Heads up — the easy free way to put an AI-built page online usually makes it public and searchable on Google. Pagelive keeps yours private →
+ +

Template · 10 min read

QBR template: a quarterly business review clients actually read

Most quarterly business reviews are slide theater: forty minutes of a deck being read aloud to a client who is checking email, followed by "great, see you next quarter." The work was real, the meeting wasted it. This is a different shape — a short, verdict-first document the client reads in three minutes, sent as one permanent link that carries the whole relationship quarter after quarter. The full template is below, section by section with the reasoning, plus a prompt that makes AI build the document for you.

+ +

What a QBR is actually for

Strip the acronym and a QBR is one thing: the recurring moment where the client decides you're worth keeping. The weekly calls cover the work; the QBR covers the relationship. Its real audience is usually not your day-to-day contact — it's the sponsor above them, the person who signs the renewal and wasn't in any of the weeklies. They have one question: are we winning, and is this partner the reason?

That reframing changes what goes in the document. A status report lists activity. A QBR renders a verdict against the goals both sides agreed to, translates the numbers into business outcomes, admits what didn't work, and proposes what happens next. Activity appears only in service of that story — which is why the most common QBR failure is the activity dump: twenty slides proving you were busy, zero sentences saying whether it mattered.

One honest caveat before the template: a QBR is only as good as the goals set at the start of the quarter. If you never agreed on targets, the scoreboard section below has nothing to score — in that case your first QBR's job is smaller and different: propose the goals for next quarter and get them agreed. The template still works; the scoreboard just starts life as a promise instead of a verdict.

+ +

The QBR template, section by section

Eight sections, in a deliberate order — verdict first, detail last. The sponsor should be able to stop reading after any section and still have the truth.

  1. Verdict. One sentence, before anything else: "Q3: pipeline goal beaten by 18%, retention goal missed — the miss and the fix are in section 5." Writing this sentence is the hardest part of the whole document, which is exactly why it earns the top slot. If you can't write the verdict, you don't know how the quarter went — and neither will the client.
  2. Scoreboard. The three to five goals you agreed on, each with target, actual, and a green/red delta. No vanity metrics, no charts of things nobody asked for. A number without a target next to it is decoration.
  3. What this means. Three or four plain-language sentences translating the scoreboard into business outcomes — money, customers, risk. This is the section that separates a QBR a sponsor reads from one they skim: metrics are your language, outcomes are theirs.
  4. What we did. The quarter's work in five or six bullets, each tied to the goal it served. Brief on purpose — this section proves the retainer, it doesn't star in the meeting.
  5. What we learned. Include one honest miss and what changed because of it. This is counterintuitive and load-bearing: a QBR with no admitted miss reads as marketing, and sponsors discount marketing. The admitted miss is what makes the wins credible.
  6. Next quarter. Proposed goals with numbers and owners — theirs and yours. This turns the meeting from a review into a decision, and it quietly re-sells the engagement every quarter without a pitch.
  7. Asks. What you need from the client: a decision, access, an introduction, a budget call. QBRs that skip this train clients to treat the vendor as vending machine rather than partner.
  8. Appendix. Everything the analyst in the room might want — full data, methodology, links to dashboards. Present so nobody can call the document thin; separated so nobody has to read it.

Two structural rules across all eight: stamp a "data through" date under the title so "is this current?" never gets asked, and keep metric definitions identical quarter to quarter — quietly redefining "conversion" between reviews destroys the one thing a recurring document builds, which is a trend the client trusts.

+ +

Deliver it as one permanent link per client

pagelive.site/p/acme-qbr same link
Q1 review v1 · Jan
Q2 review v2 · Apr
Q3 review v3 · Jul live

re-publish ↻ the bookmark never changes

acme-qbr — this week
  1. Mon 09:12

    Link sent to Dana

  2. Tue 14:03

    Opened · 4m 38s on page

  3. Thu 08:51

    Opened again · second viewer, new city

  4. Thu 09:10

    Your move: they are reading it right now

Here's the delivery pattern that makes the template compound: each client gets one URL, and every quarter's QBR replaces the last one at that same address. The sponsor bookmarks it once. There's no QBR_Q3_final_v2.pdf hunting, no "can you resend the deck," and by the fourth quarter the link itself is an asset — the standing address of the relationship, with the appendix linking back to past quarters if you keep them.

The second half of the pattern is the read signal. Send the document two or three days before the meeting, and a tracked link tells you whether the sponsor opened it and roughly how long they spent. That changes how you run the room: a sponsor who read it gets a discussion; one who didn't gets a five-minute walkthrough first, without anyone being embarrassed. It's the pre-read culture Amazon made famous, minus the awkward silent-reading ritual — the reading happens before the room.

The honest limits, because overselling tracking backfires: a link open tells you the link was opened, not who opened it — a forwarded link, an assistant, or an email scanner all count. Treat it as a prompt for how to open the meeting, not evidence for an argument. And for anything with real numbers in it, protect the page with a password rather than relying on an obscure URL.

