Strategic Planning: How to Make a Great Plan That Works

Build a strategic plan with SWOT, PESTLE, and OKRs, then fix the timelines underneath it, which are where plans usually break.

Students in the final semester of an honors thesis course were asked to predict, as accurately as they could, the day they’d hand the thesis in. They averaged 33.9 days. It actually took them 55.5.¹

Then the researchers tried the obvious fix and asked for a worst-case estimate instead. That moved the number, but it didn’t improve the estimate.

Your strategic plan probably has the same problem, one layer down from the strategy itself, in the timelines written underneath it.

What Is Strategic Planning?

Strategic planning is the process an organization or an individual uses to set long-term goals, work out which resources and actions will reach them, and record that as a plan that guides later decisions. A good one keeps resource allocation, project prioritization, and the people involved pointed in the same direction.

By convention, strategic plans run three to five years out. Stable industries sometimes stretch that to ten, and fast-moving ones pull the horizon in to one or two.

Underneath it sit shorter tactical and operational plans:

As an individual, the same three layers work: a three-to-five-year direction, this quarter’s projects, and this week’s tasks.

Start With Where You Actually Stand

You need an honest read on where you are before you can plan where you’re going. Three frameworks do that job, and which helps most depends on what’s most likely to move your plan.

SWOT: What You’ve Got and What’s Coming

SWOT stands for strengths, weaknesses, opportunities, and threats. Two of those look inward at the organization, and two look outward at what you don’t control.

• Strengths. What do you do measurably better than competitors, and what’s the proof?
• Weaknesses. Where does the work slow down, and which skills is the team missing?
• Opportunities. Where are customer needs moving, and where are competitors leaving a gap?
• Threats. Which competitor, cost change, or regulation could hurt you next year?

Here’s a free table you can copy for your own SWOT. As an individual, run the same four questions on yourself.

PESTLE: The Forces Outside the Building

PESTLE widens the frame to six things that happen whether you plan for them or not.

• Political. Which government decisions could change how you operate?
• Economic. How would a shift in rates or customer spending hit your costs?
• Sociocultural. What change in what people value would reshape your market?
• Technological. What could make your current approach obsolete?
• Legal. Which employment or industry rules are you about to be caught by?
• Environmental. How do sustainability pressures land on your supply chain?

The Balanced Scorecard: Four Angles on One Strategy

The Balanced Scorecard checks one strategy from four directions, so a plan that wins on a single measure while quietly losing on another has somewhere to show up.

• Financial. Which money goals does this plan hit?
• Customers. What problem are you solving, and how would you know if satisfaction slipped?
• Internal processes. Which processes deliver this strategy, and where do they break?
• Learning and growth. What do people need to learn to run it?

Action Step: Pick one framework, answer its questions in writing this week, then send your answers to two people who see the business differently and ask what you got wrong.

Set Objectives You Can Measure

Once you know where you stand, you pick what you’re aiming at. One common way to do it is OKRs: objectives and key results.

The objective is the qualitative half: where you want to end up, stated in a way people can repeat back. The usual advice is three to five, tied to what the organization is for.

The key results are the quantitative half: commonly two to five measurable outcomes per objective that would prove you got there. This is where the SMART test earns its keep: specific, measurable, achievable, relevant, and time-bound.

A company objective of boosting customer satisfaction might carry a key result of reaching a Net Promoter Score of 70 within 18 months. An individual might aim for $100,000 a year by 2030.

Pro Tip: Write the key result before you write the plan to hit it. If you can’t say what number would prove the objective happened, the objective is still a wish.

Turn Objectives Into Resources and Milestones

The how-to half of the plan runs in four moves.

1. Prioritize. Sort the objectives by importance and urgency, and get input from the team and leadership before you lock the order.
2. Break each one into chunks. Split every objective into tasks small enough to assign. Getting to an NPS of 70 might start with a customer feedback survey and acting on what comes back.
3. Allocate the resources. For each task, name the time, the people, and the budget it needs. That survey might mean $5,000 of software, two marketers, and a data analyst.
4. Set milestones and deadlines. Fix the interim dates on the way to the target so progress is visible while there’s still time to change course.

Then you track it and adapt. As I’m sure you know, things don’t always go according to plan.

And that’s where plans usually come apart, for a reason that has nothing to do with your strategy.

Why the Dates in Your Plan Come Out Optimistic

Go back to those thesis students. Asked for their most accurate guess, they averaged 33.9 days and took 55.5, and fewer than a third finished by the date they’d named.¹

Then the researchers asked for a worst-case estimate instead: what they’d predict if everything went as badly as it possibly could. That pushed the average out to 48.6 days, and fewer than half of them still made it. But those estimates were no more accurate than the ordinary ones—off by 23.2 days on average against 22.6.¹ Bracing for the worst moved the number without sharpening it.

The reason shows up when you ask people to write down what’s going through their heads as they estimate. Almost everyone, 93.5% of them, described how the work would go. Under one in ten mentioned anything that might get in the way, and 8.9% mentioned how similar past projects had actually gone.¹ Nobody at all mentioned how long the same kind of work had taken someone else. An estimate built that way rests on a story about how this project goes well, with your own track record barely in the room.

When people do look back, they tend to explain the miss away. Asked why an earlier project had run late, people explained their own lateness with reasons more external, more temporary, and more specific to that one occasion than the reasons they gave for a friend’s late project.¹ That’s the shape of every explanation that ends in “but that was a one-off”: the brief changed, someone was out sick, the software broke that week. Every one of those is a reason this time will be different.

