The Six Dimensions of the Velocity Model
Velocity Model v1.0 · Published September 2026 · Developed by The C2 Group
A dimension is an area of the operating system that the Velocity Model examines to determine an organization's stage. There are six, grouped into three enabler families, and each has a defined description at every stage of the curve.
The families name what kind of capability each dimension represents. The organizational dimensions describe who owns digital and who does the work. The operational dimensions describe how work and content move. The technical dimensions describe what the organization knows and what it runs on. A seventh area, Governance, Risk & AI Readiness, is deliberately not a dimension: it cuts across all six as a gate, because it measures whether additional speed is safe rather than how much speed exists. This page describes each dimension, the full stage-by-dimension matrix behind placement, and the gate in detail.
New to the model?
Start with the definitionThe stage-by-dimension matrix
The matrix is the model at full resolution: each dimension described at each of the five stages. The Steady column is the plateau signature.
The matrix is read by row and by column together. Reading a row shows how one dimension matures across the curve. Reading the Steady column shows the plateau signature: every row competent, every row static. Placement against the matrix is what the Velocity Score approximates through self-report and the Velocity Assessment establishes through evidence.
Organizational: who owns digital, and who does the work
Leadership & Digital Vision
Who owns digital, and whether the roadmap is tied to outcomes.
Why it governs velocity
Velocity is set by decision speed before it is set by build speed. Without an owner, nothing can be prioritized, so every request competes on urgency instead of value. Without a roadmap tied to outcomes, the backlog fills with releases that prove activity rather than progress. This dimension also controls whether any other dimension can improve, because changing how work moves requires someone with the authority to change it.
How it matures
This dimension pivots at Accelerating, when decision rights arrive. Everything before that is ownership without authority: first no owner at all, then a champion without a mandate, then an owner steering by releases because releases are what they can control. Everything after is authority migrating toward the team, until the team sets and revises the vision itself and leadership's remaining job is governing risk.
Common failure patterns
Ownership by committee, where every decision is a negotiation and the calendar absorbs the cost. A roadmap inherited from the last replatform and never revisited against outcomes. An owner who carries responsibility for digital without decision rights over it, which produces accountability for a system they cannot change.
Related terms: digital ownership, digital strategy, executive sponsorship, website roadmap, digital governance.
Team Capability & Capacity
Whether the team has the roles and the slack to build, sustain, and adopt.
Why it governs velocity
Capacity determines whether improvement is possible at all. A team running at full utilization on keeping the current experience alive has nothing left to change the system with, so its velocity is fixed no matter what it buys. Slack is the raw material every stage transition is built from. Roles matter for a different reason: they determine whether capability lives in the organization or in individuals, and capability that lives in individuals leaves with them.
How it matures
What matures here is where capability lives and whether any of it is spare. From Stalled through Steady the team grows from one overloaded operator into defined roles, but capacity stays fully consumed by running what exists, so nothing is left over to change anything. Slack first appears at Accelerating, and it is the hinge of the whole dimension, because a team with room to experiment can finally invest in its own operating system. By Compounding the team renews itself, continuously upskilling and absorbing new capability, AI included, as a matter of course.
Common failure patterns
Hiring only against the delivery backlog, which adds throughput at the current stage without adding the capability to leave it. A single person who knows how everything works, unbackfilled because they are always too busy to document. Core knowledge that lives at the agency, so the internal team routes work it cannot evaluate.
Related terms: digital team structure, marketing operations, capacity planning, key-person risk, upskilling.
Operational: how work and content move
Operating Model & Workflow
How work moves: sequentially and by hand, or in parallel with automation.
Why it governs velocity
The workflow is the denominator of the rate. The same piece of work takes a multiple longer moving through series handoffs than through parallel, self-service paths, and most of the difference is queue time rather than build time: work sitting between steps, waiting for the next function to pick it up. Improving any other dimension while leaving the workflow sequential pours capability into a pipe whose width has not changed.
