One person sees the whole and doles out the parts.
- Context is shared selectively
- People wait for their next assignment
- Issues travel upward before action
- Progress stops when the gatekeeper is unavailable
The Smaller Chunks Theory
Complexity is inevitable.
Confusion is not.
People do their best work when the work in front of them is small enough to understand, clear enough to act on, and visible enough to learn from.
The theory at a glance
Six principles. Choose your deep dive.
Start with the overview, explore a principle, or read the manifesto. Each chapter stands on its own; use the chapter tabs to move between them.
05 / Chunk Obeya theory · Transparent autonomy
Work held close becomes work held back.
Some people hold status, priorities, decisions, and client context close to their chest. Chunk Obeya creates an Obeya-style shared reality so the team can expose gaps, align decisions, and act without waiting for one gatekeeper.
The team sees the whole and pulls the next best action.
- Work, status, risks, and priorities are shared
- Decision boundaries are explicit
- Independent chunks move in parallel without collision
- The team adapts without waiting for work to be handed out
Autonomy requires sharedPurposePriorityBoundariesRelationshipsEvidence
Autonomy is not absence of accountability. It is accountability supported by enough shared context to act without unnecessary permission.
Safe parallelism: Parallel work does not mean individual multitasking. Focused people advance independent chunks at the same time because scope, ownership, dependencies, and decisions are visible. Work no longer waits for one person to dole it out, so the process becomes self-sustaining.
Limited attentionNo true multitaskingImperfect memoryHidden uncertaintyFinite coordination
A good process does not make one person the source of work. It makes the work visible enough for the team to become the source of solutions.
Companies that cannot see their work cannot make important decisions. Ultimately, they fail.
The Smaller Chunks Theory · Overview
Enable your team to work with greater autonomy, delivering value through small, incremental improvements along a clearly illuminated path.
Smaller does not mean trivial. A chunk is the smallest meaningful unit of progress that a team can understand, own, complete, and learn from without losing sight of the outcome.
About us
60 years of experience. One shared vision.
Our team brings a combined 60 years of experience writing software, testing software, and managing complex software projects for a Fortune 50 corporation. We have worked hands-on with Azure DevOps, CI/CD, pipelines, delivery timelines, visual metrics, and process visualization to help teams plan, build, test, and deliver.
That experience has shown us where the tools fall short. Too often, the focus is on developers, while value teams and project managers are left stitching together plans, spreadsheets, dashboards, and status reports to understand the whole picture.
We believe there should be a single solution that brings it all together: planning, development, testing, delivery, timelines, metrics, and visible processes. Smaller Chunks is built around that vision, giving value teams, project managers, and developers a shared path from intent to delivered value.
Turn ambiguity into a visible sequence of decisions.
One theory · Six implementations
Each product makes one part of the system visible.
The tools work independently, but together they connect intent, effort, relationships, delivery signals, collaboration, and release communication into one understandable flow of value.
Understand the workThenDeliver and learnVisible · measurable · adaptable
- 01Chunk Plan
Break down. Plan. Track. Deliver.
Define the change, who it serves, and the evidence of success; then shape explicit chunks ahead of and alongside implementation.
- 02Chunk Size
Estimate. Split. Deliver.
Compare the work early enough to align capacity, skills, and people before a delivery gap becomes unavoidable.
- 03Chunk Map
Plan. Connect. Deliver.
Connect outcomes, initiatives, chunks, owners, dependencies, and backlog work so impact and responsibility are visible.
- 04Chunk Plot
Visualize. Analyze. Deliver.
Use dashboards to monitor flow, work in progress, blockers, inbound dependencies, forecast drift, and quality so teams can decide before conditions become issues.
- 05Chunk Obeya
Align. See. Solve.
Use an Obeya-style shared view of value, flow, evidence, risks, decisions, owners, and countermeasures to adapt before drift becomes failure.
- 06Chunk Notes
Release. Communicate. Clarify.
Use backlog parent-child relationships to assemble completed chunks into release notes automatically, then review and publish them as a small final step.
01 / Chunk Plan theory · Value before backlog
Plan the value before implementation decides the plan.
