Contents
Use this when the same problem keeps returning despite repeated fixes — when growth stalls for reasons nobody can articulate, when well-intentioned interventions make things worse, or when teams are trapped in patterns they can feel but can't name. System archetypes are recurring structural patterns that show up across industries, scales, and centuries. Learn eight to ten of them and you'll start recognising the architecture of dysfunction before it fully manifests.
Section 1
What This Tool Does
A startup scales aggressively, hiring fast to meet demand. Quality drops. Customers churn. The company hires faster to replace lost revenue. Quality drops further. Everyone works harder. Nobody asks why working harder keeps making things worse. A city builds a new highway to relieve congestion. Traffic improves for eighteen months. Then it's worse than before — the new capacity induced new demand, and now there's a wider highway full of more cars. A pharmaceutical company discovers that its blockbuster drug is losing market share. It increases the sales force. Margins compress. It cuts R&D to protect margins. Five years later, there's no pipeline. The sales force has nothing new to sell.
These aren't three different problems. They're the same problem wearing different costumes. Peter Senge, drawing on decades of system dynamics work at MIT — Jay Forrester's pioneering simulations, Donella Meadows' leverage point analysis, the modelling tradition that stretches back to the 1950s — codified roughly a dozen recurring structural patterns in The Fifth Discipline in 1990. He called them system archetypes. Each one describes a specific configuration of feedback loops, delays, and unintended consequences that produces a characteristic behaviour over time. Fixes that Fail. Shifting the Burden. Limits to Growth. Tragedy of the Commons. Escalation. Success to the Successful. Eroding Goals. Growth and Underinvestment. The names are plain because the patterns are universal.
The mechanism is pattern recognition at the structural level. Most diagnosis operates on events — what happened, who did what, what went wrong last quarter. Slightly better diagnosis operates on trends — the error rate has been climbing for six months. System archetypes operate on the level of structure: the arrangement of feedback loops, delays, and incentives that generates those trends and events. This is the cognitive shift. You stop asking "what happened?" and start asking "what structure would produce exactly this pattern of behaviour?" Once you identify the archetype, you inherit decades of accumulated knowledge about where the leverage points are — and, critically, where the obvious interventions will backfire.
The reason human intuition fails here is that most systemic structures contain delays and feedback loops that make cause and effect non-obvious. The action that caused the problem may have occurred months or years ago. The intervention that feels most urgent — hire more people, cut costs, add capacity — often reinforces the very dynamic it's trying to fix. Senge's archetypes are a library of these traps. They function like a diagnostic manual: you observe the symptoms, match them to a known pattern, and the pattern tells you both why your current approach isn't working and what a structural intervention would look like. You don't have to derive the system dynamics from scratch every time. Someone has already mapped the territory.
The practical power is in the speed of diagnosis. A team that knows the archetypes can look at a deteriorating situation and say, within minutes, "This is Shifting the Burden — we're treating the symptom with a quick fix that's atrophying our capacity to address the fundamental problem." That sentence, spoken in a meeting, changes the entire conversation. It moves the group from arguing about tactics to examining structure. That's the whole trick.
Section 2
How to Use It — Step by Step
Instructions on the left. Worked example — a B2B SaaS company whose growth has plateaued despite increasing marketing spend — on the right.
Step 1 — Observe
Map the problematic behaviour over time
Before reaching for any archetype, document what's actually happening. Plot the key metric over time — not a snapshot, a trajectory. What's the shape? Is it oscillating? Declining despite effort? Growing then flattening? The temporal pattern is your first diagnostic clue. Also note: what interventions have already been tried, and what happened after each one? If the problem returned or worsened after a fix, that's a structural signal.
Worked example
SaaS growth plateau
CloudSync, a mid-market project management SaaS, grew MRR from $200K to $1.8M over 18 months. Then growth stalled. Over the past two quarters, MRR has hovered between $1.7M and $1.9M despite a 40% increase in marketing spend. The team's response: double down on paid acquisition. CAC has risen from $380 to $620. Net revenue retention has quietly dropped from 115% to 94%. The shape: rapid growth, then a plateau that doesn't respond to the lever being pulled.
Step 2 — Match
Compare the behaviour pattern to known archetypes
Review the core archetypes and ask: which structural pattern would produce exactly this behaviour? Key signatures to look for: Limits to Growth — early success that slows as a constraint binds. Fixes that Fail — a quick fix that works temporarily but triggers side effects that recreate or worsen the original problem. Shifting the Burden — a symptomatic solution that works so well in the short term that the fundamental solution is never pursued, and the capacity to pursue it atrophies. Growth and Underinvestment — growth that stalls because the organisation fails to invest in capacity ahead of demand. Don't force a match. Sometimes two archetypes operate simultaneously.