How it matures
Everything through Steady is refinement of a sequential system. Tickets become an emerging process, the process becomes a repeatable cadence, and each step makes the sequence more reliable without making it shorter, because marketing and IT still work in series. The pivot comes at Accelerating, when parallel work changes the shape of the system rather than its polish. Compounding is the shape improving continuously, with cycle time compressing release over release.
Common failure patterns
Process improvement that adds approval steps, tightening control while lengthening the queue. A workflow optimized for preventing mistakes rather than for throughput, which prevents both. Automation applied to a broken sequence, which produces the same waiting faster.
Related terms: marketing workflow, content approval process, cycle time, self-service publishing, workflow automation.
Content Operations & Structured Content
Whether content is structured and reusable, or rebuilt from scratch each time.
Why it governs velocity
Reuse is the mechanism that makes velocity compound. Structured content is an asset that appears in the next channel at near-zero cost; unstructured content is an expense that gets paid again every time it moves. This dimension also underwrites the gate: personalization and AI both operate on content, and content without structure, metadata, and ownership cannot be governed when either is applied to it.
How it matures
The trajectory runs from content as expense to content as asset. Through Starting, content is rebuilt each time it is needed, so every use is paid for in full. Steady is the investment stage: structure, workflow, and taxonomy exist, but mostly in service of maintaining the current experience. The return arrives at Accelerating, when structure starts spanning channels and reuse begins to pay. At Compounding the content base is evergreen and the return compounds, because each delivery adds to a library that makes the next one cheaper.
Common failure patterns
A redesign that restructures the visual layer and leaves the content unmodeled, so the new site inherits the old operations. Copy locked inside pages, which turns every campaign and channel into a new build project. A taxonomy defined once at launch and never governed, drifting until nobody trusts it.
Related terms: structured content, content modeling, content operations, content supply chain, editorial workflow, taxonomy.
Technical: what the organization knows, and what it runs on
Measurement & Experimentation
Whether decisions rest on data and testing, or on opinion.
Why it governs velocity
Measurement sets the learning rate, and the learning rate is what velocity is for. An organization that ships faster without measuring ships wrong faster. Reporting alone describes the past; experimentation is what converts delivery speed into compounding returns, because each test changes what gets built next. Without this dimension, the curve's upper stages are motion rather than progress.
How it matures
What matures is the direction measurement faces. Through Steady it looks backward: first nothing is measured, then analytics exist but go untrusted, then reporting is dependable and describes what already happened. The pivot comes at Accelerating, when testing turns measurement forward and results start changing what gets built next. At Compounding it faces forward entirely, continuous and predictive, with personalization running at the scale the data supports.
Common failure patterns
Reporting theater: the dashboards are always current and have never changed a decision. A testing tool licensed at renewal and idle in between, because running tests requires workflow and capacity the organization has not built. Measurement owned by whoever built the dashboard rather than by the people making the decisions it should inform.
Related terms: digital analytics, experimentation program, A/B testing, conversion rate optimization, personalization.
Platform & Technical Health
Whether the platform is modern and composable, or brittle and carrying debt.
Why it governs velocity
The platform sets the ceiling on what is possible, and technical debt is a tax collected on every change regardless of size. A brittle platform makes small work large; fragile integrations make every release a risk event. Composability matters at the top of the curve because it determines whether capability is swapped and extended or rebuilt around. The ceiling is real, but it is only the ceiling: an adequate platform run by a capped organization still produces Steady velocity, which is why this dimension is one of six rather than the model itself.
How it matures
This dimension has no pivot. It improves in a straight line, each stage retiring more debt and adding more flexibility, until the platform can be swapped and extended rather than rebuilt around. The straightness is worth noticing, because it makes platform progress easy to mistake for velocity progress. Platform work is visible, budgetable, and steadily rewarding in a way operating-model work is not, and an organization can hold a platform of Accelerating quality inside an operating model that has never left Steady.
Common failure patterns
Replatforming as the answer to an operating-model problem, which produces a modern stack run in series. Debt retired once in a rebuild and re-accumulated on the same schedule, because the workflow that created it did not change. Integrations that work until the one person who understands them leaves.
