Start with the piece of paper, because the paper is the retreat. Whatever else happens in the room, a strategic planning retreat for a small business has done its job only if everyone walks out with one sheet that holds five lines:
- The number. The one figure the business is steering by for the next 90 days, with today’s value written next to it.
- Three priorities. Each with exactly one owner and a due date inside the 90 days.
- The stop list. What the business will do less of, or not at all, to make room for the three.
- The failure list. The three most likely reasons this plan dies, each with the early sign that it is happening.
- Three dates. The 30, 60 and 90-day check-ins, already sitting in everyone’s calendar.
Every block of the agenda below exists to fill one of those lines. If a session you are tempted to add does not fill one, it belongs at a different meeting.
Why the horizon is 90 days, not a year
The quarter as a planning unit is not our invention. EOS, the Entrepreneurial Operating System, calls its version “Rocks”: the three to seven most important priorities a company, department or individual must complete in the next 90 days, each owned by one person, on the stated ground that when more than one person is accountable, nobody is (EOS Worldwide). Its own case for the quarter is partly a claim about attention spans, which we would treat as a rule of thumb rather than research, and partly a practical one we find convincing: if you choose the wrong priorities, you lose one quarter before correcting course instead of a year (EOS Worldwide).
For a business under ten people we would cap it at three, not seven. Seven priorities in a five-person company means most people own two or three, and the day job eats all of them by week three.
A 90-day plan still needs something to point at. If you have not yet done the arithmetic on where the business is meant to be in three years, do it before the retreat rather than during it; the method in our three-year goal breakdown turns a revenue target into a weekly capacity number, which makes a far better line one than a revenue figure you cannot directly control.
Two weeks before: three inputs, no slide deck
Most of what goes wrong at a retreat is decided before anyone arrives, when the organizer spends the prep time building a presentation. Send three things instead, and ask people to read them before the day.
The numbers, on one page. Revenue by month for the last twelve months, cash in the bank today, and the one operational figure you already count every week, whether that is jobs completed, orders shipped or billable hours. No commentary. People should arrive having seen the same facts.
One anonymous question to everyone attending. “What is the one thing that, if it stayed exactly the same for another year, would hurt this business most?” Anonymous matters. The person who sees the problem first is often the one least likely to raise it in front of the owner, and a free form tool such as Google Forms is enough to collect the answers.
Last quarter’s promises. If you made plans before, list what was said and what actually happened, line by line. It is the most useful and least comfortable page in the packet, and it stops the room from repeating the same three goals for the fourth year running.
The day, block by block
This is a one-day version, 9:00 to 3:30, with breaks. It assumes three to eight people. Say the decision rule out loud at 9:00: the team proposes and argues, and where the room splits, the owner decides and says why. Knowing that up front is what lets people argue honestly instead of politically.
- 9:00 to 9:30, read the numbers aloud. Someone other than the owner reads the one-page packet out loud. No discussion yet, only questions of fact. This fills line one: by 9:30 the room has agreed which single number matters for the quarter and what it reads today.
- 9:30 to 10:30, the anonymous answers. Group the survey answers into themes on a wall or board and read every theme out. The owner speaks last in this block, not first. Most of the real agenda for the rest of the day comes out of this hour.
- 10:45 to 12:00, candidates. Anyone can propose a priority, with one condition: they must say which number it moves and by roughly how much. Candidates that cannot name a number are parked, not debated. Cut the list to three. This fills line two, minus the owners.
- 12:45 to 1:45, the stop list. For each of the three, name what gets less time to make room for it. This is the block owners skip, and it is the one that decides whether the plan survives. A priority with nothing on the stop list is a plan to ask people to work more hours, which is not a plan. This fills line three.
- 1:45 to 2:30, the premortem. Covered in its own section below. This fills line four.
- 2:30 to 3:15, owners and dates. Each priority gets one name. The owner writes their own line in their own words, with a date, and reads it back. If the sentence would not let an outsider check on the due date whether it happened, rewrite it before moving on.
- 3:15 to 3:30, the calendar. Send the 30, 60 and 90-day check-in invitations before anyone stands up. This fills line five, and it is the fifteen minutes that most often gets dropped when the day runs late, so protect it.
The premortem, and why it goes after the plan
The technique comes from Gary Klein, chief scientist of the decision-research firm Klein Associates, who described it in the Harvard Business Review in 2007: team members assume the project they are planning has just failed and then generate plausible reasons for the failure (Harvard Business Review). Klein’s argument is that projects fail partly because people with reservations stay quiet during planning, and that making it safe for knowledgeable dissenters to speak up improves the odds (Harvard Business Review).
Here is how we would run it in a small-business room. Say: “It is 90 days from now. This plan failed. Everyone take three minutes and write down why.” Write silently, then go around the room and read one reason each until the lists run out. Group them, pick the three most likely, and for each one agree the earliest sign you would see. “A priority owner misses the 30-day check-in” is a sign. “Loss of focus” is not.
