Visioning
Visioning is the facilitated practice that states, in one short sentence, the outcome expected from an initiative and the audience for whom that outcome matters. The Agile Extension to the BABOK Guide makes it technique §7.24 and places it at the Strategy horizon. It holds three pieces: the vision statement, the workshop that produces it with the key stakeholders and two or three impact metrics showing whether the organisation is getting closer to it. The statement is then the decision criterion: a request that does not serve that outcome is turned down by quoting the sentence.
Goal
Visioning produces one sentence, short and understandable without a glossary, saying what an initiative must have changed and for whom. The Agile Extension gives it the purpose of fixing that outcome, then of directing the work towards it.
The problem it addresses is one of trade-offs. An initiative receives requests from several departments, each defensible on its own. As long as the expected outcome is written down nowhere, the order is set by the rank of whoever asks or by the date the request arrived, and the scope grows until the budget stops it. The guide states three contributions: the statement says what falls inside the scope and what falls outside it, it turns the team and its stakeholders towards the value the organisation expects and it gives the stopping criterion, the moment the organisation judges that enough product has been delivered.
The deliverable adds to the statement the record of the workshop and the measured starting point of each impact metric. The guide places the technique at the Strategy horizon, where it serves to inform the decision options, to clarify the organisation's vision and to fix the subject of a new initiative.
Usage
When to use it
- Launching an initiative: fix the expected outcome before opening the backlog and committing the spend.
- Scope reopened at every committee meeting: the statement gives the criterion each request is compared against.
- Several teams on one initiative: a shared sentence replaces as many readings as there are teams.
- An initiative with no stopping criterion: the vision, once reached, says at what point delivery is enough.
When not to use it
- Solution already settled by the sponsor: the session ratifies the choice and manufactures the confirmation bias the guide warns about; set the boundary with scope modelling and move on to the requirements.
- Regulatory obligation with an imposed deadline: the expected outcome is written in the legal text; the business case and business rules analysis carry the framing.
Description
What the statement contains
The guide describes the vision statement as a simple sentence, carried by words or by images, that says the purpose and the scope of the initiative. The expected value is expressed in it as a set of benefits, whose detail sharpens as the initiative advances and as the organisation understands better why it needs a solution.
The guide insists: the conversation that produces the sentence counts as much as the sentence. A statement written by one person and then circulated leaves everyone to read it their own way, which is the original situation with one more document in it.
The guide stops at that description and gives no template for the statement itself. Three requirements, drawn from that use in deciding between requests, make the sentence usable. It names the beneficiary, precisely enough for someone to recognise that they are excluded. It states the outcome as that beneficiary observes it, in their words. It carries a deadline, without which nothing allows anyone to say that the vision has been reached; the most quoted vision statements carry none, and they are not used to decide between requests. Highsmith adds a test of admissibility, his elevator test: the team must be able to present the initiative to an outsider in two minutes.
The workshop that produces the statement
The guide hands the drafting to a facilitated session with the key stakeholders, run by the business analyst. It names two useful exercises: the Product Box, where participants draw the product's box and then sell it to the room, and the product differentiation statement, which the guide does not define and whose reference form is Geoffrey Moore's positioning template.
| Template field | What it fixes |
|---|---|
| For (the target segment) | Who benefits from the outcome, described narrowly enough to exclude someone. |
| who (the need or the opportunity) | What that segment is trying to obtain today and does not obtain. |
| (the product) is a (the category) | The box the reader files the offering in without explanation. |
| that (the key benefit) | The outcome the segment obtains, stated from its point of view. |
| Unlike (the existing alternative) | What the segment does today for want of anything better, including nothing at all. |
| our product (the difference) | What justifies the change of habit. |
The guide states the limitation in so many words: the group's imagination, diversity, trust and ability to collaborate set what the session can produce, a weak group producing a statement that leads to poor decisions. The session fits in two hours with six to ten people: the sponsor, the person who will hold the budget, someone from the beneficiaries' side and the team that will deliver. One rule set before the session avoids deadlock: the sponsor settles the final wording, the room having the right to contest it before leaving. Where nobody contradicts the sponsor out loud, the sentence will come from the loudest voice: gathering positions in interviews ahead of the session means the facilitator can raise them.
