A closer look at executive summaries: the seven things a reader hunts for, why you write it last

For a large share of readers, the summary is not the first part of the plan. It is the whole of the plan, because it is the only part they will finish. Everything after it gets read by people who have already decided the thing is worth more of their time, and the summary is what makes that decision.

Which means it is the one page that has to work alone, in someone else's office, without you there to explain it.

What to take away

Summary, not introduction

An introduction prepares the reader for what comes next, while a summary delivers the findings and lets the reader decide whether to check the work. Almost every weak executive summary is an introduction under the wrong title.

They open with the industry, the company's founding, or a trend everyone in the room already knows.

The test is simple. If a reader stopped after your summary, would they know what you are asking for, what they get, and why it works? If not, you have written a preamble.

The conclusion goes first. What the business does, for whom, what you want from this reader, and what happens with it. Then the evidence, briefly. Background, if it earns its place, goes near the end.

What a reader is looking for

Seven things, and a reader will find them in some order whether or not you provide them.

That last item is the one people leave out, and it is the one that builds the most credibility. A summary with no acknowledged risk reads as either naive or evasive, and an experienced reader will supply the risk from imagination if you do not. Naming it lets you frame it.

Write it last, from the finished plan

The summary is generated from the plan, never the other way around. That is not a stylistic preference. It is the only way the numbers in it match the numbers in the sections behind it. A mismatch here is fatal in a way it is not elsewhere, because this is the page a reader checks against everything else.

The consistency joins that get tested across a whole document are set out in the guide to what a plan is for and who reads it; the summary is where a break in any of them becomes visible first.

The practical method: finish the plan, then pull the single most important sentence out of each section, then cut what a reader would not act on. What remains is usually close to the right length, and it is all sourced, because every line came from somewhere with working behind it.

Then read it as the reader. Not as the author checking that everything is mentioned, but as a person deciding whether to spend an hour on the rest of it.

How summaries fail

The mismatch row deserves emphasis. A reader who finds one number in the summary that disagrees with the same number twenty pages later stops evaluating the business and starts auditing the document. That shift is very hard to reverse.

Length, and the versions you need

One page is the working target for most plans, two if the business is genuinely complicated or the reader has asked for detail. The constraint is not a rule about pages. It is that a reader has to be able to hold the whole thing at once.

Three versions are worth having, all generated from the same content.

The page goes at the front of the document. The paragraph goes in an email and must survive being read on a phone. The spoken version, about a minute, is what you actually say when someone asks what the business does.

If the spoken version differs from the page, the page is probably overwritten. The spoken one is usually closer to the truth.

Tuning it to the reader

The content stays the same. The order changes, and the order is the message.

A lender reads for repayment, so cash timing and what secures the loan move up. The routes differ in what they ask you to show, as the federal summary of funding types sets out. An equity reader is looking for how large this could become, so the market and the team move up.

A landlord or major supplier wants to know you will still be trading next year, so working capital and the runway come first. A licensing body wants the compliance position, stated precisely, with the right authority named, which means the permit and license position moves to the front.

Same facts, four documents. Reordering costs an hour. Sending the wrong order costs the meeting.

The summary you write for yourself

There is a version of this page nobody outside the business ever sees, and it is the more useful one. It says what the business is trying to do this year, what has to be true for that to work, and what you would see first if it were going wrong.

It reads differently because it persuades no one. It can call the price a guess, say one customer is most of revenue, and note the second hire depends on a month that has not happened yet.

That does not belong in a lender document in those words. Yet you need all of it to run the year.

Write that version first and derive the external one from it. Doing it the other way round produces a polished page and a business whose owner cannot explain their own forecast when asked.

Common questions

Should the summary include financial projections?

The few figures that matter, not the tables. Revenue at a stated point, when the business covers its costs, the funding required, and the cash gap it covers. Whole tables belong in the financial section, where they can be examined.

Can we use it as a standalone document?

Yes, and it is worth writing so that it survives being separated from the plan. That means no references to sections the reader cannot see, and no sentences that only make sense with the rest of the document in hand.

How much of the team's history belongs here?

Only what predicts performance in this business. Relevant operating experience, a track record with something comparable, or a specific capability the plan depends on. A full biography belongs in the team section, and putting it here signals that the business case was thin enough to need padding.

What if the strongest evidence is small?

Say the small thing accurately. A handful of paying customers described precisely is more persuasive than a market projection, because it is checkable and it is yours. Readers discount large unverifiable numbers automatically; they cannot discount a fact you can demonstrate.