A systems archetype from Peter M. Senge's The Fifth Discipline (1990), developed by Daniel H. Kim (Systems Archetypes I, 1992) and John D. Sterman (2000).
Describes how a fast symptomatic fix keeps getting used in place of a slow fundamental one, while the ability to make the real fix withers.
Contractors keep the queue short while the team's own skill drains away
At rest, contractors cover 80 percent of the gap: the queue stays short (B1) while the team's own capability falls (R1). Commit staff to training whatever the queue says, and Play: the slow fix (B2) takes over and the contractors are no longer needed. Illustrative.
Work waiting, % of demand
Work done by contractors, %
Work the team can do itself, %
Month 24 of 24
- Work waiting, % of demand
- 11%
- Work done by contractors, %
- 42%
- Work the team can do itself, %
- 47%
- B1, Quick fix
- Work waits, contractors come in, and the queue shortens within a month. One negative (−) link: an odd count, so the loop balances.
- B2, Slow fix
- Work waits, staff are trained, and months later the team can do more itself, so less work waits. One negative (−) link: an odd count, so the loop balances.
- R1, Side effect
- Contractors doing the hard work drain the team's own skill, so more work depends on contractors, and the slow fix gets no pressure to start. Two negative (−) links: an even count, so the loop reinforces.
- Work waiting
- The symptom: data work requested but not done this month, as a share of the month's demand. It starts at 40 percent after the move to the new cloud platform.
- Contractor share
- The quick fix: contractors hired to cover part of the gap between demand and what the team can do itself.
- Staff in training
- The slow fix: employees learning the new platform. Some start when work piles up; more start when the committed number on the slider says so. Each takes about six months to become productive.
- Team's own capability
- The share of the work the team can do itself. It grows as trainees finish, and it shrinks while contractors do the hard work, because the team loses practice and the people who knew the work leave.
An example
A freight company moved its data work to a new cloud platform, and 40 percent of the requests from the business started waiting a month or more. The team had never used the new tools. The CIO approved contractors to cover most of the gap, and within a month the queue was short again.
Two years later the queue was still short, but contractors were doing 42 percent of the work and the team could handle only 47 percent on its own, down from 60. The two engineers who knew the old pipelines had left, and every hard job went to a contractor because only contractors had done one before. Training had been approved twice and postponed both times for a deadline. The new head of data drew the loops and ran the numbers back: had one person a month gone into training from the start, the contractors would have been unnecessary by month nineteen. She kept them, paired each one with an employee, and put one person a month into training whatever the queue said.
Why it works
The two fixes compete for the same signal. Both are set off by the symptom, and the quick one answers in weeks, so it wins every time the pain flares. Each win lowers the pressure that would have started the slow one, and the quick fix often erodes the very capacity the slow one would build. Nobody decides to become dependent; it arrives one sensible decision at a time.
The map makes the trade visible. Once the side effect is drawn, the quick fix's price includes the capability it drains, and the slow fix stops looking like a luxury. The map also points to the remedy: set the slow fix's pace on a schedule that does not depend on the symptom, and keep the relief running while it matures.
The thinking behind it
Peter M. Senge Peter M. Senge set out shifting the burden among the system archetypes in the appendix of The Fifth Discipline (1990). He drew it as two balancing loops that both relieve the same symptom, one fast and symptomatic, one slow and fundamental, joined by a side effect through which the fast loop weakens the slow one. The book places the archetypes inside systems thinking, which it presents as the discipline that ties together the other four of a learning organization.
Daniel H. Kim Daniel H. Kim, a writer and teacher in organizational learning, turned the archetypes into diagnostic templates in Systems Archetypes I: Diagnosing Systemic Issues and Designing High-Leverage Interventions (Pegasus Communications, 1992), and later, with Virginia Anderson, in Systems Archetype Basics: From Story to Structure (1998). His guides give a team a drawing to fill in and a way to move from the story it tells about a recurring problem to the structure that keeps producing it.
John D. Sterman John D. Sterman treats the structure in Business Dynamics (2000), the standard system dynamics textbook, which builds such archetypes from stocks, flows, feedback loops, and delays. Read that way, the team's capability is a stock: it fills slowly through training and drains while someone else does the work, which is why the decline can run for two years before anyone names it. His beer game experiments showed how badly managers read delays of this kind.
How to run it
- Name the symptom. Pick the number that keeps coming back, such as work waiting, defects, or missed dates.
- Name both responses. Write down the quick relief the organization reaches for and the slower fix that would remove the cause.
- Count which one gets used. Go back through the last year and note which response was chosen each time the symptom flared.
- Find the side effect. Ask how the quick relief makes the slow fix harder: lost skill, lost budget, lost urgency, or lost people.
- Fund the slow fix on a schedule. Commit its pace in advance, so it does not depend on how loud the symptom is this month.
- Keep the relief while it matures. Pulling the quick fix early brings the symptom back and turns people against the slow fix.
- Wind the quick fix down in steps. Tie each step to a measure of restored capability, such as the work the team now finishes alone.
Where it goes wrong
The first failure is banning the quick fix. Contractors, workarounds, and emergency budgets exist because the symptom is real, and removing them before the slow fix has matured brings the pain back and discredits the slow fix. Keep the relief and change what it costs the long term, for example by pairing every contractor with an employee.
The second is a slow fix that exists only on paper. Training that is approved and then postponed for each new deadline is the archetype running unchanged; spot it by counting how many people finished training this quarter. The third is treating the side effect as bad luck, so that each departure or faded skill is explained on its own and the pattern behind them never gets named.
When to use something else
- Fixes That Fail. When the quick fix itself brings the problem back through a delayed side effect, and no slower remedy is in view.
- Eroding Goals. When the pressure is being relieved by lowering the target instead of by any fix at all.
- Iceberg Model. When the team still sees only the recurring events and has not yet named the structure that produces them.
Where it sits in SOLVE eX
Go to the source
- The Fifth Discipline: The Art and Practice of the Learning Organization, Doubleday/Currency, 1990
- Systems Archetypes I: Diagnosing Systemic Issues and Designing High-Leverage Interventions, Pegasus Communications, 1992
- Business Dynamics: Systems Thinking and Modeling for a Complex World, McGraw-Hill, 2000