Welcome to Agentic Benchmarking thatkeeps one governed peer file that every deck cites.
The same institution sitting above and below peer median in one board book costs IR its credibility either way. Risely's agents hold one governed peer file, refresh it on federal release, and feed every comparison from it. An IR analyst signs each refresh.
Peer file · governed
watching federal releases
IPEDS release lands
pulled the week it publishes and never re-keyed
Retention against the peers
84.2% · 3rd of 15
vintage stamped · aspirational set excluded and labeled
Provost's deck
…
CFO's deck
…
Board book
…
The peer set is chosen once, on purpose.
The file holds fourteen peers and six aspirational institutions, chosen deliberately and governed from then on. The aspirational set stays labeled in every chart, and the era of each deck assembling its own comparison group ends.
The set is governed
assembling
14 peers · chosen and revisited deliberately
6 aspirational · marked as aspirational in every chart
A deck's own comparison group · declined, with the governed file offered instead
The set is a decision the institution made once, and every chart inherits it.
Federal data lands the week it releases.
IPEDS and Common Data Set numbers reach the peer file the week they publish, each with its vintage tracked. The office stops re-keying spreadsheets, and a chart can always say which year of which release it stands on.
Federal releases land
0/3
Nobody re-keys a spreadsheet, and nobody quietly mixes two vintages in one chart.
Every deck reads the same rank.
The provost's deck and the CFO's draw the same ranks from the same file, so the institution stops sitting above and below peer median in the same board book. The credibility cost of that split always landed on IR.
Retention vs peers · 84.2% · 3rd of 15
the governed line at one vintage
Provost's deck
…
CFO's deck
…
The institution stops sitting above and below peer median in the same board book.
A restatement flags everything it touched.
When a peer restates, every chart built on the old number flags itself before a boardroom sees it. Risely's agents draft the refresh with the correction noted, and an IR analyst signs it into the file.
A peer restates retention
the old number is now wrong in three charts
The stale comparison dies before it reaches a boardroom, and an analyst signs the refresh.
One thing moves. The file re-runs.
A peer restates a figure, a new IPEDS year lands, the provost adds an aspirational school. Any of them changes a comparison, and every one runs the same way.
A peer restates retention
the correction lands in the federal file
The peer file updates
with the vintage stamped
The rank recomputes
still 3rd of 15
Stale charts flag themselves
before a boardroom sees one
The analyst signs the refresh
one review makes every deck current
What changes, and how it runs.
One peer file behind every deck
The provost and the CFO read the same rank, because it is the same file.
Comparisons stay labeled honestly
Aspirational is marked as aspirational, and vintages are tracked.
Stale numbers die before the boardroom
A restatement flags every chart it built.
The technical read
Connects read-only to your data warehouse and the public federal releases to start
An IR analyst signs every refresh of the peer file
Every action lands on an append-only log
Runs alongside what you have. Nothing rips out.