Commentary in management reporting: how to write the story behind the numbers (2026)

Commentary in management reporting: how to write the story behind the numbers

The story behind the numbers: what belongs in it, how to structure it and how it gets a fixed place in the monthly process.

What is management commentary in a management report?

The management commentary is the written story behind the numbers of the management report: per main topic what happened, why it happened and what management is doing about it. It is the part that connects loose tables into one readable picture of the month.

‍

In the structure of a good management report, the commentary is the seventh and final fixed section — but in reading order often the first thing the reader turns to. Management and investors do not read the report to admire the tables, but to understand what is going on and which decisions are on the table; exactly that is what the commentary delivers.

‍

The commentary does not stand alone. It is fed by the variance analysis, which determines which deviations deserve a story, and it substantiates the summary up front, which condenses the conclusions into three to five messages. Summary, variance analysis and commentary form one chain: the analysis selects, the commentary explains, the summary concludes.

‍

Why this section deserves the attention: in practice it is the weakest part of most reports. The numbers side is usually in order — certainly when automated — but the story gets written in the last minutes before the deadline, describes what the table already shows and avoids the painful topics. Whoever takes the commentary seriously lifts the whole report to another level — at zero extra system cost. The rest of this article covers, in order, the content, the structure, the writing style, the length and place, the pitfalls and the spot in the monthly process.

What belongs in the commentary — and what does not?

The commentary covers only what needs a story: the material deviations, the decisions on the table and the developments that will hit the coming months. Everything else is already in the tables — and may stay there.

‍

Include. The explanation of every deviation above the variance analysis threshold, with the timing-or-structural label. The decisions being asked of management, with the trade-off attached. The outlook: what this month’s events mean for the rest of the year, tied to the latest estimate. And the developments outside the numbers that the reader needs to know — a canceled contract, a departed key person, a shifted product launch — precisely because they only show up in the numbers next month.

‍

Leave out. Descriptions of what the table already shows (“revenue grew 8%”) — that is reading aloud, not explaining. Explanations of lines below the threshold: whoever explains everything buries what matters. Financial jargon without explanation, because the reader is not necessarily financially trained. And vague language that dodges responsibility — “margin was under pressure” says nothing; who or what pressured it, and what are we doing about it?

‍

The litmus test for every paragraph: would the reader take a decision differently, or ask one question fewer, if this paragraph were not there? If not, it can go. The commentary is not a record but a filter — and its quality sits, just as with the variance analysis it draws on, mainly in what is not there. So dare to cut: the reader will not notice, and that is exactly the point.

How do you structure the commentary: the what-why-action triptych

Every passage in the commentary follows the same triptych: what happened (one sentence, with the amount), why it happened (the cause, with the timing-or-structural distinction) and action: what management is doing about it, with an owner and a deadline. That fixed form forces completeness without length — and makes the commentary scannable for the reader with little time.

‍

The difference is clearest in an example. Not like this: “Revenue came in at EUR 1.42 million, 8% below budget. Margin also developed below expectations.” This tells the reader nothing the table does not already show, and ends without cause or action.

‍

Like this instead: “Revenue fell EUR 120,000 short of budget. That comes almost entirely from two items: customer X pushed the planned order into August (timing, EUR 90,000) and price pressure in product group Y persisted (structural, EUR 30,000). X’s order has since been confirmed; for Y, sales is revising the pricing policy and the proposal is on the September agenda.” Three sentences, and the reader knows what happened, how much it weighs, what heals itself and where a decision is needed.

‍

Work per main topic — revenue and margin, costs, working capital and cash, and where relevant the group’s entities — and keep to at most three triptychs per topic. More stories rarely means more insight; it usually means the threshold in the variance analysis was not strict enough. The triptych is also a control instrument: any passage without an action part automatically raises the question whether anyone has thought it through.

How do you write for management and investors?

Write for the smart reader without a financial background: someone who understands the business perfectly well, but does not think in ledger accounts. That starting point sets the tone of every good commentary.

‍

Concretely, that means five things. Active sentences with an owner: not “it was decided to shift the campaign” but “marketing is shifting the campaign to Q4”. Rounded amounts: EUR 120,000 reads, EUR 119,847.23 does not — the precision is already in the table. Consistent terms: the same words for the same items every month, so the reader never has to guess whether “gross margin” means the same thing as last month. Explained jargon where it is unavoidable: “DSO — the average number of days customers take to pay — rose to 52”. And one voice: even when several people contribute, one writer edits the whole, so the commentary reads as one story.

‍

The most important principle is honesty, and it must be symmetrical. A commentary that celebrates windfalls at length and buries setbacks in a subclause loses exactly the trust it is meant to build — investors read straight through it, and management takes decisions on a filtered picture. The rule is simple: the setback gets at least as many words as the windfall, and every setback ends with an action instead of an excuse.