The impact metrics
An impact metric is objectively measurable information showing whether the organisation is reaching its vision. That is the name the guide gives it; in the sense used by performance measurement, the object is an indicator, whose metric is the level read on a given date. The guide attaches a caveat to it that changes how it is used. An indicator moving in the right direction signals that the initiative is heading towards the vision; it does not establish that the initiative produced that movement. Attributing the movement takes another instrument, a comparison group or an A/B test where the product lends itself to one.
Two or three indicators are enough. Each carries a starting point measured and dated, a target with a deadline, the source that produces it and the name of the person who takes the reading. An indicator that no system produces today leaves two ways out: replacing it with one already measured or bringing its instrumentation inside the scope of the initiative.
What makes the technique fail
The tick-box exercise
The guide puts at the head of its limitations the session held once, whose result is never read back nor revised in the light of feedback. It adds the twin limitation: the effort returns nothing as long as the team does not use the statement to decide and to prioritise. The corrective is a meeting that already exists. The quarterly revision of the product roadmap reads the statement back before placing the themes again, and the minutes record what was turned down by quoting it.
The statement that names the solution
The guide flags two neighbouring limitations. The technique can fix the group on a single solution and stop it learning from what it observes; it can also narrow the field to one option when several were still open. Both are triggered at drafting time. "A self-service claims portal" names the means and closes the discussion. "A policyholder files and tracks a claim in under three minutes" names the outcome and leaves several ways of getting there open, some of which cost a fraction of the portal. The check fits in one question asked in the session: what in this sentence describes what the beneficiary gets?
The sentence nothing can contradict
"Improving the experience of our policyholders" passes every review and turns down no request. The check is again made in the session: name a plausible request that the sentence excludes. If nobody finds one, the sentence is a slogan and the work of deciding between requests is untouched.
AI considerations
A language model serves the preparation and the review. On the raw notes of a session, it produces five or six candidate wordings for the room to compare, which moves the discussion from a blank page to a choice. It spots the statements that name a means instead of an outcome, a question of form on which it rarely gets things wrong. It proposes candidate indicators from the stated outcome, leaving the team to say which of them an existing system produces. It sets the statement against the backlog and lists the items serving no part of the outcome, a sorting job nobody runs by hand across eight hundred tickets and one that prepares the agenda of the revision.
What the model does not replace is the conversation. A consensus sentence that no participant argued for leaves the disagreements intact under an acceptable wording. A target proposed by a model commits nobody; the one that counts is negotiated with those who will have to hold it. Telling correlation from causation calls for a judgement on what was already moving in the market, information absent from the text submitted. A vision statement also carries unannounced market entries, volumes and prices: these are removed before anything goes to a public model.
Examples
Valmonta Assurance-maladie SA opens an initiative on the reimbursement of its policyholders' claims. Twelve per cent of claims arrive through the mobile application already in use, which the planned portal is to replace. The visioning session brings eight people together for two hours and comes out with two texts: the positioning statement, then the vision sentence drawn from it.
Positioning statement, produced by the exercise
For policyholders who already use the mobile application, who want to file and track a claim without sending receipts through the post, MonAssurance is a self-service reimbursement space that acknowledges receipt in under a minute and settles the majority of claims with no attachment. Unlike sending them by post, our service asks for neither paper nor a phone call.
Vision statement, adopted by the session
By the end of the year, a policyholder files and tracks a claim in under three minutes, with no paper.
The beneficiary: the policyholder, who does or does not recognise themselves in the sentence.
The observable outcome: filing and tracking, in under three minutes, with no paper.
The deadline: the end of the year.