Define the outcome, shape the value into explicit chunks, and make scope, sequence, dates, ownership, dependencies, and evidence visible before the backlog turns assumptions into commitments.
Separate but Equal
Plan ahead. Couple deliberately.
The value team must be able to plan in parallel with—and often ahead of—the implementation backlog. Chunk Plan keeps value planning independent, then connects every chunk to its delivery work through a simple Azure DevOps integration, regardless of project, team, process, or backlog structure.
Value planCoupledImplementation backlogAzure DevOps · any structure
A useful plan makes explicitOutcomeScopeSequenceDatesOwnersDependenciesEvidence
The plan is not the backlog. It is the value model that tells the backlog what its work is meant to deliver.
- 01Intent
Frame the outcome
State the change, who it serves, and the evidence that will prove it worked.
- Name the customer or team who benefits.
- Describe the change they should experience.
- Agree on observable success criteria.
- 02Boundary
Shape value chunks
Break the outcome into meaningful increments that can be understood and delivered independently.
- Define what is inside and outside each chunk.
- Keep each increment useful on its own.
- Make completion small enough to verify.
- 03Path
Sequence relationships
Expose prerequisites, dependencies, milestones, and the order in which value can safely emerge.
- Identify prerequisites and blocking dependencies.
- Order milestones around usable outcomes.
- Show where work can proceed in parallel.
- 04Commitment
Make ownership and dates explicit
Record who moves each chunk, when it is needed, and which assumptions support the timeline.
- Give each chunk an accountable owner.
- Distinguish needed dates from forecasts.
- Record assumptions behind each commitment.
- 05Coupling
Connect plan to implementation
Relate every value chunk to its Azure DevOps work without forcing a new backlog hierarchy.
- Link value chunks to Azure DevOps work items.
- Keep existing project and team structures.
- Trace delivery progress back to the outcome.
- 06Learning
Revise with evidence
Keep the outcome stable while delivery evidence improves boundaries, sequence, scope, and forecast.
- Compare delivered results with success criteria.
- Update scope and forecasts as evidence changes.
- Make the next planning decision explicit.
02 / Chunk Size theory · Estimable work
Estimate effort. Establish cadence. Forecast delivery.
Chunk Size gives the people doing the work a shared way to estimate the effort of each chunk. As chunks finish, management can use the team’s observed cadence to forecast completion and organize capacity, skills, and teams before delivery is at risk.
Team sizing loopEvidence based
- Boundary
- Agree on the chunk being estimated.
- Compare
- Size effort relative to work the team understands.
- Uncertainty
- Expose assumptions, unknowns, and differing opinions.
- Commit
- Record the team’s shared effort estimate.
- Observe
- Measure how much estimated work actually completes.
- Forecast
- Use observed cadence to estimate a completion range.
- Align
- Adjust capacity, skills, and team composition early.
The people doing the work size it.
Conversation exposes complexity and uncertainty that top-down estimates miss.
Comparison creates a usable signal.
A shared scale makes effort comparable without pretending it is precise time.
Cadence turns estimates into forecasts.
Completed effort over time supports a realistic range—not a false promise.
Forecasts make capacity actionable.
Management can align people, skills, and teams while there is still time to adapt.
Chunk Size connects team judgment to management decisions without turning estimates into individual scorecards.
04 / Chunk Plot theory · Visual decision system
See the signal. Decide before it becomes an issue.
Chunk Plot makes the delivery process transparent through metrics that expose flow, load, friction, and drift. Teams can see conditions changing, understand what is affected, and act while options still exist.
Signal → context → decision, while options still exist.
Work stops moving
Compare current age with normal cycle time. Unblock, swarm, split, resequence, or stop the chunk before delay spreads.
Work in progress rises
Expose competing demand before focus collapses. Finish existing work, reduce intake, or add targeted capacity.
Blocked time accumulates
Show the owner, cause, duration, and affected work. Resolve, reroute, or escalate before the needed-by date is missed.
Delivery rate changes
See whether completed effort is trending outside its normal range. Investigate scope, staffing, quality, or system constraints.
