Know who has capacity before you overcommit the team

Who this is for

This playbook is written in agency and professional-services language: studios, consultancies, implementation shops, and retainers where people are the inventory. If a principal can sell a project the floor cannot staff, or if “we will squeeze it in” is the scheduling system, you are in the right place.

It is not a manufacturing OEE guide and it is not an HR performance program. If the pain you feel is extras that never become invoices, read unbilled scope in the same month — full people and free work often arrive together. If the pain is a yes that dies after the sales call, start at pipeline follow-up. Capacity starts when the commercial yes is about to become a delivery yes.

Symptoms

If several of these are true in the same month, this is a live operational leak — not a tooling preference.

• A seller books a kickoff and delivery learns about it from the client’s calendar invite.
• “Who has bandwidth?” is a hallway question with three different answers.
• Retainers are treated as leftover time even though the same people are on two launches.
• PTO, sales work, and internal rebuilds are invisible, so the sheet looks open when the week is not.
• The team is tired and the pipeline still looks thin because committed work was never written down.
• Someone quoted a utilization target from a blog and wants every person managed to that number.
• New work is pulled forward to “keep people billable” while last month’s extras still have no home.

What done looks like

Done is a working operating change, not a purchased seat or a dashboard nobody opens.

• Each person or role has a stated available time for the current week and the next two.
• Committed work is listed: sold jobs, named retainer load, and non-negotiable internal work.
• Available versus committed is visible to the people who can say yes to new work.
• A weekly review happens before new commitments, with a named owner.
• Utilization is used to plan load, not to rank or punish individuals.
• A yes that does not fit is delayed, restaffed, or declined — not absorbed in silence.

Available versus committed, in agency language

Utilization talk goes wrong when it starts with a target. Start with two piles. Available is the time a named person can actually give: after PTO, after the sales day the principal already booked, after the all-hands you already know is happening. Committed is work you have already promised: a sold project, a retainer with a real load, a launch date the client has. Pipeline is not committed. Hope is not committed.

The gap between those piles is the only number that matters this week. If committed is already above available, you are not “busy in a good way.” You are making promises with time you do not have. If available is far above committed, you have a sales or scoping problem, not a heroics problem.

Write people by name when the team is small. Write roles when names would turn the sheet into surveillance. The test is whether a seller can answer “can we take this kickoff in two weeks?” without pinging four people. If they cannot, the view does not exist yet.

There is no universal utilization target

You will be offered a benchmark. Seventy percent, or some other round number, will show up in a deck as if it were a law of professional services. It is not. Mix, seniority, sales load, and the kind of work you sell change what a sane week looks like. A principal who still closes and delivers cannot be planned like a dedicated producer. A retainer studio and a project shop do not share a target.

If someone insists on “the” utilization number, ask what they counted as available and what they counted as committed. If they cannot answer, they are repeating a slide. Use your own week. The question is whether the next yes fits, not whether you match a blog.

Do not invent a utilization percentage to justify a hire or a tool. Describe the overcommit: two launches on the same three people, a retainer that was sold as leftover, a principal who is already in sales meetings four days. That is enough to decide.

Review weekly, before the next yes

A monthly staffing meeting is a postmortem. The damage is already in the calendar. Put a weekly look on the same rhythm as the work. Thirty minutes is enough if the view exists. Who is over committed. Which yes is still unsigned. Which date will slip if nothing is declined.

The people who sell have to see the view before they promise dates. Hiding capacity from sales “so they stay hungry” produces a full floor and a surprised client. If sales and delivery cannot share one picture, you will keep buying project-management software to store the argument.

Onboarding is where an overcommit becomes a client experience. A kickoff booked into a full week feels like disorganization even when the work is good. Automate client onboarding without the email chase needs a date that the capacity view already blessed. Do not automate a kickoff sequence that ignores load.

