An executive summary page is defined by what you leave off it. Every other page in your report answers “what happened”. This one answers “what should we do about it” — and it only works if a reader can get that answer without scrolling, clicking or asking you a question.
This covers what earns a slot on the page, how to build it in Data Studio, and the part nearly everyone skips: writing the summary itself.
What an executive summary page is
It’s a single report page, placed first, that states the headline performance, the one trend that explains it, what it means, and what happens next. Three or four numbers, one chart, a written summary, a short list of actions.
The distinction that makes it work is the one between exploratory and explanatory reporting. The rest of your report is exploratory — it lets someone dig. The summary page is explanatory: you have already done the digging, and you are handing over the conclusion. Confusing the two is how a consultant loses authority, because a dashboard presented as an analysis implies you had nothing to add.
So the test isn’t “is this metric interesting”. It’s “would I say this out loud in the first minute of the meeting”.
Build it on day one, not when you need it
The usual approach is to build the detail pages first and add a summary later, when somebody senior asks for one. That’s backwards, and the reason is a scheduling problem rather than a design one.
If your day-to-day contact is reasonably senior, they will need a holistic view anyway. And the moment they introduce you to their boss — which happens with no notice, usually in a meeting you didn’t schedule — you need the page to already exist. Building it under time pressure, from a report designed for a specialist, produces exactly the cluttered page this article is about.
The architecture that works: one holistic, top-level page covering every channel, which links out to project-level pages underneath it. The summary is what the CEO or head of marketing sees. The detail is one click away for anyone who wants it.
The five-second rule
Here is the test the rest of this article keeps returning to: if a reader can’t state the headline within about five seconds of the page loading, the page has failed.
Not “can they find the number” — can they say what it means. Those are different, and the gap between them is where most summary pages die. A page with eleven scorecards contains the answer somewhere. A page with four scorecards and a sentence is the answer.
Applied honestly, this rule removes about half of what people put on a summary page, and it settles arguments about whether something earns a slot faster than any principle about clarity or minimalism.
What earns a slot
Three or four numbers, and no more
Each one needs a comparison. A metric without a previous period is a fact, not information — nobody can tell whether 31,900 sessions is good news without knowing what it was last month.
Pick numbers your reader is accountable for. This is where output metrics belong: revenue, orders, leads, conversions — the backward-looking measures of whether the thing worked. Input metrics (pages published, links earned, issues fixed) matter enormously, but they belong on the project pages, because an executive can’t act on them directly.
The rule of thumb that catches most over-stuffing: if you can’t say why a number is on the page in one sentence, it isn’t.
One chart — and it’s almost always a time series
One. Not four. The summary page shows the shape of the story; the detail pages show the breakdown.
A time series is usually right because the executive question is nearly always directional — is this getting better or worse, and since when. If your one chart is a bar chart, check whether the real question is actually “which channel” rather than “what’s happening”, because that’s often a sign the page is trying to do a detail page’s job. The chart types post covers picking the right one from the shape of your data.
The written summary — the part everyone skips
This is the single biggest differentiator, and it’s covered properly further down, because it’s the hardest part and it deserves more than a bullet.
Short version: a page of charts with no words is a stats dump. Sending one out, however well-designed, isn’t storytelling — it’s delegation of the interpretation to somebody less equipped to do it than you.
What happens next
Owned, dated actions. Not “consider improving internal linking” — who is doing what, by when, and what you expect it to move.
It is your job as the consultant to supply these. Leaving a client to work out their own next steps from a dashboard is the most common way a good analysis produces no change at all.
How to build it in Data Studio
1. Make it a separate page, and put it first
Add a new page (Page → New page) and move it to position one. Give it a name that says what it is — “Summary” or “This month at a glance” — because that name appears in the report’s navigation.
If your report has more than about three pages, set the navigation to a left-hand menu rather than tabs, in Theme and layout → Layout. Tabs run out of room and start truncating exactly when the report gets big enough to need them.
2. Scorecards with a comparison date range
Add a scorecard per headline metric, then in Setup set the Comparison date range to previous period or previous year. This is the setting that turns a number into information, and it’s off by default.
Two things worth knowing:
- Previous period or previous year is a real choice. For anything seasonal — retail, travel, education — previous period will mislead. December against November is not a comparison anyone should act on.
- A sparkline earns its space here. A scorecard can carry one, which gets you the headline and the shape in the footprint of a headline.
3. One time series, with the target on it
Add the chart, set its date range explicitly rather than leaving it on Auto — Auto means the connector’s default, which differs between connectors and is a common cause of two charts on one page showing different periods.
