For agencies

Ten clients, one collection.

Citino measures categories, not accounts. The daily run for Warehouse Management Software · US is a single collection — 54 prompts, 1,785 answers in thirty days — and every one of the 17 brands in that category is measured by it, whether or not anybody in it is paying.

Put a client on a seat and they are reading a series that already exists, with its history behind it. Put nine more on seats and the collection still runs once.

Read the methodology →
Meridian Growth is fictional, like every brand on this site.
Roster · Meridian GrowthIllustrative
Client seats3 of 10 used · 1 opening
ClientCategoryStatus
TallyfoldWarehouse Management Software · USlive · 8 of 50
HarbourlineField Service Management · USlive
Kestrel SupplyHR & Payroll Software · USlive
FrostlineNew categorybuilding · 5–10 days
Categories collected once · 54 prompts daily · not metered
Client portal · TallyfoldWhite-label
Visibility10.3%±3.0
Rank9thof 17
30-day Δ+1.7within noise
“…and Tallyfold has a lighter setup that smaller 3PLs tend to prefer.”
Roster · Meridian Growth · 3 of 10Illustrative
SeatClientCategoryCustom
1TallyfoldWarehouse Management Software · US8 of 50
2HarbourlineField Service Management · US— of 50
3Kestrel SupplyHR & Payroll Software · US— of 50
QueuedFrostlineBuilding · opens in 5–10 business days—
4–10OpenAny category
Category prompts · 54 daily · not metered on any seat
Tallyfold · seat 1custom prompts8 of 50category prompts 54not metered
The unit

A seat is a client, not a credit.

One seat is one client brand. It carries:

  • their category, with the whole collected history from the first day they log in
  • every category prompt in it, daily, never metered — 54 of them here
  • five competitors tracked alongside them
  • 50 custom prompts, the questions only that client wants asked
  • Growth's full feature set: source gap, prompt value, briefs, measured experiments, and alerts that fire only on a move that clears its own band

Nothing on that list is a credit, a token or a report allowance. The one counter is that client's own custom prompts — Tallyfold is using 8 of 50 — because a custom prompt is a real engine bill every day, forever. Seats for your own people are unlimited and free, on top of the ten client seats.

Category prompts are collected once for the category, so they are not metered on any seat, on any tier.

Why the roster price is flat

The tenth client costs the same to measure as the first.

Collection added · one more client
$0.00to measure one more client in a category already running. The answers were collected before they signed.
Agency · price per month$499 flat
12345678910
Today · 3 seats$166.33per client
→
Full roster · 10 seats$49.90per client
Client 10prompts addedTallyfold, Warehouse Management Software · US.Category prompts added to serve this client: 0. Custom prompts: 8 of 50 in use.
Client 20prompts addedA second brand in the same category.Both seats read the same 1,785 answers; all 17 brands are already measured.
Client 100prompts addedAny category already live.A new category is collected once, then serves every future subscriber in it.

So the measurement cost of a roster tracks categories, not clients — and the price tracks neither.

Agency is $499 a month at three seats and $499 at ten. Meridian Growth is paying $166.33 per active client today. At a full roster that is $49.90. The tenth client is the cheapest one they will ever add.

A tool that runs a collection per account has to bill per account, because for them the tenth client really does cost ten times the first. That is not a pricing policy they could change; it is what their architecture charges them.

Prices as published on Pricing: $499/month, or $4,491 annual. Figures illustrative, brands fictional.

What the client logs into

A portal scoped to one project.

The client logs into their own view, not into yours. A portal is scoped to a single project — one brand, its competitive set, its GA4 property — and there is no route from inside it to another client's data, to your roster, or to the fact that a roster exists.

White-label means your mark, your colours, your domain. The one convention that survives your branding is the typographic one: engine text stays in the serif, everything Citino generated stays in the sans. The first time a client asks whether ChatGPT said that or the tool said it, the answer needs to be visible on the page.

What they do not see is your other clients, your seat count, your margin, or which categories you have been quietly reading.

One project per portal. The scope is a boundary, not a saved filter.

Meridian Growth · for TallyfoldIllustrative
Warehouse Management Software · US10.3% ±3.0+1.7 · within noise · n = 1,785
“…and Tallyfold has a lighter setup that smaller 3PLs tend to prefer.”
Meridian Growth · for Harbourline
Field Service Management · US—±3.0 · n = —
Meridian Growth · for Frostline
buildingopens in 5–10 dayscollection starting
What arrives on the client's date

The monthly report is
generated, not assembled.

The report is the product rendered on a schedule. The league table with its bands and sample sizes. The month's movement, with within-noise moves printed grey and labelled rather than rounded into a story. The source gap. Prompt value in sessions and conversions. Any experiment that read out, with its net effect and its verdict. It arrives on your client's date, in your branding, as a PDF and as a live link.

A monthly client report built by hand is an afternoon of screenshots, pasting and re-typed numbers, and most of that afternoon is transcription — which is also where the errors live. Here the figures in the report are the same objects the dashboard is reading, so there is no version of the meeting where the deck says one thing and the product says another.

Meridian Growth · Tallyfold · SeptemberIllustrative
Crateline31.4% ±2.8▲ +2.9
Northport WMS24.8% ±3.1+1.1
Palletworks18.2% ±2.6▼ −3.4
Shipmatic14.0% ±2.9−0.8
Tallyfold10.3% ±3.0+1.7
Bindley Ops8.1% ±2.2+0.3
Experiment read-out+7.8 pts net21.1% → 29.5% · category +0.5 · measuring · next read 5 Oct
Prompt valuebest warehouse management software for a 3PL · 186 sessions · 9 demo requests
Source gapsoftwareadvice.com · absent · 41 retrievalsreddit.com/r/3PL · unknown · 28 retrievals
Every panel in the report inherits the app's discipline — bands printed, sample sizes printed, within-noise moves grey and unarrowed.
The harder conversation

Some months the report will say nothing moved.

Every rule that makes this product honest is a rule you will eventually have to explain to a client. A gain of 1.7 points against a band of ±3.0 prints grey and labelled within noise, in your colours, on your report, in a month when a green arrow would have been convenient. An experiment can close inconclusive. Neither is editable by you or by us.

That is one harder conversation and a considerably easier relationship afterwards. A client who has watched the tool refuse to flatter them has an actual reason to believe the next number it prints.

One operational limit: a client in a category nobody has claimed yet waits while the collection starts. Frostline, Meridian Growth's fourth client, is building now and opens in 5 to 10 business days. Categories already running open immediately with their full history behind them.

→ Read the methodology

Monthly report · Meridian Growth for Tallyfold · SepIllustrative
Movement this monthNo moves outside noise.
+1.7pts · visibilityWithin noise
−3.00+3.0
Band ±3.0 · n = 1,785 · the move sits inside it, so it prints grey.
Experiment · closed this monthInconclusive
Editable by the agency or by CitinoNo

Put your roster on one index.

Ten client seats, 50 custom prompts on each, category prompts unmetered, white-label reports and client portals. Self-serve from the pricing page — no call, no gate. Claim your first client's category and check whether it is already collecting. Most are.

$499/ month · ten client seatsRead the methodology →
Pricing →