DIBB Decision Framework

The action every executive and management consultant must take is direct: replace opinion-first decision-making with data-first decision sequences. Define your current organizational bets explicitly. Trace each bet back through the DIBB chain to its originating data. Where that trace breaks down — where a bet exists without a belief, or a belief exists without an insight, or an insight exists without data — you have identified a structural gap in your decision-making architecture. Close those gaps before committing resources. Decisions without a complete DIBB chain are not strategy; they are speculation.

Origins in the Spotify Rhythm

Spotify developed the DIBB framework as a component of the Spotify Rhythm — its organization-wide system for establishing strategic alignment across a rapidly scaling, highly autonomous structure. Henrik Kniberg, Spotify’s agile coach and one of the primary architects of the Rhythm, described the challenge Spotify faced as it grew: how do you maintain coherent strategic direction across dozens of semi-autonomous squads without recreating the hierarchy and bureaucracy that autonomy was designed to escape?

The Spotify Rhythm answered that question with a cascading alignment architecture. Company-level beliefs defined the strategic context for the next three to five years. North Star metrics and two-year goals provided measurable direction aligned to those beliefs. Bets — at the company, functional and team levels — translated belief and direction into funded, time-bound initiatives. The DIBB framework was the reasoning chain that justified each bet and connected it to the data and insights that made it credible. It served simultaneously as a decision tool, a communication protocol and a governance standard.

The DIBB framework’s durability comes from its discipline. It imposes a reasoning sequence that organizations cannot shortcut without exposing the gap. When a team is required to state the data underlying their insight, the insight underlying their belief and the belief underlying their bet, the quality of the reasoning becomes visible — to the team itself, to its stakeholders and to the leadership allocating resources across competing priorities. That visibility is precisely what organizations need when they must allocate finite attention and capital across a large portfolio of potential initiatives.

Data

Data in the DIBB framework means observable, verifiable facts about the world — market conditions, customer behavior, product performance metrics, competitive signals, operational statistics, financial trends. Data is not interpretation. It is the raw evidentiary material from which reasoning begins. Spotify’s application of the DIBB framework emphasized a specific discipline at this stage: the data must be objective and current, not selectively curated to support a pre-formed conclusion.

In practice, this discipline is harder to maintain than it appears. Organizations that operate with strong functional cultures frequently gather data to confirm existing positions rather than to challenge them. A product team that has already decided to build a feature will find user data that supports it. A business unit leader who believes a market is attractive will weight the evidence that confirms that belief. The DIBB framework’s data stage requires the reverse: gather the full picture first, then allow the data to generate insights, rather than generating insights first and marshaling data behind them.

The practical scope of data at the DIBB stage is deliberately broad. Spotify teams drew on market trend data, user engagement metrics, streaming behavior statistics, competitive positioning data and customer research. For an executive team applying DIBB to a strategic allocation decision, the data stage might encompass total addressable market (TAM) sizing, win-rate analysis by segment, net revenue retention by customer cohort, technology cost trajectories and macroeconomic indicators relevant to the decision domain. The criterion is relevance and objectivity, not volume.

Insight

Insight is the interpretive step that converts raw data into meaning. An insight is not a data point — it is a pattern, a relationship, an anomaly or a trend that the data reveals when examined with analytical discipline. Spotify’s DIBB framework defined insight as an actionable observation derived from the something specific enough to generate a hypothesis about what it implies for the organization’s decisions.

The distinction between data and insight is one of the most frequently blurred boundaries in organizational decision-making. Leaders who present data as if it were insight, or who treat data summaries as analysis, consistently under-exploit the information they hold. A retention rate of 73% is data. The observation that retention rates among users who engage with a specific product feature within the first 30 days are 40 percentage points higher than among those who do not is an insight — because it implies something actionable about where onboarding investment should go.

Generating genuine insight requires analytical capability and a willingness to surface findings that challenge assumptions. In Kniberg’s description of the Spotify Rhythm, insights were expected to do exactly that: they were meant to challenge the team’s current beliefs, not simply confirm them. Organizations that cultivate this norm — rewarding analysts and team members who surface inconvenient patterns — generate higher-quality insights than those where delivering positive news is the implicit social expectation. The DIBB framework makes the insight stage explicit precisely so that this norm can be applied and evaluated.

Belief

A belief in the DIBB framework is a hypothesis — a statement about what the organization thinks is true, based on the insights it has developed. Kniberg was clear about the epistemic status of a belief: This is not a fact. It’s a belief. The distinction matters. A fact requires no further testing — it is already confirmed by evidence. A belief is a reasoned position that carries conviction based on available evidence but remains open to revision as new evidence arrives.

