Foundation · AI Value Associate companion

Value Orchestration Workbook.

A working document for the person who has to turn AI demand into value that can be seen. One partner area, one quarter, three value conversations. It saves as you type, on this device only, and prints to a clean PDF for the review.

For: Business Relationship Managers, value and benefits owners, demand and portfolio leads
Takes: About ninety minutes the first time; twenty minutes a month after that
Use with: Toolkit canvases 06, 09, 18 and 19, and the AI Value Associate micro-course
Saves: In your browser as you type; export a copy before you clear it

Before you startHow to use this workbook

Pick one partner area and one quarter. The workbook is built for a quarter. A year is too long to correct and a week is too short to measure.
Work the pages in order the first time. After that, go straight to whichever value conversation is next.
Write numbers as ranges with conditions, in the partner's unit. Hours a week, cases a month, days of cycle time, or money. A number with no condition attached is a promise you did not agree to keep.
Name a person in every owner box. A team name is not an owner. Nobody in a team answers for a number.
Print it for the review, and export a copy before you clear it. Everything you type stays in this browser.

Partner area
Quarter
Completed by

Page oneValue map: one line from ambition to value

List the AI demand in this partner area, live or proposed, up to six items. For each, name the organisational ambition it serves, write the item as a need sentence with no tool in it, say which gate it is at, and record the committed value, its owner and the state of the evidence.

If a row has a tool in the demand column and nothing in the ambition column, it has not been shaped yet. Send it back to the front door before it goes any further.

	Ambition it serves	Demand item (need, no tool)	Gate	Committed value	Business owner	Evidence
01
02
03
04
05
06

Page twoThe three value conversations

Value is not agreed once. It is agreed before commit, measured after handover and revisited at the review, and each of those is a different conversation with a different question at its centre. Plan all three for this quarter: who you will have them with, when, the evidence you will bring and the decision each one has to produce.

Value planning

Before commit

"What is the lowest figure you would still act on, and what has to be true for the upper half to arrive?"

With whom When Evidence I will bring The decision it produces

Value realisation

After handover

"What did we measure, against what we committed, and who is answering for the gap?"

With whom When Evidence I will bring The decision it produces

Value optimisation

At the review

"Which of these do we scale, fix or stop, and what goes back to the front door as a lesson?"

With whom When Evidence I will bring The decision it produces

Page threeValue leak map

Value goes missing at predictable places, one at each gate. For each leak, write the symptom you actually see in this partner area, the item where it is happening now, the fix, and who owns the fix by when. Leave a row blank if you genuinely cannot find the leak; do not invent one to fill the page.

A fix that only works while you personally chase it is not a fix. It is a leak with a person standing in it.

Leak	Symptom you see	Where it is happening now	The fix	Owner	By when
Built to the wrong needDiscover: the request was taken as the need
Committed without evidenceAssess: readiness scored on hope, not on what was seen
Ranked by volume or by voicePrioritise: not by value against the whole portfolio
No owner and no number at commitDesign: funded without anyone to answer for it
Deployment mistaken for useAdopt: log-ins counted, behaviour unchanged
Month nine never happensRealise: the review is never booked, so the number is never tested

Page fourAttention allocation

Relationship attention is the scarcest thing you have, and it concentrates where the noise is rather than where the value is. Place this quarter's items in the four boxes, then write the routing rules that decide, in advance, what a person handles and what is routed elsewhere.

High value, high attention

Protect

High value, low attention

Move attention here

Low value, high attention

Route or stop

Low value, low attention

Leave alone, review quarterly

A person handles it when
It goes to self-service or automation when
It is declined, with reasons, when

Page fiveValue evidence ledger

One row per committed benefit. Write the unit and the baseline before anything else, then the committed range, then what has been measured so far. Say plainly whether a figure is measured or estimated, and write avoided cost in the same row as the spend it is set against, never on its own.

An estimate that has sat in the ledger for two reviews without becoming a measurement is telling you something about ownership.

	Benefit	Unit	Baseline	Committed range	Measured so far	Basis	Measurement owner	Next review
01
02
03
04
05
06

Avoided cost, beside spend

Page sixThirty-day value plan

Three moves, no more. Each belongs to one of the value conversations and fixes one gate. Write the first action small enough to do this week, and the evidence that would show it worked.

	The move	Conversation	Gate it fixes	First action	Evidence of progress	Date
01
02
03

Page sevenMonthly value review

Run the review with the ledger open. Tick each question once it has been asked out loud and answered, then record what was decided.

Which committed numbers gained a measurement this month, and which are still estimates?
Which item moved through a gate, and did the reasons hold up afterwards?
Where did attention go this month that value did not justify?
Which leak showed up again, and is the fix systemic or personal?
What did the most recent month-nine review send back to the front door?
Which alternative to a build was taken seriously enough to have an owner and a cost?
What is the one value decision for the next thirty days?

	Observation or decision	Evidence	Owner and next action
01
02
03

Value orchestration is not the job of filling in boxes. It is the habit of asking, at every gate, what this is for, what it is worth in the partner's unit, and who will answer for it after launch. The workbook exists to make that habit visible to the people who fund the work.

The GOVBRM Value Orchestration Workbook is part of the AI Demand Toolkit and is free to use, print, adapt and share for your own or your organisation's internal use. Grounded in the BRM Body of Knowledge (BRM Institute). A generic model: adapt the gates and scales to your own governance.