Request a briefing

Essay · Portfolio Strategy

Let the Work Move: Flow Efficiency and Portfolio Choices for Leadership Teams

How leadership teams get customer work finished sooner by starting less of it.

Let the Work Move is an essay on flow efficiency and portfolio management for leadership teams whose people are fully booked while customer work waits. It shows how limiting work in progress, finding where failure demand starts and measuring lead time get value to customers sooner.

A leadership team that tries to keep every person busy all the time makes customer work wait. Full utilization of each specialist is a local optimum: each person looks productive while requests pile up in the queues between them. Flow efficiency vs resource efficiency is a choice about whether the people or the work waits. Managing for resource efficiency, the share of each specialist's available time in use, leaves requests waiting for whoever comes free. Managing for flow efficiency, the share of a request's elapsed time in which someone adds value to it, sometimes leaves people idle so that work does not pile up. In our experience, leadership teams that manage their portfolio for flow deliver more customer work with fewer projects active.

Flow efficiency falls when silos and utilization targets make the work wait

Many enterprises are organized for resource efficiency, with specialists in functional silos. Each hand-off between silos adds a queue, and a request can spend days waiting for hours of work.

In our experience, fully booked teams keep starting new work. When a system's finishing rate stays steady, the average time an item spends inside it rises in proportion to the average work in progress. That relationship describes averages across the system, and a single project's finish date needs its own forecast. When requests arrive unevenly and vary in size, waiting also climbs steeply as utilization nears one hundred percent, because no spare capacity remains to absorb a burst. A highway at full use is a traffic jam, and in a jam even an emergency vehicle waits.

In our view, keeping everyone fully utilized has hardened into orthodoxy, and most organizations still manage for maximum utilization. The utilization rule grew up on the factory floor, where scientific management set a standard output for each worker and machine, and cost accountants priced idle machine hours. Management accounting made those standards routine measures of efficiency, and in our experience business courses still teach them to new managers, many of whom apply them to offices and product teams.

The utilization rule still fits in two places: at the constraint, the step or team whose capacity sets the pace for the system, and in repeatable work with steady demand, where little variation produces little queueing. Work that requires learning arrives unevenly and is partly new each time, so keeping people fully booked there lengthens the queues.

Failure demand competes with value demand for the same people

Some queued work exists because something earlier went wrong. Value demand is what customers come to an organization for. Failure demand is the work that comes back when something a customer needed never arrived, or arrived broken, unclear or incomplete, such as a customer service call about a confusing bill. The caller still deserves an answer. Each hour spent answering failure demand is an hour taken from value demand, and removing the cause returns that hour.

A value stream map, which traces the work from a customer's need to the moment that need is met, shows where the work waits and often where the errors behind failure demand begin. In our experience, much of the waste sits in hand-offs between functions that no single manager owns.

A portfolio exists to make the trade-offs visible

A leadership team makes its trade-offs in the portfolio, where projects draw on one limited pool of people. A portfolio that funds every request is a wish list. The portfolios we have seen work best keep every commitment on one wall the whole team sees, with a rule agreed in advance for which project gets a specialist both need. At each portfolio review, leaders give every project one of three answers: staff it fully, end it and release its people, or recharter it with a new scope. A project worth doing that no one can staff yet waits in a visible queue, outside the active count, until a team comes free.

Projects that share specialists finish sooner when fewer are active. With ten active, each specialist switches among them and all ten progress slowly. When three are active and seven wait, the three finish and free their people. The constraint sets how many projects can be active at once, so leaders shield it from interruptions and from work that is unnecessary or not ready. Maintenance, learning and recovery still earn a place in the constraint's time. To keep the constraint supplied, every other step needs spare capacity to recover from delays.

Limiting work in progress lets leaders choose the slack an organization needs to adapt

A work-in-progress limit caps how many items are under way, so new work starts only when earlier work finishes. The limit makes overload visible, and leaders decide how much slack it leaves. Set below the load that would keep everyone busy, the limit leaves room to handle a surprise, fix a defect at its source or try a better way of working. Smaller batches help too: they finish sooner, bring downstream feedback earlier and put less at risk in each release, down to the size where the cost of preparing each release outweighs those gains.

Cost of delay puts a price on waiting: the value lost for each week the work stays unfinished. When finished work frees a place under the limit, divide each waiting item's cost of delay by the weeks of constraint time it needs and start the highest result first. The ratio suits independent items whose cost of delay stays steady from week to week, and a fixed deadline or a dependency between items can change the order.

Lead time shows how long value takes to reach the customer

Lead time, our headline measure of flow, runs from the decision to create value to the moment a customer receives it. We also count how long a request waits before anyone accepts it, so moving the starting line cannot hide delay. Measuring lead time changes the question managers ask. A manager watching resource utilization asks who has spare capacity. A manager watching lead time asks the team, "What can I do to help your throughput?" then clears the blockers.