The belief stage serves a specific function in the DIBB chain: it makes the organization’s assumptions explicit and testable. When a leadership team states a belief — We believe that expanding our content library in the audiobooks category will attract a segment of users we currently underserve — they create a falsifiable proposition. That proposition can be tested by a bet, evaluated against defined success metrics and revised based on what the test reveals. Without the belief stage, the bet that follows has no stated assumption to test and the organization cannot learn from its outcome because it never stated what it expected to discover.

Beliefs in the Spotify Rhythm were hierarchical. Company-level beliefs described the organization’s strategic understanding of the world over three to five years — the broad market, competitive and behavioral dynamics that should guide all resource allocation. Team-level beliefs were narrower, more near-term and more specific to the domain in which the team operated. Both levels shared the same logical structure: they were reasoned positions derived from insight, stated explicitly and treated as working hypotheses rather than fixed truths.

The belief stage also performs an alignment function that the data and insight stages alone cannot achieve. Two teams can look at identical data and derive identical insights but hold different beliefs about what those insights imply. One team believes the insight implies accelerating investment in an existing market. Another believes it implies entering a new market before a competitor does. Both beliefs are defensible from the same data and insight. The DIBB framework surfaces that divergence explicitly — at the belief stage, not after resources have already been committed to conflicting bets.

That surface-level alignment function has direct organizational governance value. Executives who require teams and business units to state their beliefs explicitly before presenting resource requests create an environment in which strategic disagreements are diagnosed at the right level — at the level of competing hypotheses about the world — rather than at the level of competing project proposals. The former is a strategic conversation. The latter is a resource negotiation. The DIBB framework converts the latter into the former.

Bet

A bet in the DIBB framework is a funded, time-bound commitment to act on a belief. Kniberg described a bet as an action to either test or capitalize on a strong belief. The language of betting — rather than planning, deciding or investing — is deliberate. It acknowledges uncertainty without allowing uncertainty to become a reason for inaction. An organization that requires certainty before committing resources will commit resources very rarely, because strategic decisions almost never admit certainty. An organization that frames commitments as bets acknowledges the uncertainty, defines the expected outcome and proceeds with accountability.

Spotify structured its bets into three tiers. Company bets were large cross-organizational initiatives typically running six to 12 months, requiring coordinated effort across multiple functions and aligned to company-level beliefs. Functional bets were function-level initiatives aligned to company bets, typically scoped within a single team or domain. Team bets were smaller, faster initiatives that teams could execute and evaluate within shorter cycles. This tiering ensured that each level of the organization was making bets that were coherent with — and traceable to — the bets above them in the hierarchy.

The bet stage requires three elements to be valid within the DIBB framework. First, the bet must state its expected outcome — what the team believes will happen if the bet succeeds. Second, the bet must define measurable success metrics — the data points that will tell the organization whether the outcome has been achieved. Third, the bet must acknowledge the belief it is testing — so that the organization knows what assumption to update if the outcome does not materialize. Without all three elements, the bet generates activity but not learning.

Spotify operationalized its bets through a Bet Board — a shared, visible display of active bets organized in a format similar to a portfolio kanban. The board showed company-level and functional-level bets in a Now, Next and Later structure, indicating their current status, their relationship to higher-level bets and their defined success metrics. The board was not a planning artifact maintained by a central function; it was a live governance tool used by teams and leadership alike to maintain visibility into the full portfolio of organizational commitments.

The Bet Board creates an organizational condition that most planning processes fail to establish: genuine transparency about what the organization has bet on and why. When every bet is visible, traceable to a belief and measurable against defined outcomes, the leadership team can evaluate the portfolio’s coherence, identify conflicting bets, surface resource contention and make reallocation decisions based on evidence rather than advocacy. The Bet Board converts strategic intent into an auditable, shared reality.

DIBB Across the Organization

The DIBB framework scales vertically across organizational levels and horizontally across functions. Vertically, each organizational tier — executive leadership, functional leadership, product and operational teams — applies the full DIBB chain within its decision domain and aligns its beliefs and bets to those operating at the level above. A product team’s bet should be traceable to a functional belief, which should be traceable to a company-level belief grounded in market and competitive data.

Horizontally, the DIBB framework creates a shared decision language that reduces the translation friction between functions. A technology team and a commercial team operating with the same DIBB chain can disagree productively — at the level of competing beliefs about what the data implies — rather than talking past each other with incompatible frames of reference. That shared language is one of the framework’s most practically valuable organizational contributions.