Session of 4 August 2026, eight participants. Sponsor: the claims division.
Two impact metrics go with the statement.
| Indicator | Starting point measured | Target at 12 months | Source of the measure | Who takes the reading |
|---|---|---|---|---|
| Share of claims filed digitally | 12% (4'800 out of 40'000 per month) | 65% (26'000 out of 40'000 per month) | Filing logs of the application and of the portal | Head of the digital channel |
| Average processing time for a claim | 9 days | 3 days at most | Claims management system | Head of claims |
Processing a paper claim costs CHF 14.50, scanning, data entry and the return letter included; filing it digitally costs CHF 3.20, a difference of CHF 11.30. Once the target is reached, the 21'200 claims moved each month, the difference between 26'000 and 4'800, are worth CHF 239'560, an annual run rate of CHF 2'874'720. The first year takes only a fraction of that, the digital share climbing gradually towards 65%. The committee therefore sets two numbers against the CHF 850'000 asked for building the portal: the gain of the first year and the run rate reached afterwards.
The guide's caveat bears on how this table is read a year from now. A digital share risen to 65% indicates that the organisation is heading towards its vision; it does not establish that the portal produced the shift, since a rise in postal charges or a competitor's application pushes policyholders the same way.
Visualisations
The statement is shown on its own, in large type, at the head of the initiative's space, with its date and the name of the group that wrote it. Without those two mentions, a reader takes the sentence for a slogan from management. The guide allows images as much as words, and the drawing that comes out of a Product Box holds that role for an audience that will not read a document.
The positioning template is shown filled in, its six fields marked, which lets each participant contest the field that concerns them. The impact metrics are read as a table, one row per indicator. The two columns most often missing are the source and the person who takes the reading; without them the next reading does not happen and the table becomes a dated statement of intent.
Cost
| Phase | Level | Justification |
|---|---|---|
| Preparation | Medium | Bringing together the people who decide, taking the starting figures and preparing the template take a few days spread over two weeks. |
| Execution | Low | A two-hour session, then one review per quarter grafted onto a committee that meets in any case. |
| Documentation | Low | One sentence, a filled-in template and a table of two or three indicators. The cost sits in taking the readings period after period. |
Tooling
The wall remains the tool of the session: cards, markers and the box of the Product Box when that exercise is the way in. It assumes a room and participants gathered in one place, and it produces nothing that circulates without a photograph and a transcript.
The collaborative whiteboards (Miro, Mural and their equivalents) hold the session for a group spread across several sites. They keep the record of the wordings set aside, material the next revision reads back when the question returns.
The documentation space (Confluence, SharePoint and their equivalents) carries the statement, dated and signed, at the head of the initiative's space. A sentence buried on page eleven of a framing document settles no trade-off.
The backlog tool (Jira, Azure DevOps and their equivalents) displays the statement at the head of the board and makes it quotable at the moment a request arrives. It is the one place where the vision meets the daily decision to take an item or to turn it down.
That leaves the system that produces the measures: product usage measures, a business management system or a decision-support dashboard. Without it, the impact metrics stay declarative and the vision is never verified.
Sources
- IIBA, Agile Extension to the BABOK Guide, §7.24 Visioning: the purpose of the technique, its three elements, being the vision statement, the visioning session and the impact metrics with the caveat on correlation, together with the three strengths and the five limitations it states. §4.7.1 places the technique at the Strategy horizon and table 7.0.1 files it among the communication techniques, with the team and with the stakeholders outside it.
- Geoffrey A. Moore, Crossing the Chasm: Marketing and Selling Disruptive Products to Mainstream Customers, 3rd edition, Harper Business, 2014: the six-field positioning template, the reference form of the product differentiation statement.
- Jim Highsmith, Agile Project Management: Creating Innovative Products, 2nd edition, Addison-Wesley Professional, 2009: the product vision built in session around the product box and the elevator test, two minutes to present the initiative, as the criterion of admissibility for the statement.