‍

Finally, show discipline in the outlook. Promises that slide month after month (“recovery expected in Q3”) undermine credibility faster than the setback itself; tie expectations to the latest estimate and name it explicitly when an earlier expectation did not come true — that restores more trust than quietly moving the date ever does.

How long should the commentary be and where does it go?

The norm: three to five paragraphs per main topic, and one to two pages in total — fitting within the ten pages that guide the whole report. Whoever needs more is usually explaining too many lines or too verbosely; the detailed substantiation belongs in the appendices, not in the story.

‍

For placement, a two-tier model works best in practice. Brief callouts on the numbers pages themselves: two or three lines under the P&L, the balance sheet and the cash flow that interpret the most striking movements on that page — so the reader does not have to flip between table and text. The coherent story as its own section, after the variance analysis and the KPIs: there the loose callouts come together in the triptych per topic, with the decisions and the outlook.

‍

The division of labor with the summary up front is strict. The summary gives the conclusions — the three to five messages of the month, with the charts — and is written for whoever reads one page. The commentary delivers the substantiation for whoever reads on: why those messages hold and what sits behind them. Whoever lets both tell the same story forces the reader to read everything twice; whoever lets them cross-reference (“see revenue commentary”) keeps both short.

‍

And keep the layout identical every month: the same topics, the same order, the same place. The reader who knows where the margin story lives finds it in ten seconds — and for a report that returns monthly, that reading ease is worth more than variety or literary ambition. The story may surprise; the format should not.

How do you avoid the well-known pitfalls?

Four patterns recur in almost every weak commentary — and all four can be caught with a simple rule. They share one root: they arise not from incompetence but from time pressure and discomfort — which is exactly why fixed rules work better than good intentions.

‍

Retelling the table. “Costs rose 6%” is not commentary but a caption. The rule: every paragraph must add something that is not in the table — a cause, context, an action. If it does not, the paragraph is redundant.

‍

Structural optimism. Framing setbacks as incidents and windfalls as trends is the fastest route to a credibility problem. The rule: the timing-or-structural label comes from the variance analysis, not from wishful thinking — and a deviation that has been “incidental” for three months is by definition structural.

‍

Copy-paste commentary. Whoever lightly edits the same template story every month trains the reader to skip the section. The rule: write each month from the question “what really happened this month?” — and dare to drop a topic when there is nothing to report. An empty paragraph is more informative than a recycled one.

‍

The commentary as an afterthought. Whoever crams the writing into the last day delivers the weakest section at the most important moment. The rule: reserve fixed writing time right after the variance analysis, and plan a short review by someone who did not produce the figures — they read what is actually there, not what was meant. How that fits the monthly rhythm follows in the next section.

How does the commentary get a fixed place in the monthly process?

A good commentary is not a matter of writing talent but of calendar: the story needs time, and that time has to come from somewhere. In a manual reporting process, almost the entire lead time goes into collecting and reconciling figures — and the writing gets squeezed out.

‍

That is why a better commentary starts with an automated data feed. With Finstack, the actuals from packages such as Exact Online, AFAS and Twinfield are ready on day one — refreshed daily, for groups including automatic intercompany elimination, clickable through to the source booking. The days that come free are exactly the days the variance analysis and the commentary need; how that shift works is covered in automating management reporting.

‍

The writing process itself gets three fixed moments in the closing calendar. Right after the variance analysis: writing the triptychs per topic, while the analysis is fresh. A day later: the review by a second reader and the final edit toward one voice. And at delivery: sharpening the summary up front based on the final commentary — in that order, because the conclusions follow from the story and not the other way around.

‍

AI can deliver the draft work in this process: a first version of the triptychs based on the figures and the analysis. Judgment stays human — the context beyond the numbers, the curation and the final edit toward management and investors; a dedicated article on management commentary with AI follows within the AI cluster. And sharing happens without email rounds: stakeholders get access to the dashboards and reports in Finstack, with access rights per user. Finstack starts from EUR 39 per month: live in 5 minutes, set up within a day.

The 3 most common mistakes in management commentary

Three patterns we see again and again in commentaries. Each makes the story weaker than the figures beneath it; each is preventable with a fixed rule.

Sources and provenance

• Finstack — Reporting & insights (dashboards, Excel/Sheets sync): finstack.io/solutions/reporting-insights
• Finstack — Integrations with Exact, AFAS, Twinfield and other packages: finstack.io/solutions/integrations
• Finstack Help Center — Sources & connections (sync frequency, security): help.finstack.io
• Finstack — Pricing (from EUR 39/month): finstack.io/pricing

Last reviewed: 25 July 2026 · Next review: October 2026