Improvement programs
A number nobody acts on is just a receipt for a problem. We take the priority list your audit produced and ship it — your pages and content, the facts engines repeat about you, and the search, social and paid channels that feed them. Then we read the same panel again, the same way, and show you what moved.
One Panorama asks 50 questions of 21 engines, 7 times each — 7,350 cells in all, read across the 20 places an answer can show your brand.
What stopping at the number costs. If nothing changes on the pages the answers cite, next quarter's reading repeats this one and the measurement budget is lost along with it. The CiteAngle team that read the grid does the work, then re-reads it under sealed, identical conditions, which is why a change here can be shown on the same panel that found it.
There is a second cost, quieter and larger. Without execution attached, the findings age — the questions move, the pages that won the answers get updated by their owners, and six months on you are paying to re-measure a report you never acted on. Work that starts the same month it is found is the only version of this that compounds.
Measurement has a built-in dead end, and most of this market stops right at it. A report tells you which questions you lose and which pages the winning answers cited. Then it goes to a team that already has a quarter of work booked, and the findings sit there. Nothing in the answers changes, so next quarter's reading looks like last quarter's, and the conclusion drawn in the room is that the measurement was not worth buying. The reading was fine. Nobody moved.
We do. A sprint turns your audit’s priority list into pages we patch and proof we publish. Each item carries a work order and a receipt when it lands. The five places we read are the five we work on: search, AI answers, content, social and paid. A program keeps that loop running month after month, with the reading and the moves on one calendar.
The check is in the contract. On every program where we do the work, the before-and-after read is a line item you are owed.
A hard push on the base layer: your pages, the technical gates, and the proof an answer engine needs before it can use you as a source at all. What follows is a list of what lands, not a list of what we work on.
Week 1
Priority set and work orders What the audit found becomes a ranked move list. Each move is scoped to one asset, with a target, the change we will make, and the test it has to pass.Week 2
Technical gates and brand facts Crawler access, page rendering and markup patches. Plus a fact sheet that states your names, aliases, domains and products the same way in every place.Week 3
Core pages and proof assets Priority pages built or patched, and new proof published — the material an answer engine can quote and credit back to you.Week 4
Receipt ledger and matched second read The change ledger closes with before-and-after captures, links and time stamps. The panel runs again on the sealed rules, and the memo on what moved goes to the decision table.We connect Search Console, Bing and GA4 during the sprint. That way the closing read sits beside numbers you can add up yourself, in your own accounts.
Two waves of work instead of one. And a scoring plan that we file before any of the data comes in.
Three reading points
A read in the middle, between the two waves. The second wave is then ordered by what the first wave actually moved, rather than by the plan we drew at the start.
Filed in advance
Part of the panel is held back from the work, and part is read on a delay. Both are filed before the data comes in, so a claim that things got better is one we can check, not just say.
Sources that carry weight
A target source list drawn from the citation data we measured, with material ready for sign-off. Then a written memo each week on what to scale, hold or stop.
The Category Sprint can start with no paid audit behind it — its own baseline read opens the work. Starting from a sealed baseline you already have just makes the first week shorter.
Where the reads sit in each sprint
Three gates, in order. Each one is here because saying the work paid off is cheap and proving it is not — so the proof is built into what we hand over.
Before any work starts, your panel’s baseline is sealed into a ledger. Questions, places, repeat count and judging rules are fixed and receipted, so later movement reads against a record no one can quietly edit.
Work orders, before-and-after captures, links and time stamps tie each change to the proof behind it. The change ledger ships with the work and stays yours.
The closing read runs your panel again under matched rules — rivals included, as they were at baseline — and publishes what moved, with the range around each figure. The statements behind our public copy sit on the public claims registry, so you can check the verdict without taking our word for it.
A sealed t0 is the starting point, locked at the first read. It covers your rivals as well as you, on the same panel and the same runs, so the closing read shows your movement next to theirs. That is what turns the ending into a number rather than an opinion. A paid audit on the same domain in the previous thirty days supplies that starting point, and its full list price is credited toward your first program payment.
The credit rules, the programs it covers and a calculator for your own mix are on the pricing page.
We build on assets you own or can give us access to: pages, profiles, markup, crawler and rendering setup, and the proof material we publish. Media buying, ad platform spend and third-party posting fees are contracted on their own.
In-app search, voice and personalized screens ride in the work layer, and the coverage map names them there.
The month-by-month scope, in full.
All three are contracted engagements. A person works on your account every month, so each one starts from a written agreement — get in touch and we scope it with you first. A twelve-month term is priced at ten months.
Every engagement that changes anything closes the loop with matched remeasurement — a claimed improvement is checked against the record, and the protocol holds the verdict.
Verification doesn’t end with our report — it ends with the results set beside the numbers in your own Search Console.
What that comparison settles is the direction of business results — how visits from search moved, and how often people looked your brand up by name. Whether we read the citations correctly is settled somewhere else. The sealed ledger, a rerun under the same conditions and the published question list carry that part.
The full monthly measurement, four moves and an authority campaign every month — ending in a board-ready deck.
Eligible for the audit credit.
Every month: a wide-angle Panorama, a written call on what to do, and two moves executed with receipts.
Eligible for the audit credit.
A fresh competitive read every month, on the full Compass grid.
Observation-led — your team executes.
Turn the audit’s priorities into executed work — a one-time sprint for a concentrated push, or a subscription that keeps measurement, decisions and execution in one monthly loop.
Annual measurement, governance, approval workflow and executive reporting scaled to brand, market, region and language count. Scoped and contracted by quote, with term, governance and invoicing fixed in writing before it starts.
Within thirty days of payment, the full list price of your most recent paid audit — Radar, Compass or Panorama — comes off your first program payment — automatically, in full. Either sprint, Growth Operations, Category Leadership: all four carry it, and your audit spend becomes execution budget.
Week-by-week deliverables and the three verification gates behind them: how the execution work runs. What stays fixed so two runs stay comparable: the monitoring contract.
Programs are measured in the US market · 21 engines · 20 places an answer can name you.
Observation-led · never miss a competitor move
Flagship program · measurement and execution in one loop
Four moves a month + authority operations · built for the board
Enterprise engagements are measured in the US market · 21 engines · 20 places an answer can name you.
What it includes
What lands each month
What lands each month
What lands each month
Ready to turn US visibility evidence into a market plan? Tell us the US buyer, target state and decision date.
Request a written proposal