Plan your design team at 80%. A team booked to 100% can only react.

Your design team is late on things and the obvious read is that you need another designer. Before you open the role, check a number almost nobody writes down: how much of each person’s week is actually assigned to work. If the answer is all of it, you do not have a headcount problem yet. You have a planning problem, and adding a person to a plan built at 100% gives you a bigger team that is late on more things. Plan at about 80% instead, leave the rest open on purpose, and defend it like it is committed work. It is.

For me that is arithmetic, not a philosophy. I plan about 32 of each person’s 40 hours a week for project work. The other eight cover meetings, the asks that arrive Tuesday and are needed Thursday, and the hours it takes to learn a tool that did not exist last quarter. A dashboard flags anyone running over. I run this with seven people at a multi-billion-dollar apparel platform, a team I built from one designer over two years, and the trade I am making is deliberate. I would rather the team be consistently reliable than occasionally extraordinary.

The last 20% of capacity is the most expensive capacity you can buy

Anyone who has sat in traffic already understands the mechanism. A highway at 70 percent of capacity moves. The same highway at 95 percent, same lanes, no accident anywhere, crawls. Nothing broke. The road ran out of the room it needed to absorb ordinary variation, so every small slowdown propagates backward instead of dissolving. Work queues behave the same way. As you approach full utilization, the wait time in front of every task climbs much faster than the utilization does.

Design work carries far more variation than traffic does. A two-day flow turns out to need a conversation about the data model. A research recruit falls through and the study slips a week. A stakeholder who owns one decision is out until Monday. None of those are failures, they are Tuesday. At 32 planned hours, each one gets absorbed and the date holds. At 40 planned hours, each one pushes into work that was already spoken for, and the slip travels down the line until it reaches the thing your CEO is watching.

A plan with no room in it is not an ambitious plan. It is a forecast that becomes wrong the first time anything is late.

The teams that run this best are not design teams

The clearest version of this discipline I know of is written down by Google’s site reliability engineering org, which runs systems where being late is measured in outages. Their rule is in chapter one of the SRE book (opens in new tab): "Google places a 50% cap on the aggregate ’ops’ work for all SREs," meaning tickets, on-call, and manual tasks, and "an SRE team must spend the remaining 50% of its time actually doing development." They are direct about the reason. "This cap ensures that the SRE team has enough time in their schedule to make the service stable and operable."

Read that as a management move rather than an engineering one. The reserve is not there because the work is hard on people, though it is. It is there because reactive work expands to fill whatever space you give it, and the only way to guarantee the proactive half happens is to make the reactive half illegal past a line. If the most operationally serious engineering org most of us can name writes that line at 50%, a design team asking for 20% is not asking for a lot.

The same company made the more famous version of this argument twenty years ago. In the 2004 founders’ letter (opens in new tab) that went out with Google’s IPO, Larry Page and Sergey Brin wrote: "We encourage our employees, in addition to their regular projects, to spend 20% of their time working on what they think will most benefit Google." I would not hold that up as a model to copy, because the honest history is that it got squeezed as the company scaled and plenty of people never took it. That is the useful part. Unprotected open time is the first thing a hot quarter eats, which is exactly why I set mine as a planning number that shows up in the schedule rather than a benefit in an onboarding deck.

What the eight hours actually absorb

Start with the urgent asks, because those are the ones leaders assume are cheap. They are not, and the cost is not just the hours. Gloria Mark, Daniela Gudith and Ulrich Klocke ran a controlled experiment on this and published it at CHI 2008 as The Cost of Interrupted Work (opens in new tab). People finished interrupted tasks in less time, with no drop in quality, by compensating: working faster. The price showed up somewhere else, in more stress, higher frustration, more time pressure and more effort. So an interrupt does not bill you two hours. It bills you two hours plus a tax on everything the person touches for the rest of the day, and if there is no reserve, that tax gets paid out of their evening.

The base rate for that is already bad before you add anything. In Microsoft’s 2024 Work Trend Index (opens in new tab), a survey of 31,000 knowledge workers across 31 markets, "68% of people say they struggle with the pace and volume of work, and 46% feel burned out." A team planned to the last hour is not neutral against that number. It is a multiplier on it. The reserve is the cheapest retention program I have ever run, and I have never had to ask anyone to stay late to make the plan true.

Then there is learning, which is the line item that quietly decides whether your team is current. When the business picked a new search platform for the whole site, I had every designer learn what it could and could not do before the work was assigned to us. Not one hour of that was on a ticket. It came out of the open time. When the integration finally landed on our busiest pages, most of the product page redesign we had in flight carried over, because search changed what loads on the page rather than the design underneath it. A team with no slack would have met that platform on the day the ticket arrived and eaten a redesign.

The same reserve is why a vendor proposal never became a contract. Our research team wanted a faster way to pull answers out of its studies, and a vendor quoted a per-seat yearly fee for the one feature we needed, bundled with a dozen we did not. With over a hundred people on the platform, that was a six-figure bill every year. I built and tested the one feature the same afternoon the proposal arrived, then turned it into an assistant the digital team uses to look things up on its own. That afternoon existed because it was planned to exist.

The part people get wrong: incubation does not live in the slack

Most versions of this idea say the leftover time is where the exploratory work happens. That framing is how exploratory work dies. Anything funded by leftovers is the first thing a hot quarter takes, and a designer who is behind is never going to spend their Friday on a speculative prototype. So on my team, incubator work is planned inside the 32 hours as real project work with a real owner. Every designer carries at least one. The reserve’s job is not to host the future work. Its job is to protect the plan that already contains the future work.