Follow-up has the same dependency. A pipeline that is moving is good only if the next close can be staffed. Follow up before good prospects go cold is the motion that keeps the next step visible. Pair it with a capacity check before the close email promises a start date.

A worked week, without a benchmark

Picture a ten-person studio. Three people are on a launch that already slipped once. Two are on a retainer that answers in the same day. The principal is in sales conversations four half-days. A seller wants to promise a new brand kickoff “next Monday.” Do not reach for a utilization target. Write available time for those three launch people — after the slip work you already promised. Write the retainer as committed, not leftover. Write the principal as partly unavailable. The new kickoff does not fit. The honest answers are a later date, a different team, or a no.

The same week can look “open” if you only count project codes. The launch extras, the retainer, and the sales days were the real committed pile. A sheet that hides them will green-light the kickoff and surprise the floor. That is why the weekly review has to happen before the close email, not after the invite exists.

Write the decision down: delayed two weeks, or declined. A verbal “we will see” is how overcommit returns on Thursday. The client can handle a later date more easily than a kickoff that nobody knows how to staff.

Do not weaponize utilization

The moment the sheet becomes a scoreboard, people hide work. They stop logging internal time. They stop flagging extras. They look busy in the tool and still miss the launch. A planning signal that punishes honesty will become fiction.

Use the view to move work, decline work, or change a date. Do not use it to rank humans in a public channel. If a person is consistently over committed, the first question is what was sold onto them, not what is wrong with them. If a person is consistently open, the first question is whether the work is invisible — unpaid extras, sales assist, cleanup — not whether they are lazy.

This is why billing and capacity belong in the same conversation. Unpaid extras fill the week and leave the utilization sheet looking “fine” or “soft,” depending on whether you count them. Catch scope creep before it never hits the invoice so the committed pile includes the work you are actually doing, not only the work you remember to sell.

What people usually get wrong

They buy a resourcing platform before they can list this week’s commitments on one page. The platform will store the same hallway answers in a prettier grid.

They count only billable project codes and ignore sales, management, and the rebuild that already ate Thursday. Available time that ignores known work is a lie.

They treat retainers as slack. A retainer with a real response expectation is committed work. If you sold it as leftover, you sold a fiction.

They chase a benchmark and miss the local question: can we staff the thing we are about to promise? The benchmark cannot answer that.

DIY vs hire

DIY a two-week view. List the people who deliver. Write their available time. Write the sold work and the retainer load. Sit sales and delivery in one short meeting and mark the next three yeses as fit, delay, or decline. No new tool is required. A shared page with an owner is enough to learn whether the leak is visibility or something larger.

Hire help when sales and delivery cannot share a picture, when overcommit is chronic across several teams, or when a staffing purchase is about to happen without a weekly review. The $997 assessment is a live consultation and a written analysis. You leave with two to four improvements you can run yourself — a weekly view, a yes rule, a stop on weaponized targets — and a later roadmap quoted only if you want us on a staffing or delivery system. Implementation is not in the fee.

Controls before automation

Recommendations that touch customers, money, contracts, private data, or systems of record need human review, limited permissions, a log, and a fallback. Do not automate a broken path because the tool is ready.

• Do not publish individual utilization as a ranking or a shame metric.
• Do not promise a start date the weekly view has not blessed.
• Keep PTO and known internal work in the available-time math.
• A staffing or PSA write-back needs a named reviewer and an undo — no silent overwrite of assignments.
• If the sheet and the calendar disagree, stop saying yes until one picture is true.
• Leave a fallback: if the tool is down, the weekly list still names who is over committed.

How this sits next to the other playbooks

Capacity is a revenue-motion room in the fourteen mid-market AI asks: keep the next commercial step honest about delivery. Follow-up moves the prospect. Onboarding turns the yes into a start. Billing makes sure the work you squeezed in can still become an invoice. None of those paths work if the team was already full when you said yes.

If you want that order written with you, book the assessment. Bring the last three yeses that hurt. Leave with a weekly view and a rule for the next one.