Related terms: technical debt, composable architecture, digital experience platform, replatforming, integrations.
Governance, Risk & AI Readiness: the Safe Speed gate
Whether risk is managed and AI is governed, or absent and shadow-used. The gate determines when an organization can safely introduce AI-enabled speed.
Why it is a gate, not a dimension
The six dimensions measure how much velocity an organization can produce. The gate measures whether producing more of it is safe. It cuts across all six, because governance failures surface wherever speed does: in content shipped without review, in tests run without controls, in tools adopted without anyone accountable for their output. A dimension can be weak and simply cap velocity. The gate, when weak, converts velocity into exposure.
The gate across the curve
Through Steady, the gate accumulates thinking without action. Shadow use at Stalled means ungoverned tools already operating on ungoverned content, with no one accountable for the result. Curiosity without policy follows, and then the plateau's tell: use cases defined, reviewed, and never deployed, an organization that has done the thinking and cannot act on it. The pivot is deployment. At Accelerating, governance stops being a document and becomes operational, covering testing and carrying risk-managed AI into production. At Compounding the gate sets the pace, with an audit cadence that admits new capability, agentic included, only as fast as the organization can absorb it.
What the gate implies
AI readiness is reached through operating maturity rather than purchased, because AI multiplies whatever it is applied to, weaknesses included. The gate is why the model sequences AI into production at Accelerating and not before, and it cuts in both directions. An organization at Steady with a disciplined gate is closer to safe AI adoption than a faster-moving organization without one, because the discipline is the prerequisite and the speed is not. Read this way, the gate is not a brake on the curve; it is the reason the upper stages are worth reaching.
Common failure patterns
An AI policy that is a prohibition memo, which does not stop use but relocates it into the shadows the policy cannot see. Pilots that never deploy because no governed path to production exists, so review becomes where use cases go to age. An adoption program measured by tools purchased rather than by use cases governed.
Related terms: AI governance, AI readiness, shadow AI, AI policy, risk management.
How the dimensions interact
The dimensions are ordered by constraint, not by importance. The organizational dimensions constrain the operational ones, because workflow cannot be rebuilt without ownership and capacity to rebuild it. The operational dimensions constrain the technical ones, because measurement and platform capability only convert into velocity through the way work and content move. Improvement that ignores this ordering produces familiar results: a modern platform run in series, an experimentation license without an experimentation practice, a content model nobody has the slack to adopt. This is why placement against the matrix leads to a remediation sequence rather than a tool recommendation, and why the dimension furthest upstream of the constraint is usually where the work starts.
See how the stages read as whole patternsFrequently asked questions
Are all dimensions weighted equally?
The model treats the dimensions as a pattern rather than a weighted sum, and the matrix carries no weights. The Velocity Score and the Velocity Assessment apply C2's scoring to produce placement from that pattern, and the scoring, including question design, weighting, and stage thresholds, is proprietary and not published.
Which dimension should we fix first?
The one capping the pattern, which is usually the dimension sitting furthest behind the others, read with the constraint ordering in mind: ownership before workflow, workflow before tooling, because each constrains the next. The general heuristic breaks down in specific organizations, which is what an evidence-based placement is for. Identifying the capping dimension and the sequence to relieve it is the core output of the Velocity Assessment.
How does this relate to platform maturity models?
They grade different things and can be used together. A platform maturity model measures how fully an organization has adopted a product's capabilities. The Velocity Model measures the organization itself, which is why a team can score well on a platform model while sitting at Steady on the curve: adoption of features is not the same as the ownership, workflow, and governance to convert them into velocity.
Is Compounding realistic for a small team?
Yes, within a scope matched to the team's size. Compounding describes how a team works, not how large it is: a small team with reusable content, self-service workflow, and governed AI adoption compounds inside its surface area the same way a large one does. Small teams reach the stage by narrowing what they own rather than by adding headcount.
Where does your team sit on the curve?
The Velocity Score places you on the curve in about five minutes and names the dimension most likely to be holding you there.
For an evidence-based placement across all six dimensions, see the Velocity Assessment.