If you have a target, put it on the chart as a reference line. A trend against a target answers “are we on track” in one glance; the same trend without one only answers “did it go up”.
4. The commentary text box
Insert a text box beside the chart, not underneath it. Underneath, it reads as a caption and gets skipped; beside it, it reads as the point.
There’s a real objection to putting commentary in a dashboard at all, and it deserves a straight answer. The dashboard updates automatically; your commentary doesn’t. On the first of the month you’re showing last month’s notes against this month’s data, which is worse than no commentary.
Two things fix it:
- Keep the commentary in a Google Sheet with a date column, and blend it into the report as a data source. Then updating the story is editing a spreadsheet row, not editing a report.
- Publish your update cadence on the page itself. A small text component saying “commentary updated on the first and third Monday of each month” converts a stale-looking dashboard into a documented one. It sets the expectation instead of leaving the reader to discover the gap.
The side effect is worth having: over a year you accumulate a detailed log of expert commentary against specific time points, which is its own asset at renewal time.
5. Page navigation, so people can get to the detail
The summary page works because things are missing from it — which only holds if the missing things are one click away. Link out to the project pages explicitly, either through the report navigation or with buttons on the page.
If the same date control should apply everywhere, select it and use Arrange → Make report-level. The reader’s selection then carries across pages, so setting the summary to last quarter doesn’t leave the detail pages on last month.
Writing the summary itself
This is the highest-value four hundred words on the page, and the part that no competing guide shows an example of.
A sentence pattern that works
What changed → by how much → why → what we’re doing about it.
That’s it. It’s dull as a formula and it works, because it answers the four questions an executive is going to ask in that order anyway.
Three summaries, rewritten
Weak: “Organic traffic was down this month.”
Better: “Organic sessions fell 12% month on month, driven almost entirely by the blog. Two of our top ten landing pages lost position after the March core update. We’ve rewritten both and resubmitted them; we’d expect recovery within six to eight weeks.”
Weak: “Great month for conversions, up 24%!”
Better: “Conversions rose 24%, but 18 points of that came from the brand campaign that ran 8–22 March. Underlying non-brand conversion is up 6%, which is the number to plan against. The campaign ends in April.”
Weak: “Rankings continue to improve across target keywords.”
Better: “We now rank top three for 14 of the 40 target terms, up from nine. Those 14 account for 61% of the tracked search volume, so the revenue impact should be larger than the count suggests. Next is the nine terms sitting at positions four to seven.”
Notice what the good versions do: they name a cause, they separate signal from noise, and they end with something happening. None of them are longer than four sentences.
Numbers in prose
- Round. “Up 12%” is more usable than “up 11.87%”, and nobody is making a different decision at the second decimal place.
- Give the comparison in the sentence. “31,900 sessions” means nothing; “31,900 sessions, up from 28,400” means something.
- Be honest about noise. If a 4% move is within normal variance for that metric, say so rather than narrating it as a trend. Doing this consistently is what makes people believe you when you say something is significant.
When the news is bad
Report it plainly, early, and with a route out.
Two habits cause most of the damage here, and both are about the consultant rather than the data. The first is simply being afraid of the conversation — softening a bad month, or hoping nobody asks. The second is subtler: anticipating how someone will react and shaping the narrative to suit that expectation, either going in combative or quietly downplaying. Both produce a subjective story, and objectivity is the whole basis of being believed next time.
The practical version: if performance is down because of something on the client’s side — a migration that shipped without you, three months of recommendations not implemented, a budget cut — that belongs in the summary. Framed as a fact with a consequence and a fix, not an accusation. Underperformance that sits within the reader’s own control to resolve is the most actionable thing you can put in front of them.
And don’t save it for the quarterly review. A conflict that surfaces once a quarter has had three months to compound.
Executive summary pages by role
The page changes depending on who opens it, and the difference is mostly about what counts as an outcome.
Reporting to a CMO or head of marketing
They care about channel mix, efficiency and whether the plan is working. Lead with the outcome metric they present upward — usually revenue, pipeline or cost per acquisition — and make the one chart show the trend against target. Keep channel breakdowns off this page; that’s what the click-through is for.
Reporting to a founder or CEO
Shorter still. They want to know whether this is working and what it costs. Two or three numbers, one trend, three sentences. Anything requiring a paragraph of setup belongs elsewhere. They will not use a filter control, so don’t build the page around one.
Reporting to a client contact who reports upward
This is the most common case and the one most people get wrong. Your contact isn’t the final audience — they’re going to paste your numbers into their own deck.
Build for that: make the page exportable, write the summary in sentences they can lift verbatim, and give them the comparison figures they’ll be asked for rather than the ones you find interesting. The kindest thing you can do for a day-to-day contact is make them look well-prepared to their boss.
