of your commitment.

Team capacity is another area where leaders routinely overcommit. “Yes, we can take on that initiative alongside the three others we’re already running” — said every CTO who then watched all four initiatives slip. I’ve learned to be ruthless about capacity planning. It’s better to fully deliver three things than to half-deliver five.

The Overpromise Spiral

The most dangerous pattern I’ve observed — and experienced — is the overpromise spiral. It goes like this.

You miss a deadline. You lose some credibility. The board or your stakeholders start asking more questions, requesting more status updates, getting more involved. You feel the increased scrutiny as pressure. To relieve that pressure, you make an even bigger promise: “We’ll make up the lost time and deliver the next milestone early.” You almost certainly won’t. You miss again. Credibility drops further. The scrutiny intensifies. The promises get more desperate.

I’ve watched talented leaders destroy their own positions this way. The exit from the spiral is painful but simple: stop. Acknowledge the miss honestly. Reset expectations to something achievable. Deliver on the reset. It takes two or three cycles of this to rebuild trust, but it works. The spiral only ends when you choose to break it.

The Discipline of “I Don’t Know Yet”

The hardest sentence in a leadership meeting is “I don’t know yet, but I’ll get back to you with a realistic estimate by Friday.”

It feels weak. It feels like you should have the answer. But compare the outcomes: committing to something on the spot that you later have to walk back, versus taking 48 hours to come back with a well-reasoned estimate that you then beat.

I’ve made it a personal rule: I don’t give timeline commitments in meetings where I haven’t had time to consult with my teams. If pressed, I give a range — “somewhere between Q2 and Q3, I’ll narrow it down this week” — and I follow up. Every time I’ve broken this rule and committed on the spot, I’ve regretted it.

The same goes for customer commitments. Sales teams will always want a date they can put in a contract. Give them one you’re confident in, not one that makes the deal easier to close today but creates a delivery crisis tomorrow. A customer who gets their feature a month early becomes an evangelist. A customer who gets it a month late becomes a detractor. Same feature, same team, wildly different outcomes — determined entirely by what you promised.

Compounding Trust

The beautiful thing about this discipline is that it compounds. After a year of consistently beating your own forecasts, something shifts. The board stops questioning your timelines. Stakeholders stop asking for status updates every week. Your teams feel the reduced pressure and, paradoxically, often deliver faster because they’re not operating under artificial deadlines.

You earn the most valuable currency in leadership: the assumption of competence. People stop second-guessing you and start backing you. That’s when you can truly lead — when your credibility is so established that you can take genuine risks, propose bold initiatives, and ask for patience on hard problems, because everyone in the room knows that when you commit to something, it happens.

The Energy Connection

I once had a discussion about leadership with Satya Nadella, who was my supervisor long before he became CEO of Microsoft. His view was clear: leadership is fundamentally about creating energy. Not managing processes, not setting strategy — creating the energy that makes everything else possible.

That insight connects directly to expectations management. When you overdeliver, you create positive energy. The team feels capable. The board feels confident. Customers feel valued. That energy is self-reinforcing — it attracts talent, accelerates execution, and compounds into momentum.

Underdelivering does the exact opposite. It destroys energy. The worst case isn’t just missing a target — it’s the negative downward spiral that follows. The team loses confidence. Stakeholders lose trust. The increased scrutiny creates pressure that leads to more overcommitting, more misses, more energy destruction. I’ve seen this spiral consume entire organizations. Good people start leaving. The remaining team operates in a defensive crouch, focused on survival rather than ambition.

The way out is almost absurdly simple: stay in the comfort zone of what you can reliably deliver. Underpromise. Then overdeliver. The positive energy this creates is transformative — and it’s entirely within your control.

It starts with the smallest discipline: promise less than you think you can do. Then do more than you promised. Every single time.