It has to come after the three priorities are chosen, not before, because the point is to attack a specific plan. Done first, it turns into a general list of worries. The signs it produces are also the natural raw material for a proper contingency plan, if the business does not have one; the trigger-first format in our 90-day contingency plan build takes exactly this kind of early sign and turns it into a three-line play.
Who belongs in the room
At least one person who does the work customers see: the person who answers the phone, runs the jobs or packs the orders. They know where the plan will break, and the premortem is worth far less without them. A retreat that only includes the owner and whoever manages the money tends to produce a plan built around the owner’s view of the business, which is the view the anonymous question was designed to get past.
If you work alone, you are not the exception here, you are the typical case. Of the 36,207,130 small businesses in the United States, 82.3 percent have no employees, according to the Small Business Administration’s Office of Advocacy (SBA Office of Advocacy). A solo retreat still works, but the premortem and the stop list need a second person to push back, and there are two no-cost places to find one. SCORE mentoring is a free service for owners of US-based businesses, online, by phone or in person (SCORE). Small Business Development Centers run nearly 1,000 local centers offering no-cost business consulting, funded in part by Congress through a partnership with the SBA (America’s SBDC). Ask either to sit in for the afternoon blocks.
The wall: paper or a board
For a room where everyone is present, a pack of sticky notes and a blank wall is the best tool there is, and it costs a few dollars. Go digital only if someone is joining remotely or you want the board to survive the day. Three real options, prices read from each vendor’s pricing page today:
- Miro has a free plan with 3 editable boards; Starter is $8 per member per month billed yearly and Business is $20 per member per month billed yearly (Miro pricing).
- FigJam, Figma’s whiteboard, is available on Figma’s free Starter plan, and the Collab seat that includes full FigJam access is $3 per month on the Professional plan (Figma pricing).
- Mural has a free plan with 3 editable murals at a time; Team+ is $9.99 per member per month billed annually or $12 billed monthly (Mural pricing).
For a once-a-quarter retreat, every free tier above is enough. Photograph the physical wall at the end of the day either way; the photo is the backup.
Where an AI assistant earns its place
Three jobs in this process are tedious and suit an assistant well. Grouping twenty anonymous survey answers into themes before the day. Turning a vague priority (“improve customer service”) into a sentence an outsider could check on a date. And drafting the check-in notes for the 30, 60 and 90-day meetings from the finished sheet, so nobody has to rebuild the context from memory.
The Strategic Planning Retreat Designer prompt at BusinessPrompter.com is listed under Strategy & Planning as a prompt to design and facilitate retreats that produce actionable plans; it is a Pro prompt with a single-prompt version and a five-step chain. Its own description leans toward early-stage founders building a burn-rate analysis and runway plan, so it fits best if cash is the number on line one of your sheet, as preparation for the 9:00 block rather than a script for the whole day.
What an assistant cannot do is the stop list. Deciding what the business does less of is a judgment about people’s time and the owner’s priorities, and it only sticks if the people affected were in the room when it was decided.
What we would cut from most retreat agendas
This section is opinion.
We would cut the second day. Two-day retreats tend to spend the extra time on the parts that feel productive, the vision statement and the big-picture discussion, and the difficult blocks get compressed into the last tired hour, which is where stop lists get softened. One day with a hard end forces the choices.
We would cut the icebreakers and the SWOT grid, because the anonymous question does the useful part of both in less time. And we would cut any priority whose plan amounts to the same team doing more. The best outcome of a planning day is a small team that ends the quarter doing more valuable work, with some of the repetitive work taken off its plate, not a team stretched further to hit a number chosen in a room.
Between the check-ins, the plan needs a weekly pulse or it drifts. The lightest version we know is the one in our piece on weekly reports to an accountability partner: three lines a week from each owner, sent to one person who is expecting them.
Frequently asked questions
How long should a strategic planning retreat be for a small business?
One focused day is enough for a team of three to eight people, provided the numbers and an anonymous one-question survey go out two weeks ahead. The day ends when there is a one-page sheet with the steering number, three owned priorities, a stop list, a failure list and three check-in dates in the calendar.
What should a strategic planning retreat agenda include?
Seven blocks: reading the numbers aloud, reviewing the anonymous answers, choosing three priorities that each move a named number, a stop list, a premortem, owners and dates, and sending the 30, 60 and 90-day check-in invitations before anyone leaves. The premortem, described by Gary Klein in the Harvard Business Review, asks the team to assume the plan has already failed and list plausible reasons why (Harvard Business Review).
Do I need an outside facilitator?
Not if the team is honest with each other, but a solo owner needs a second person for the stop list and the premortem. SCORE mentoring is free for owners of US-based businesses (SCORE), and Small Business Development Centers offer no-cost consulting at nearly 1,000 local centers (America’s SBDC).
One test tells you whether the day worked, and it happens the Monday after. Ask each priority owner, without letting them look at the sheet, to say their priority, the number it moves and the date it is due. If all of them can, you have a plan. If they cannot, you have a document, and the first 30-day check-in should start by rewriting it.