Completion begins to drift
Update the likely range early. Rebalance teams, change sequence, reduce scope, or reset the commitment while choices remain.
Bug arrival accelerates
Compare bug arrival with resolution and severity. Reserve capacity, slow intake, or stop the line before instability compounds.
Transparent by design
Every metric must lead to a decision.
- SignalWhat changed, and is it a trend?
- ContextWhich chunks, outcomes, dates, and teams are affected?
- ThresholdWhen does the condition require attention?
- DecisionWho can act, and what options remain?
03 / Chunk Map theory · Editable relationships
See every work item. Edit every link.
Chunk Map turns work-item links into an editable visual map. Follow parent-child structure, trace related work across branches, and correct links without losing context.
One view / Editable work map
Understand the structure. Change it in place.
Select any story to see its parent, children, and related work. Add, change, or remove links from the same view so the map stays aligned with the work.
- Parent / childTrace the hierarchy from outcomes and stories to delivery tasks.
- RelatedExpose work that matters across branches without forcing a false hierarchy.
- Edit linksAdd, change, or remove relationships while their impact is visible.
Action: Select a work item, inspect every link, and fix the relationship in place.
04 / Chunk Plot theory · Metrics and decisions
Metrics are decision instruments.
Metrics make invisible conditions observable. They should tell a team where attention is needed, which assumption changed, and what decision must be made—without turning estimates into promises or people into scorecards.
Metrics do not replace judgment.
They make judgment better informed. A metric earns its place only when the team knows what question it answers, what context it needs, and what action a meaningful change could trigger.
Cycle time
How long does a chunk take from active start to verified completion?
Work in progress
How many chunks are competing for attention at the same time?
Blocked time
Where and how long does work wait for a decision, dependency, or environment?
Completion predictability
How often do chosen chunks finish within the planning horizon?
Scope change
How often does a chunk expand after work begins, and what did we misunderstand?
Outcome evidence
Did the completed chunk create the behavior, learning, or result it was intended to create?
Management questionEvidence to examineDecision it supports
Work in progress, aging, cycle time, blocked time
Finish, unblock, split, or stop work before adding inventory.Throughput range, scope stability, dependency confidence
Forecast a range, change the sequence, reduce scope, or decline.Bug arrival and resolution rates, severity, escapes, reopens
Investigate, reserve capacity, slow intake, or stop the line.Outcome evidence, adoption, behavior change, cost of delay
Continue, adapt, pivot, or stop based on observed value.A vital guardrail
Never use a single metric as a target. When a measure becomes a quota, people optimize the number instead of the system. Read measures together, look for trends, and ask better questions.
05 / Chunk Obeya theory · The early warning system
Visibility should expose trouble while there is still time to act.
A useful work system does more than describe what has already happened. It makes weak signals visible, connects them to the work they threaten, and gives the team a clear next decision.
SignalWhat changed?
MeaningWhat is now at risk?
DecisionWhat will we do next?
Signal 01
Expose stuck work
Work can appear active long after movement has stopped. A visible system should distinguish healthy progress from silent aging.
Make visible
- Current age compared with expected cycle time
- Time since the last meaningful change
- Blocker, owner, and next unblock action
- Downstream chunks and dates now at risk
Decision: unblock it, split it, swarm it, change the sequence, or stop it. Never let “in progress” become a hiding place.
Image placeholder 03Expose Stuck Work
Suggested visual: a delivery flow where one aging red chunk blocks a growing queue, with age, last movement, owner, and impacted work called out.
Signal 02
Expose inbound dependency completion
A plan is only as reliable as the work it expects from outside the team. Dependencies must be treated as tracked commitments, not footnotes.
Make visible
- The provider, required outcome, and needed-by date
- Evidence of completion—not a percentage or verbal assurance
- Confidence, last confirmation, and current forecast
- Every chunk that cannot start or finish without it
Decision: confirm the handshake, resequence work, create a fallback, reduce scope, or escalate before the dependency becomes a surprise.
Image placeholder 04Inbound Dependency Radar
Suggested visual: external commitments flowing toward a team plan, showing completed evidence in green, uncertain forecasts in amber, and a late dependency threatening connected chunks.