What to leave off
- Pivot tables. They reward digging, and this page is for people who won’t.
- Anything needing a paragraph to explain. If the metric requires a definition, it isn’t a summary metric — put it on a detail page with the definition next to it.
- Daily-granularity data. Day-level detail in a monthly report is noise wearing a chart. Match the granularity to the reporting cycle.
- Position tracking. Not because it’s a vanity metric — it’s genuinely useful — but because it’s mis-audienced here. It belongs on the specialist page. An executive doesn’t act on a rankings fluctuation, and putting it on the summary page invites a conversation about position 4 to 6 instead of about revenue.
- Everything you’re proud of building. The hardest cuts are the charts that took longest. They still belong on a detail page.
Taking the page into a deck
Most summary pages end up in a slide sooner or later. Two things make that painless:
- Build at a page size that exports cleanly. If the report is mainly read as a PDF or pasted into slides, set the canvas to A4 or a 16:9 size in Theme and layout before you build — changing it later doesn’t reposition anything for you.
- Remember the export captures what’s rendered. A table showing ten of four hundred rows exports as ten rows. On a summary page that’s rarely a problem, which is another argument for keeping tables off it. More on export and sharing.
The mistakes
Showing everything. The most common failure, and it usually comes from a good instinct — wanting to demonstrate the work. Demonstrate the work on the detail pages.
Starting from the chart. Deciding you want a nice-looking dashboard and then finding metrics to fill it. Start from the decision the reader has to make.
No context on any number. Comparisons, targets and denominators are what convert a metric into a judgement. A number on its own is something the reader has to interpret, which is the job you were hired for.
No takeaway. A page that ends on a chart ends on a question. Put the answer on it.
A narrative the data doesn’t support. The one genuine risk of getting good at this. A compelling story built on a correlation is more dangerous than a boring one built on a fact, precisely because it’s more persuasive. If you’re inferring rather than measuring, say which one you’re doing.
A note on the name
Google renamed the tool Looker Studio and then renamed it back to Data Studio, so guides under either name describe the same product. We use Data Studio throughout because that’s the current name. Looker without “Studio” is a different product entirely — Google’s enterprise BI platform, with its own visualisations and its own modelling layer.
Frequently asked questions
What is an executive summary page in a dashboard?
A single report page, placed first, that states the headline performance, the one trend that explains it, what it means and what happens next. Typically three or four scorecards with comparisons, one chart, a short written summary and a list of actions. It’s defined by what you leave off — everything requiring exploration belongs on a detail page.
How many metrics should be on an executive dashboard?
Three or four headline numbers, each with a comparison. If you can’t say in one sentence why a number is on the page, it doesn’t belong there. The test is whether a reader can state the headline within about five seconds.
What’s the difference between an executive summary page and the rest of the report?
The rest of the report is exploratory — it lets someone dig into the data. The summary page is explanatory: you’ve done the digging and you’re presenting the conclusion. Mixing the two up is the most common structural mistake in client reporting.
How do I keep dashboard commentary up to date when the data updates automatically?
Two things. Keep the commentary in a Google Sheet with a date column and blend it into the report, so updating it means editing a spreadsheet rather than the report. Then publish your update cadence on the page — a text box saying commentary is updated on the first and third Monday sets the expectation instead of leaving readers to find stale notes.
Should I build one dashboard or one per channel?
One holistic top-level page covering every channel, linking out to project or channel pages beneath it. Most client contacts are generalists who want to see everything; the specialists who want depth can click through.
When should I build the executive summary page?
At the start of the engagement. The day your stakeholder introduces you to their boss you won’t have time to build one, and a summary assembled under pressure from a specialist’s report is exactly the cluttered page you’re trying to avoid.
Should I use previous period or previous year for comparisons?
Previous year for anything seasonal — retail, travel, education — where a month-on-month comparison will mislead. Previous period for steadier businesses and for spotting recent change. Pick one and stay consistent, or your reports won’t be comparable to each other.
How long should the written summary be?
Three or four sentences per point. Follow the pattern: what changed, by how much, why, and what you’re doing about it. Longer than that and it stops being a summary.
Should rankings go on an executive summary page?
Usually not. Position tracking is genuinely useful, but it’s aimed at the wrong reader here — an executive can’t act on a rankings fluctuation, and including it steers the conversation away from revenue. Put it on the specialist page.
Can I export an executive summary page to a slide deck?
Yes, as a PDF or an image. Set the canvas size before you build if the page is mainly destined for slides, since changing it later doesn’t reposition anything. Note that exports capture what’s rendered, so a paginated table exports only the visible rows.