Goal Setting Framework for Small Business: Four Rules

In the early 1990s, Sears set its auto repair staff a sales goal of $147 an hour, and the staff hit it by overcharging customers and doing repairs nobody needed. The case opens the Harvard Business School working paper Goals Gone Wild (Ordóñez, Schweitzer, Galinsky and Bazerman, 2009), which records that the company’s chairman later acknowledged the goals had pushed employees to deceive customers. Nothing about the goal was vague. It had a number, a unit and a deadline, and that is exactly why it did so much damage.

That is the problem a goal setting framework has to solve in a small business, where the owner is usually the person setting the goal, chasing it and judging it. Specific, hard goals are one of the most replicated findings in management research: Edwin Locke and Gary Latham’s 35-year review in American Psychologist (2002) summarizes the core result, and the Goals Gone Wild authors restate it in one line: “Compared to vague, easy goals (e.g., ‘Do your best’), specific, challenging goals boost performance.” So the framework below is not a template for making goals more specific. It is four rules for deciding which specific goals are safe to set, and what to write next to each one so it does not become a Sears goal at a smaller scale.

Rule 1: A number, a date and a place to read it from

“Grow the maintenance side of the business” is a wish. “Twelve maintenance contracts active by March 31, counted in the invoicing software” is a goal. The third part matters as much as the number. If the figure cannot be read off a system the business already uses, such as the invoicing app, the bank feed or the booking calendar, the quarter ends in an argument about what counts, and the goal quietly becomes “do your best” again.

Pick the horizon before the number. A quarter is short enough to feel and long enough to move something; anything beyond a year belongs in a separate exercise, and our three-year goal playbook covers how to turn a long horizon into a 90-day number you can check.

Write it like this: the number, the date, and the system the number will be read from.

Rule 2: If nobody in the business has done it before, set a learning goal

This is the rule most owner-written goal lists miss, and it comes from the same research tradition. Gerard Seijts and Gary Latham argued in the Academy of Management Perspectives (2005) that when a task mainly requires acquiring knowledge and skill rather than more effort and persistence, an outcome goal can hurt performance, and the right move is a specific, challenging learning goal that shifts attention to discovering strategies that work. In plain terms, an outcome target on a task you have never done tells you nothing about how to do it, and it invites you to abandon the search the first month you miss.

A hypothetical to make it concrete: a two-person cleaning company has never run paid ads. “Twenty leads from online ads by June 30” is an outcome goal set on a task nobody in the business knows how to do. “By June 30, find out which of three ad formats produces a booked job for under $60, running each for two weeks” is a learning goal. Once the second question is answered, the first becomes a fair target for the following quarter.

Write it like this: “By [date], find out [one specific question],” followed by the test you will run to answer it.

Rule 3: Mark every goal committed or aspirational, and stop at four

Google’s re:Work guide to OKRs suggests three to five objectives with about three key results each, grades each one from 0.0 to 1.0, and says the sweet spot is 60 to 70 percent, because consistently scoring higher means the goals were not ambitious enough. That advice was written for a company that can afford to miss. A small business cannot run payroll on 70 percent. What makes the method usable at this size is a distinction Google itself draws, explained in WhatMatters’ OKR FAQ: a committed OKR must be fully achieved, graded pass or fail, and a team that cannot deliver escalates early; an aspirational OKR is expected to average about 70 percent and rolls forward until it is done.

Our view: in a business under 20 people, a committed goal is anything with a consequence attached: the cash balance, a tax deadline, a contract delivery date, keeping a key employee through the busy season. Everything with a consequence is committed and graded pass or fail. Then allow one aspirational goal a quarter, scored on the 70 percent scale, and only one, because a second stretch goal competes for the same ten spare hours a week. Three committed goals and one aspirational is a full quarter for most owners. If you run a planning day, a one-day strategic planning retreat is built to end with exactly that page.

Write it like this: a C or an A in front of every goal, and never more than four goals on the page.

Rule 4: Write the side-effect line before the goal goes live

Goals Gone Wild lists five documented side effects of goal setting: a narrow focus that neglects everything outside the goal, a rise in unethical behavior, distorted appetite for risk, damage to the culture, and lower intrinsic motivation (Harvard Business School working paper 09-083). The authors’ conclusion is that goal setting should be treated like a prescription drug rather than an over-the-counter remedy: dosed carefully, with the side effects known in advance and someone supervising. For an owner, the practical version is one question asked of every goal before it is adopted: what would a reasonable person do to hit this number that would hurt the business?

For a revenue-per-hour target, the Sears answer was unnecessary repairs. For “forty five-star reviews by December,” it is pressuring customers or quietly steering unhappy ones away from the review link. For “cut material cost 10 percent,” it is the cheaper supplier whose product fails inside a year. The line you write next to the goal is the counter-measure, and it is usually a second number watched beside the first: refund rate next to revenue per job, review requests sent next to reviews received, warranty callbacks next to material cost. A goal that can only be hit by leaning on your staff or your customers is a bad goal, however specific it is.

Write it like this: “Could be hit by [the shortcut]; watch [the second number].”

The one-page quarter sheet

Put the four rules together and a quarter fits on one page with four columns. Below is a filled-in example for a hypothetical six-person heating and cooling company. The numbers are illustrative, not benchmarks.

Two committed goals carry the consequences, the learning goal covers the one thing nobody in the business has done before, and the single aspirational goal is where the ambition sits. The sheet is only alive if it is read weekly. The strongest evidence on that is covered in our accountability partner playbook: a short progress report sent to one person every week beat a written plan on its own. The same sheet should feed the candidate list in a weekly priorities system, so Monday’s plan is pulled from the quarter rather than from the inbox.

Where the sheet lives, and what it costs

It needs to live somewhere the team already opens, with a column per week so the grades accumulate. Three options, from free to dedicated:

• Google Sheets. Free with a personal Google account. If the business runs on Google Workspace, Business Starter is $7 per user per month and Business Standard is $14, billed monthly, and both include Sheets (Google Workspace pricing). Four rows and thirteen weekly columns is the entire system.
• Perdoo. A dedicated OKR tool whose Free plan covers up to five users and includes OKR and KPI tracking (Perdoo pricing). That is enough for one owner and a small team; the paid plans are shown in euros by default, so switch the page’s currency toggle before comparing.
• Asana. Its Goals feature, which ties objectives to the tasks underneath them, is only on the Advanced plan at $24.99 per user per month billed annually or $30.49 billed monthly; the free Personal plan covers two users but does not include Goals (Asana pricing). Worth it only if the team already runs its work in Asana.

An AI assistant earns its place in one step of this: drafting the side-effect line. Paste the four rules and your rough goal list into ChatGPT, Claude or Gemini, all of which have free tiers, and ask for the shortcut and the second number for each goal. The assistant drafts, and you keep only the lines that describe something a real person in your business might do. It should not pick the goals; it does not know which consequence you can live with.

If you want a structured walkthrough for the first draft, BusinessPrompter.com publishes a Goal Setting Framework prompt, a Pro (paid) prompt in its Personal Development section, offered as a single prompt and a five-step chain. Its description is written for individuals and entrepreneurs working on personal and career goals rather than a company quarter, so bring the four rules with you and apply them to whatever it returns.

Before adopting any goal this quarter, put it through the question Sears never asked: name the shortcut that would hit the number and hurt the business. If you cannot name one, the goal is probably a learning goal in disguise. If you can, write it down next to the target, because the shortcut you just named is the one someone will reach for in week eleven.