+ +

Build the document with AI (prompt included)

You don't need to hand-build this in slides every quarter. Give an AI the structure above plus your actual numbers and it will produce a clean, self-contained HTML document in one pass — you review it, fix what's off, and publish. The prompt below encodes the whole template, including the two rules that matter most: verdict first, and never invent a number.

Build it with AI

Copy, fill in the brackets with your quarter's real data, paste into Claude or ChatGPT:

qbr-prompt.txt
You are producing a quarterly business review (QBR) as a single, self-contained HTML file.

CONTEXT — replace everything in brackets:
- Client: [client name]
- Period: [Q3 2026]
- Prepared by: [your name / company]
- The goals we agreed for the period, each with target and actual:
  [goal — target — actual, one per line]
- What we did: [3-6 bullets of the work delivered]
- One thing that didn't work: [the honest miss + what changed because of it]
- Proposed goals for next period: [list]
- Asks of the client: [decisions, access, or input you need from them]
- Brand: [primary hex color] + [logo URL, or "text wordmark"]

REQUIREMENTS:
- Open with a one-sentence verdict on the quarter before anything else, then a scoreboard table of goals vs. actuals with green/red deltas.
- Follow the scoreboard with a "What this means" section in plain business language — outcomes and money, not tool metrics.
- Keep every number exactly as I gave it. If something is missing, leave a clearly marked [TBD] — never invent data.
- Sections in this order: Verdict · Scoreboard · What this means · What we did · What we learned · Next quarter · Asks · Appendix.
- Self-contained HTML: inline CSS, no external scripts, readable on a phone, printable. Stamp "Data through [date]" under the title.
- Tone: direct, no filler, no adjective where a number will do.
  1. 1 Paste the prompt (with your details filled in) into Claude or ChatGPT. You get a finished, self-contained HTML document — review it, correct anything, iterate until it's right.
  2. 2 Publish it as a permanent tracked link: with the Pagelive connector in Claude, say "publish this page" and the link comes back in chat — next cycle, "update it" replaces the content at the same URL. Or paste the HTML at app.pagelive.io.

The output is a document, which means everything in it is yours to sign off on. The one thing AI reliably gets wrong in business reviews is inventing plausible-sounding context around the numbers — the prompt's [TBD]-never-invent rule contains that, but read the "What this means" section with particular care before it goes to a client. That narrative is your judgment; the AI only drafts it.

+ +

The mistakes that kill QBRs

  • Reading the deck aloud. If the meeting is the first time the client sees the material, you've guaranteed a passive audience. The document goes out before; the meeting is for the discussion the document provokes.
  • The activity dump. Listing everything you did, ordered by effort rather than impact. The sponsor doesn't renew because you were busy.
  • No verdict. Forty slides that never risk a sentence-long answer to "so, was it a good quarter?" — forcing the client to assemble the verdict themselves, usually less charitably than you would have.
  • Surprise bad news. The QBR is the wrong room to reveal a miss for the first time. Bad news travels in the weekly the moment it's known; the QBR is where you show what you did about it.
  • Shifting definitions. "Qualified lead" meaning something new each quarter, so no trend survives. Clients notice, and it reads as hiding.
  • A deck nobody can find later. Six months on, someone asks "what did we agree in the Q2 review?" and the answer is an attachment archaeology dig. A standing link makes the QBR citable — which is precisely what an agreement wants to be.
+ +

What to build it in

Slides (PowerPoint, Google Slides, Pitch) remain the default because meetings default to screens. They're fine when you'll genuinely present — but they push you toward theater, they're painful to pre-read on a phone, and every quarter mints another orphaned attachment.

Docs and wikis (Google Docs, Notion) pre-read well and keep history, and if the client already lives in your shared workspace they're a solid choice. Sharing outside that workspace is where they get awkward — permissions, logins, and no real read signal beyond "someone viewed."

Customer-success platforms (Gainsight, Vitally and peers) have QBR and EBR tooling built in, wired to live product data. Right answer for a CS team running dozens of accounts on one stack; overkill for an agency, consultancy, or fractional exec with a handful of clients and no seat-based budget to spend.

An HTML document at a permanent tracked link is the pattern this post describes, and it's the one that fits the AI workflow above. Pagelive is built for exactly this: publish the QBR as a private page, optionally password-protected, re-publish next quarter to the same URL, and see whether it was opened before the meeting. The honest limits: there are no data connectors — you (or your AI, or a script) bring the numbers each quarter — and "current" means "as of your last publish." If you need live-refreshing dashboards, that's a BI tool's job; plenty of teams run both, the dashboard for poking at data and the QBR link for the story and the decision.

+ +

Frequently asked

A QBR is a recurring meeting-plus-document where a vendor, agency, or consultant reviews the quarter with a client: what was promised, what happened, what it meant for the client's business, and what's planned next. Done well it's a renewal instrument, not a status report — the audience is the executive sponsor who wasn't in the weekly calls.

Related: the always-live client report · did the client open my proposal? · how to write an executive summary

+ +

One QBR link per client. Updated every quarter, read before every meeting.

Build the review with AI, publish it as a permanent tracked link, and know the sponsor read it before you walk into the room. Free for 5 pages, no card.