Flow Thinking improves the whole system one small experiment at a time

Flow Thinking, as we use the term, is a relentless focus on the flow of value through an organization to the customer. The practice draws on four traditions: Lean from the Toyota Production System, Agile from software development, the Theory of Constraints and design thinking. Its first steps are to make all the work visible, map how value reaches the customer and measure where the work waits, so the work-in-progress limit fits the real queues. Teams then set the limit and improve through small experiments, checking after each one whether lead time fell and whether defects and failure demand stayed level. We have watched teams pour concrete around their first design for flow and turn it into the next orthodoxy. Keep the map, the limits and the measures open to revision as the constraint shifts.


The workshop Portfolio Strategy, Flow and the Constraint helps leadership teams make these portfolio choices on their own work. Request a briefing.

Questions leaders ask about flow efficiency

What is the difference between flow efficiency and resource efficiency?

Resource efficiency is the share of each person's available time in use. Flow efficiency is the share of a request's elapsed time in which someone adds value to it. Managing for resource efficiency keeps specialists booked while requests queue. Managing for flow limits work in progress, so people sometimes wait for work. In our experience, leaders who choose flow finish more work by starting less.

Why does keeping everyone fully utilized slow work down?

Full utilization slows work for two reasons. When requests arrive unevenly and vary in size, waiting time climbs steeply as a team nears full capacity, because no spare time remains to absorb a burst. In our experience, fully booked teams also keep starting new work, and at a steady rate of finishing, every added item in progress stretches the average time an item spends in the system.

How can a leadership team limit work in progress across its portfolio?

A leadership team limits work in progress by deciding, at each portfolio review, which projects are active and which wait in a visible queue. The team first makes all the work visible, including failure demand. The team then caps active projects at what its constraint, the step that sets the pace, can finish, and ranks waiting work by cost of delay divided by its duration, checking deadlines and dependencies.

How do you calculate flow efficiency?

Flow efficiency is the time someone spends adding value to a request divided by the request's total elapsed time, shown as a percentage. For example, a request that takes ten working days from arrival to delivery and receives one day of work has a flow efficiency of 10%. The other nine days are waiting time, and a value stream map shows where that waiting happens.

References (11)

This essay is developed from William Haas Evans's manuscript Designing Resilience: The Strategy, Structure, and Spirit of Enterprise Agility.

  • Niklas Modig and Pär Åhlström, This Is Lean: Resolving the Efficiency Paradox, Rheologica Publishing, 2012. The pairing of resource efficiency and flow efficiency.
  • John Seddon, Freedom from Command and Control: A Better Way to Make the Work Work, Vanguard Consulting, 2003. The terms value demand and failure demand.
  • Mike Rother and John Shook, Learning to See: Value-Stream Mapping to Add Value and Eliminate Muda, Lean Enterprise Institute, 1998. https://www.lean.org/store/book/learning-to-see/
  • Johanna Rothman, Manage Your Project Portfolio: Increase Your Capacity and Finish More Projects, Pragmatic Bookshelf, 2009. The second edition appeared in 2016. The decision to commit to, kill or transform a project at each portfolio review, and a parking lot for projects not yet staffed. https://www.oreilly.com/library/view/manage-your-project/9781680500134/
  • Eliyahu M. Goldratt and Jeff Cox, The Goal: Excellence in Manufacturing, North River Press, 1984. Later editions are subtitled A Process of Ongoing Improvement. The constraint that sets the pace of a whole system, and the spare capacity every other step needs. https://openlibrary.org/books/OL2540561M
  • Preston G. Smith and Donald G. Reinertsen, Developing Products in Half the Time, Van Nostrand Reinhold, 1991. Putting a price on time in product development.
  • Donald G. Reinertsen, The Principles of Product Development Flow: Second Generation Lean Product Development, Celeritas Publishing, 2009. Cost of delay, ranking work by cost of delay divided by duration, and the economics of batch size.
  • Wayne E. Smith, "Various Optimizers for Single-Stage Production," Naval Research Logistics Quarterly 3, no. 1 and 2 (1956), pages 59 to 66. The ratio rule for sequencing jobs on one resource, which ranking by cost of delay divided by duration applies. https://econpapers.repec.org/RePEc:wly:navlog:v:3:y:1956:i:1-2:p:59-66
  • John D. C. Little, "A Proof for the Queuing Formula: L = λW," Operations Research 9, no. 3 (1961), pages 383 to 387. https://doi.org/10.1287/opre.9.3.383
  • J. F. C. Kingman, "The Single Server Queue in Heavy Traffic," Proceedings of the Cambridge Philosophical Society 57, no. 4 (1961), pages 902 to 904. The steep rise in waiting time as utilization nears full capacity when arrivals and task times vary. https://doi.org/10.1017/S0305004100036094
  • H. Thomas Johnson and Robert S. Kaplan, Relevance Lost: The Rise and Fall of Management Accounting, Harvard Business School Press, 1987. The origins of efficiency measures in scientific management and cost accounting.

William Haas Evans

Founder and chief strategy officer of Fugue Strategy Advisors. He works with leadership teams on strategy design, strategic foresight and organization design. More about William

Read next

Request a briefing

What is standing between your team and its goals this year?

Tell us in a 30-minute briefing, and we'll help you decide where to start.

Request a briefing