That distinction is the whole difference between a design team that services a roadmap and one that helps write it. My team runs a discovery framework I built, tests with customers and internal stakeholders, and then hands product management a validated opportunity rather than an opinion. Eighty percent of our 2024 roadmap came from work the team incubated the year before. One opportunity we surfaced on our own became a top priority for an executive the following year. Some of those threads run twelve to eighteen months of research and iteration before a customer ever sees them, which is time that simply does not exist inside a fully booked plan.

A team planned to the last hour will always be excellent at answering the question in front of it, and will never once be the reason a better question got asked.

Consistently reliable beats occasionally extraordinary

I will say the stance plainly, because it is the whole argument and it costs me something to hold. Given the choice between a team that hits its dates every quarter and a team that produces something spectacular twice a year and misses in between, take the reliable one every time. Not because ambition is bad. Because reliability is what pays for ambition.

Watch what actually happens in an organization. The team that hits its dates gets invited into planning earlier, and then earlier again, until they are in the room where the work is being decided rather than the room where it is being handed out. The team with the brilliant quarter and two misses gets managed. Somebody adds a check-in, then a status doc, then a product manager to keep an eye on things. The organization remembers the misses with far more precision than it remembers the brilliance.

Trust is the mechanism. The right to walk into a planning meeting with an idea nobody asked for, and have it taken seriously instead of politely parked, is earned entirely by having been right about your dates for a year. That is also the foundation under any real case for design, which I get into in how I prove design actually moved the business. Extraordinary is a thing that happens to teams that are trusted enough to be pointed at hard problems. Reliable is how you become one.

How to run it so it does not become a story you tell

Write it as hours. An 80% target is a slogan until it is 32 and 40 in a spreadsheet next to a person’s name, with the assumption stated out loud: up to eight hours a week goes to meetings, interrupts and learning. Say the assumption, because the argument you will have is always about what counts, and settling that once beats settling it every sprint.

Make overcapacity visible before it is felt. We run a dashboard that flags anyone over their planned hours, which turns a conversation that used to be a confession into a number on a screen. A designer should never have to volunteer that they are drowning. The system should say it first, and a leader who only learns about overload in a one-on-one is learning about it three weeks late.

Fix intake, because unclear work is how a 32-hour plan turns into a 45-hour week. We had an under-scoped handoff cost the team several extra design cycles, and my fix was a more rigorous intake process so design asks its questions before the work starts instead of discovering them in the third review. The rework you avoid is capacity you did not have to hire for, and the shortcuts you skip are the ones that would have turned into design debt you pay interest on for years.

And never re-plan the reserve. Every quarter someone does the arithmetic, notices eight unassigned hours per person across seven people, and proposes a project that fits perfectly in the gap. It always fits perfectly. Six weeks later the interrupts have nowhere to go and the team is back to working late, except now there is also a project to finish. The reserve is only a reserve if it is boring and untouchable.

Who each path is for

If you run the team, measure before you announce anything. Take the last two months, add up planned project hours per person against calendar hours, and look at the number. Most teams I have looked at are planned somewhere between 105 and 130 percent and are covering the gap with evenings, which is why the plan looks fine right up until someone quits. You do not need permission to fix that. You need a spreadsheet and one uncomfortable planning meeting.

If you are the executive approving the headcount, ask for planned hours against calendar hours before you approve the req. A team planned at 125% does not get faster when you add a person, because the new person joins the same broken plan and needs a quarter of someone’s time to ramp. Correct the plan first, watch one full cycle, then hire. You will often still hire. You will just know what you are buying.

And if the real gap is that nobody in the building owns how design is planned, staffed and pointed at the roadmap, another designer will not fix it and neither will a dashboard. That is an operating model that does not exist yet, and it is cheaper to borrow than to buy. I lay out the sequence of hires, the roles that come in what order, and the operating model underneath them in how to build a design team. Start there before you write the job description.

When running at 100% is the right call

There are real cases where I would book the team flat out, and I would rather name them than pretend the rule is universal. The first is a dated crunch. A launch, a trade show, a compliance deadline with a legal date on it. Run hot, name the end date before you start, and give the time back on the other side. What breaks teams is not the crunch. It is the crunch with no end date, where the emergency turns out to be the operating model.

The second is a very small team. With one or two designers there is no statistical smoothing to protect, and formal capacity planning on two people is overhead dressed up as rigor. The principle still holds in a simpler form: do not assign all five days. But the tool is the founder’s judgment, not a dashboard.

The third is a billable model. If you sell hours, the reserve comes straight out of margin, and that is a pricing decision rather than a planning one. Agencies that run designers at 95 percent utilization are making a defensible choice about their business. The tradeoff is that a team booked that tightly has no room to incubate, so the ideas have to come from the client.

One more, and it is the honest limit on everything above. If your team is planned at 80% and still missing dates, the reserve is not your problem and adding more of it will not help. Look at scope discipline, at whether requirements are settled before work starts, or at whether the skill on the team matches the difficulty of what you are asking for. Capacity planning fixes overload. It does not fix ambiguity, and it does not fix a mismatch between the work and the people.

Here is the question I would put to any design leader you are evaluating, and I would want it put to me. What did your team work on last quarter that nobody asked them for? If the answer is nothing, that team is fully booked and fully reactive, and every roadmap it touches will keep arriving pre-written by somebody else. The hours you leave open are not a cushion or a perk. They are the only part of the week where your design team gets to be ahead of you.