What needs to be finished first
The decision is clearly defined, with alternatives and stakeholders, and data access is available.
What needs to be ready before start
DIBB template with four sections (Data, Insight, Belief, Bet); source directory for data; stakeholder review plan; versioning system (Git or Notion); success measurement setup.
One owner (strategy lead, PM, or departmental head) as author; one to three reviewers from neighboring areas; sponsor with decision mandate; data owner for fact check.
Decision question in one sentence; available data and sources; known assumptions from previous decisions; Bet time horizon; success metric for the Bet (what shows success at 6, 12, or 24 months).
1-2 hours for first draft, plus reviews
Create the DIBB template. Prepare source section for data. Schedule review slot in calendar. Activate versioning tool. Definition of done: sponsor signs Bet.
The one question this method answers
On what basis, and with which Beliefs and Insights from which data, does the team take which Bet (decision)?
Marker: Sektion
| Step | Duration | Action | Hint |
|---|---|---|---|
1Section 1: Data | 30-45 min | Collect hard data, facts, and observations. For each data point define source (dashboard, study, interview quote, market report). Add collection date and context. No interpretation in this section. | Interpreting in Data mixes layers. Data should be numbers, quotes, facts. Interpretation belongs in Insight. If merged, it is unclear later what is evidence and what is interpretation. |
2Section 2: Insight | 30-45 min | Derive key findings from data. Insight links multiple data points into one picture. Keep to 3-5 insights; more dilutes the result. Reference underlying data for each insight. | An insight tied to only one data point is often just a restatement. Real insight combines at least two data points into a statement. |
3Section 3: Belief | 30-45 min | Belief is a derived conviction from Insights. A Belief is a position: "We believe that ...", often with implicit assumptions. Make assumptions explicit. Reference Insights for each Belief. | Beliefs without explicit assumptions become untestable later. If "the market keeps growing" is hidden as an assumption, that belief cannot be updated when the trend changes. |
4Section 4: Bet | 30-45 min | Define what we do, what we do not do, what we invest in, and over which time horizon. Define Bet success measurement (which metric and when shows success). Name explicit exclusion options. | A Bet without success measurement is hope. Each Bet needs at least one metric and one horizon. A Bet without non-action options is wishful thinking. |
5Section 5: Review and signature | 30-60 min | Run reviewer round with written comments. Owner revises where needed. Sponsor signs Bet in writing. Record date, version, and review date (for example 6 months). | Reviews often reveal where data is weak or Beliefs are vague. Substantive critique triggers revisions, not polish. Sponsor signature creates commitment, not ceremony. |
6Section 6: Learning and updates | Quarterly, 1 h | Quarterly review: which Beliefs are challenged by new data? For each Belief set status: confirmed, weak, disproven. If disproven, adapt or reconfirm Bet with new rationale. | Teams that never review Beliefs keep Bets alive artificially. Visible update of Beliefs protects against sunk-cost bias. |
What comes out at the end
DIBB document with four sections, data source directory, assumptions list per Belief, Bet with success metrics and horizon, reviewer comments, sponsor signature, and versioning. Quarterly update log.
Use SemVer per DIBB document. Major version for Bet changes, minor for Belief updates, patch for data refresh. Archive prior versions with diff view. New Bet reversal creates a new DIBB with a reference.
Compact working template for DIBB with context, input, output artifacts, and next step.
Spotify framework: Data, Insight, Belief, Bet, to structure decisions in a traceable way.
When and for what do we use this method?
Which data, observations, decisions, or materials are available?
Short notes along the runsheet.
Owner, date, and success signal.
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
We invest 180k EUR over 6 months for Austrian market entry: local UI adaptation, FinanceOnline integration, performance marketing in Austria.
Belief status Q3 (future): Test B1 against customer acquisition rate and B2 against CAC comparison.
Recognize symptoms and steer against them
Data section contains interpretations, Insight section adds new data points.
Keep sections strictly separate. Data is only facts and sources. Insight combines multiple data points. Reviewers validate separation.
Belief reads as a closed statement, assumptions are not explicit and are later disconfirmed.
List assumptions explicitly under each Belief. Review quarterly. A belief without assumptions is not review-ready.
Everything is done, no trade-offs are visible, resource clarity is missing.
Name explicitly what is not done for each Bet (market, feature, initiative). Trade-offs force clarity.
Bet is clear but no one can state what success is measured against.
Metric and horizon are mandatory. If no metric can be found, the Bet is not thought through. Use leading indicators if necessary.
DIBB document is created once and never reviewed; Beliefs stay unchanged for six quarters.
Make quarterly review mandatory in calendar. Set explicit Belief status (confirmed/weak/disproven). Update Bet when a belief is disproved.
Reviewer comments are only cosmetically applied, and substantive criticism is not addressed.
Add a written response for each reviewer comment: accepted, rejected with reason, or parked. Smoothing is method misuse.
Done signals checkable in under a minute
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.