The Fix Is Narrower Than It Sounds

In a later experiment, students estimating when they’d finish a computer assignment were split three ways. One group simply estimated. A second was asked to recall how similar past assignments had gone. A third was asked to recall a past assignment and describe how this one would play out if it went the same way.

Recalling the past on its own barely moved anything: 38.1% of that group finished within their own estimate, against 29.3% of the group that did nothing special. The third group hit 60.0%, and their average estimate landed within a tenth of a day of what actually happened.¹

That third instruction did two things at once. It pulled the target date out of the students’ own summary of past projects, and it built the expected snags from problems they’d actually run into before. The researchers combined the two deliberately and say they can’t tell which half did the work.¹

Two limits, because this is easy to oversell. The estimates were less biased but no more accurate in absolute terms, off by roughly two days either way, and the researchers are explicit that the exercise gave nobody greater insight into the prediction they were making. What improved was how often people landed on the date they’d named. It also didn’t change when anyone finished: all three groups delivered at about the same time.¹

Action Step: Before the next date goes into your plan, name one past project that resembles it and write down how long that one actually took, start to finish. Then say out loud how this project goes if it goes the same way.

Get a Second Estimate Before You Commit

Hand the same task to somebody outside it and they’ll give you a longer number. In the last version of the study, students estimated when they’d finish the computer assignment while other people estimated it for them, working only from written information about the student and the task. The students averaged 5.5 days, the outsiders averaged 8.5, and the work took 6.8.¹

So the outsiders overshot. Their estimates ran long by about 1.7 days, and in absolute terms they were no more accurate than the students’ own.

What changed was the direction of the error, and direction is what a plan lives on. Fewer than a third of the students finished inside their own estimate, while around two-thirds of the outsiders’ predictions were met.¹

The difference came from what each side looked at. Nearly half the outsiders brought up the student’s own history of finishing things, and among the students themselves, 2.4% did.¹ The outsiders were also far more likely to raise something that might go wrong.

One group of outsiders saw the student’s own plan and the student’s own predicted date, with nothing about how past projects had gone, and still said 8.0 days against the student’s 5.5.¹

Try This: Ask someone outside the project for a date before you put yours into the plan. Then treat the two numbers as the edges of a range. If they’re far apart, the gap tells you how much of your estimate is a story.

Communicate the Plan So People Can Act On It

A plan that lives in one person’s head is a private opinion about the future. The better you can lead and communicate with your team, the further it travels. Four things matter most.

1. Give every key result an owner. Name the department, team, or person responsible for each one, and say it out loud so nobody is guessing whose it is. If you’re planning solo, decide instead what you’ll hand off to someone else.
2. Set the check-in schedule before the work starts. Decide how often you’ll review each objective and put the dates in the calendar now. Use those check-ins to look at the metrics, name what’s stuck, and change tactics when the numbers say so.
3. Set real interim deadlines, and attach something to them. In that same research, the thing that moved when people actually finished was the deadline they’d been handed, and meeting it was a condition of getting credit: a one-week deadline produced completion in 4.3 days on average, while a two-week deadline stretched it to 9.1.¹ Estimating better changes what you predict. A deadline that carries a consequence changes when the work lands.
4. Celebrate the milestones. When the team reaches an OKR, say so. Acknowledging people’s hard work tends to help morale, and it can make the next check-in something people show up to willingly.

Then revisit the objectives often enough that people can say them back to you without looking.

Watch Out for These Pitfalls When You Plan

Three failures show up again and again, and all three are cheaper to prevent than to repair.

Nobody Outside the Room Saw It Coming

A leader writes the plan, and the people it lands on find out when it’s finished. What comes out of that is a plan that’s internally consistent and disconnected from how the work actually runs.

Bring representatives into the drafting, while the plan can still change. A short workshop at each phase is cheaper than finding the mismatch a quarter later.

The Analysis Got Skipped

Going straight to objectives without an honest read on where you stand feels efficient, and it usually isn’t. The frameworks above exist because the questions they ask are the ones people skip when they’re in a hurry.

Action Step: Before you draft anything, run one of SWOT, PESTLE, or the Balanced Scorecard properly, and write the answers down somewhere the team can see them.

The Plan Never Reached the People Doing the Work

Even a well-built plan only works if the people executing it know what they’re responsible for. Your business is a collection of people, after all. It runs better when everyone is clear on their role and the team is pointing in the same direction.

So spell out who owns what and by when, put it where people will actually look, and update it when things change.

Frequently Asked Questions About Strategic Planning

Strategic Planning Takeaways

1. Run one framework properly. Pick SWOT, PESTLE, or the Balanced Scorecard, answer its questions in writing, and leave the other two alone.
2. Write the key result before you write the plan to hit it. For every objective, name the number that would prove it happened and the date it happens by.
3. Pull a comparable project out of the archive. Before you commit to any date, find the closest thing you’ve already delivered and look up how long it really took.
4. Take the date from that project, and list the snags you hit last time. Doing both together is what produced unbiased estimates in the research; recalling the past in general barely helped.
5. Get a second estimate from outside the project. Treat their number and yours as the edges of a range, and read the gap as a measure of how much of your date is a story.
6. Put interim deadlines in the calendar with owners’ names attached. Give each one a consequence that lands if it slips, because that’s the part the research showed moving real delivery dates.
7. Book the check-ins now, while the dates are still easy to change.

Before the next plan goes out, spend ten minutes on a single number: find the closest project you’ve already finished, and check what it actually cost you.

Then go work on the thinking that feeds the plan in the first place, with our guide to strategic thinking skills.