Skip to content

Innovation Time

Recurring protected-time process — instantiates Slack Capacity Design

Reserves recurring, self-directed work time for exploration and prototyping, so new-value creation is not permanently crowded out by delivery pressure.

Innovation Time carves out capacity for the one kind of work that a full backlog always defeats: discretionary exploration whose payoff is uncertain. Its defining move is that the protected capacity is self-directed and off the delivery ledger — people choose what to pursue, and the time cannot be reclaimed ticket by ticket when the sprint gets tight. The purpose is to generate options and new value, which sets it apart from slack meant to absorb variation or restore the system: nobody is buffering a shock or repairing anything here, they are making things that did not exist. Without the protection, exploratory work loses every collision with a committed deliverable, forever.

Example

A product-engineering org runs at full tilt against a quarterly roadmap, and everyone agrees "we should experiment more" while no one ever does. They declare every second Friday open: the day is calendar-blocked, taken off the sprint board, and owes no deliverable. An engineer irritated by flaky CI spends three successive Fridays prototyping a test-flakiness detector. It would never have survived triage as a backlog item — it always lost to the next feature — but on protected time it grows into an internal tool that cuts rerun waste for every team. The value came entirely from capacity that routine prioritization could not have allocated. (The pattern is the one popularly associated with Google's "20% time" and 3M's "15% culture" — real practices, cited here only as named examples.)

How it works

The distinguishing features are three: the time is pre-committed and recurring (a standing fraction, not a someday-maybe), it is self-directed (the worker picks the work, so genuinely novel bets get made), and it is shielded from delivery accounting so it cannot be quietly converted back into feature work under pressure. Output is optionality — prototypes, spikes, improvements — not a scheduled result.

Tuning parameters

  • Fraction of capacity — anywhere from 5% to 20%. More exploration, but more drag on committed delivery; set it to how much your value actually depends on discovery.
  • Cadence and batching — a day every week keeps the habit alive; a full week per quarter protects deep work. Distributed sustains momentum; batched enables things a single day can't finish.
  • Direction latitude — fully open versus themed to a strategic area. More latitude yields more surprise and less alignment.
  • Output visibility — whether explorations must surface at a demo day. Visibility adds accountability but can chill the riskiest bets.

When it helps, and when it misleads

It helps precisely where value comes from discovery that never survives contact with a prioritized backlog — the improvement no one would fund as a ticket. Its central failure mode is erosion: innovation time that exists on paper is the first thing sacrificed every crunch, so it quietly becomes zero exactly when delivery pressure is highest. The classic misuse is running it backwards — relabeling ordinary catch-up or cleanup as "innovation time" to look generous without protecting anything. The discipline that guards against both is structural protection (keep it off the delivery ledger) plus measuring whether the time is actually taken, not merely promised. The deeper argument is DeMarco's: an organization run at total utilization has no capacity left to reinvent itself.[1]

How it implements the components

  • slack_purpose — its purpose is explicit and specific: exploration, experimentation, and option creation, not shock absorption.
  • protected_capacity — the recurring, calendar-blocked hours held off routine demand are the reserve itself.
  • slack_type_catalog — it supplies a particular type of slack: self-directed attention and time aimed at creative work, distinct from material, budget, or staffing reserve.

It does not operate the release-and-governance components release_condition, replenishment_rule, or slack_use_log — those are Slack Release Review — nor utilization_ceiling and slack_visibility_metric, which belong to Utilization Ceiling Dashboard.

  • Instantiates: Slack Capacity Design — Innovation Time is the exploration-purpose reserve within the broader slack system.
  • Sibling mechanisms: Learning Time · Schedule Slack · Maintenance Window · Buffer Resources · Capacity Pool · Reserve Staffing · Slack Release Review · Surge Roster · Unallocated Budget · Utilization Ceiling Dashboard · Emergency Reserve

Notes

Easy to confuse with Learning Time: both protect recurring time, but innovation time is for creating new value while learning time is for absorbing capability. A team that only ever "explores" during learning blocks has neither.

References

[1] Tom DeMarco, Slack: Getting Past Burnout, Busywork, and the Myth of Total Efficiency (2001) — argues that an organization scheduled to 100% utilization retains no capacity to change or reinvent itself, so some deliberate slack is a precondition for innovation rather than a waste of it. ↩