Signal 03
Expose bug upticks
A rise in defects is not merely a support statistic. It can be the earliest evidence that delivery speed, product quality, or system stability is deteriorating.
Make visible
- New bugs against the normal baseline and rolling trend
- Severity, affected customers, component, and release
- Reopened defects and escapes from recent work
- Bug arrival rate compared with resolution rate
Decision: investigate the change, connect bugs to the responsible release or component, reserve capacity, slow new intake, or stop the line.
Image placeholder 05Bug Uptick Early Warning
Suggested visual: a quality signal rising above its normal band, tied directly to a recent release, affected chunks, severity mix, and the growing gap between bugs opened and resolved.
The operating rule
No critical signal should exist without context, ownership, and an available decision. A red indicator that cannot change behavior is decoration, not visibility.
06 / Chunk Notes theory · Backlog-native release communication
Release notes should come from the work—not become more work.
Completed backlog items already contain the structure of the story. Chunk Notes follows parent-child relationships, gathers the completed chunks beneath each outcome or feature, and turns that delivery evidence into release notes automatically.
Release communication belongs inside the delivery process. Generate it from the backlog, review it as a team, and publish it as one more small chunk.
Keep the hierarchy meaningful
Connect implementation work to its parent chunk, feature, initiative, and intended outcome in the backlog.
Let delivery create the evidence
Completion status, acceptance evidence, links, and child work identify what actually shipped.
Generate the release narrative
Group completed children beneath their parent value and transform technical records into a structured draft.
Review one small final chunk
Confirm wording, audience, and evidence, then publish without rebuilding the release by hand.
The automated notes contract
Release notes should be:
- Backlog nativeBuilt from the work records teams already create and complete.
- Relationship awareGrouped through parent-child links so details remain connected to value.
- Completion drivenBased on verified finished work rather than memory or meeting notes.
- Audience readyTechnical child items summarized in language appropriate for customers and teams.
- Human reviewedAutomated drafting with a small, explicit approval step before publication.
- TraceableEvery statement links back to the backlog evidence that produced it.
Let relationships and completion evidence find, group, and structure the work.
People confirm meaning and audience; they should not spend days reconstructing delivery history.
If producing release notes requires a separate reporting project, they are not integrated deeply enough into the backlog.
09 / What this is not
Smaller chunks are not smaller thinking.
The method fails when decomposition becomes fragmentation. The outcome must remain visible while the next unit of action becomes clearer.
Breaking work into meaningless tasks
Activity without an observable result creates motion, not progress.
Estimating with false precision
Sizes support comparison and conversation; they are not certainty disguised as math.
Planning every detail up front
Plan enough to move safely, then let evidence improve the next decision.
Maximizing individual utilization
A full calendar is not flow. Optimize the movement of value across the whole system.
Hiding the relationship to strategy
Every chunk needs a visible line back to the outcome that justified it.
Using tools as a substitute for judgment
The tool should preserve context and reveal tradeoffs. People still make the decision.
The Smaller Chunks Manifesto
Six products. One visible path from intent to value.
01
Chunk PlanTurn outcomes into explicit chunks. Plan ahead, then couple intent to implementation without surrendering either.
02
Chunk SizeEstimate effort together. Make uncertainty visible so capacity can align before delivery is at risk.
03
Chunk MapMake every relationship visible. Connect outcomes, work, owners, dependencies, and backlog implementation.
04
Chunk PlotMonitor the process. Make the decision. Use dashboards to expose changing delivery conditions and act before they become issues.
05
Chunk ObeyaClose delivery gaps together. Give the team one Obeya-style view for evidence, risks, decisions, owners, and countermeasures.
06
Chunk NotesGenerate release notes from the backlog. Use parent-child relationships to assemble completed work, then review and publish it as one small final chunk.
Estimate it.
Plan it. Connect it.
Show it.
Deliver it. Share it.
The promise
Less held in memory.
More shared in reality.
The goal is not to make important work feel small. It is to give people a practical way to move through complexity together.
See the tools built around the theory