Skip to main content
The Multi-Site Dashboard appears in the sidebar once your organisation has more than one site. Agencies see the same page as Agency Dashboard, covering every client site they manage, their own brand included.

​
The tabs

What you see depends on who you are:
  • A multi-site organisation (one company, several sites) gets Overview and Prompt Matrix.
  • An agency gets Overview, Cohorts and Shared Taxonomy. The Shared Taxonomy tab holds four views of its own, on the row beneath it: Overview, Sources, Prompt Matrix and Attention Queue. It unlocks once a shared taxonomy is configured.

​
Overview

Choose a period from 7D to All, and narrow to some sites if you like.
  • Visibility Trend: every site as a line, coloured to match its rows in the rankings. Switch the metric. On 6M, 9M and All, once there are more than about three months of results, each point is a site’s latest result that week.
  • Client performance (agencies): every client at its latest completed run, with Prompt Visibility Rate, Share of Voice, Sentiment and the change since the previous run. A client’s change is left blank until it has a previous run to compare. A client with several sites is blended across them, weighted by how many brand-naming answers each contributed, so a flagship is not averaged flat against a small second location. Select a row to open that client’s workspace.
  • Performance Matrix: every site crossed with every platform, retinted by the metric you choose.
  • Top Rankings: your sites ranked against each other per platform.
For an agency, the Overview always covers every prompt each client tracks, so each client is measured on its own questions. To compare clients on the same questions, use a shared taxonomy and the Shared Taxonomy tab.

​
Cohorts

Agencies only. The Cohorts tab puts each client into one of four cohorts, based on its Prompt Visibility Rate and whether that is rising or falling. The cohort suggests the most useful next move for the client. It is not a grade. With a shared taxonomy, cohorts always compare every client on the same shared prompts, and the line at the top of the tab says how many. If your agency keeps more than one set, a picker at the top of the tab chooses which one, and every view follows the same choice. Without a shared taxonomy, each client is measured on its own prompts.
  • The boundary and the trend: each cohort card shows the visibility figure that divides strong from behind, and Kevin’s greeting and the note under Client Growth Matrix say which period the rise or fall is measured over.
  • Updated every day: cohorts are dynamic. Clients move between them automatically as new results come in.
  • Kevin’s Observations: the first panel on the tab. Kevin’s greeting explains how cohorts work, with today’s boundary, then he picks out what deserves attention: the Protect client closest to slipping into Recover, the Accelerate client closest to reaching Scale, or an AI platform where your Recover clients trail the rest of the portfolio. Select an observation to narrow the cohort cards, the chart and the client list to that client or cohort, and select it again to show everyone. Ask Kevin about any client in the box underneath; he answers for the shared taxonomy on screen. The panel folds into its title bar when you scroll down, and opens the way you left it next time.
  • Where every client sits: a card per cohort, listing its clients with their current Prompt Visibility Rate, highest first. Select a card to narrow the chart and the client list below to that cohort, or select a client on a card to pick out just that client. Select the card again, or Clear filter on the client list, to show everyone.
  • Client Growth Matrix: each client is a point, placed by current Prompt Visibility Rate from left to right and by its change in points from bottom to top. The dashed lines mark the boundary and no change, so each corner is a cohort: Accelerate top left, Scale top right, Recover bottom left and Protect bottom right. Your own agency’s brand is a reference only.
  • Client benchmark: every client ranked by current Prompt Visibility Rate, with its Change in points, Portfolio rank, Cohort and Data freshness. Search for a client or filter by cohort, and select a client’s name to open its workspace.
Clients outside the four groups stay in the Client benchmark, and a line under the cohort cards counts them. The cohort filter reaches each kind:
  • New: a current result with nothing yet to compare it against. Establish a baseline before judging movement. New clients are not plotted on the matrix.
  • No current data: no current result to place the client.
  • Excluded references: paused and draft clients, your own agency’s brand and clients not on the shared taxonomy. Data freshness says which, and none of them moves the boundary.

​
Prompt Matrix

  • One brand, many locations (multi-site organisations): the Prompt Matrix tab compares the same prompts across every location of one brand, so you can see which towns the brand wins and which are weak. It fills in once your organisation has two or more locations of the same brand, such as a brand set up through Multi-site onboarding.
  • Shared taxonomy: for an agency with a shared taxonomy, the Prompt Matrix view inside the Shared Taxonomy tab shows every client site on the set against every shared prompt. See Shared taxonomy.

​
Sources and Attention Queue

Both live inside the Shared Taxonomy tab for agencies. Sources shows the domains and pages AI cites across every client on the set, so you can spot placements that would lift several clients at once. Attention Queue flags the clients whose data shows something to act on, grouped by the kind of work involved, with the evidence behind each flag and a preferences dialog to choose which signals you want to see. See Shared taxonomy for the full walk-through.

​
Where it reads from

The same stored rollup that powers the Looker Studio connector, so a chart you build in Looker always matches the numbers here.