Worked example
Two archetypes at work
CloudSync's pattern matches Limits to Growth: the reinforcing loop of "more marketing → more trials → more revenue" ran into a balancing loop — the product's onboarding experience can't handle the volume of less-sophisticated users that paid acquisition attracts. Trial-to-paid conversion has dropped from 22% to 11%. But there's a second archetype layered on top: Shifting the Burden. The "symptomatic solution" is increasing ad spend. The "fundamental solution" is fixing onboarding and investing in customer success. Every dollar shifted to paid acquisition is a dollar not spent on the fundamental fix — and the team's attention follows the budget.
Step 3 — Map
Draw the feedback loops that constitute the archetype
Sketch the causal loop diagram specific to your situation, using the archetype as a template. Label each variable. Mark reinforcing loops (R) and balancing loops (B). Identify delays — where does cause precede effect by weeks or months? The act of drawing forces precision. "Marketing isn't working" becomes "increased ad spend → higher volume of unqualified trials → lower conversion rate → higher CAC → pressure to spend more on ads." Once the loop is on paper, the team can see the trap.
Worked example
CloudSync's feedback structure
Reinforcing loop (R1): More marketing spend → more trial signups → more paid conversions → more revenue → more marketing budget. This loop powered the first 18 months. Balancing loop (B1 — the limit): More trial signups → onboarding team overwhelmed → slower response times → lower trial-to-paid conversion → growth stalls. Shifting the Burden loop: Growth stalls → increase ad spend (symptomatic fix) → brief uptick in trials → conversion drops further → growth stalls again. Meanwhile, the fundamental fix (invest in onboarding, hire customer success, improve self-serve activation) receives less attention and less budget with each cycle.
Step 4 — Locate
Identify the high-leverage intervention point
Each archetype has a known leverage point — the place where a structural change breaks the dysfunctional pattern. For Limits to Growth, the leverage is in the balancing loop: remove or weaken the constraint before it binds. For Shifting the Burden, the leverage is in strengthening the fundamental solution while deliberately weakening reliance on the symptomatic fix. The archetype tells you not just where to intervene, but where not to — which is often more valuable, because the obvious intervention usually reinforces the problem.
Worked example
Where CloudSync should intervene
The obvious move — spend more on marketing — is precisely wrong. It feeds the symptomatic loop. The high-leverage interventions: (1) Cap marketing spend at current levels to stop the bleeding. (2) Redirect $15K/month to a dedicated onboarding redesign — in-app guidance, automated activation sequences, a self-serve knowledge base. (3) Hire two customer success managers focused exclusively on trial-to-paid conversion. (4) Set a leading indicator: trial-to-paid conversion rate must return to 18% before marketing spend increases again. The archetype makes the logic legible: invest in the constraint, not the accelerator.
Step 5 — Monitor
Track whether the structural intervention is shifting the pattern
Structural interventions take time — often quarters, not weeks. Define leading indicators that will show whether the underlying dynamic is changing before the lagging metrics (revenue, growth rate) respond. Watch for the temptation to revert to the symptomatic fix when results don't appear immediately. That reversion is itself part of the archetype's pull. Name it explicitly with the team: "If we go back to increasing ad spend before conversion recovers, we're re-entering the Shifting the Burden loop."
Worked example
CloudSync's monitoring framework
Leading indicators: trial-to-paid conversion rate (target: 18% within 90 days), time-to-first-value for new trials (target: reduce from 4.2 days to 1.5 days), onboarding completion rate (target: 70%). Lagging indicator: MRR growth resumes within two quarters. The team agrees on a rule: no increase in paid acquisition budget until trial-to-paid conversion exceeds 16% for two consecutive months. The CEO writes "Shifting the Burden" on the whiteboard in the weekly growth meeting. It stays there for five months.
Section 3
When It Works Best
✓
Ideal Conditions for System Archetypes
| Dimension | Best fit |
|---|---|
| Problem type | Chronic, recurring problems that resist repeated fixes. If the same issue keeps returning — or if well-intentioned interventions consistently produce unintended consequences — you're almost certainly dealing with a structural pattern, not a one-time failure. The archetypes are built for exactly this. |
| Organisational maturity | Teams that have already tried the obvious solutions and are willing to question their own assumptions. System archetypes require a degree of intellectual humility — the willingness to consider that the team's instinctive response might be part of the problem. First-time founders in firefighting mode rarely have the patience. Operators who've been burned by the same pattern twice are the ideal audience. |
| Time horizon | Situations where the relevant dynamics play out over months or years, not days. Archetypes involve delays — the side effect of today's fix shows up next quarter. If you need a decision by Friday, this isn't the tool. If you're trying to understand why the last three years of strategy haven't produced the expected results, it's exactly the tool. |
| Stakeholder alignment | Cross-functional teams where different groups are pulling in different directions without understanding why. The archetype diagram makes the structural conflict visible. Marketing sees their loop. Product sees theirs. The diagram shows how the loops interact — and why optimising one function at the expense of another produces system-level dysfunction. |
| Complexity level | Problems with 3–8 interacting variables and feedback loops. Simpler problems don't need the overhead. Problems with dozens of interacting variables may require full system dynamics modelling — the archetypes are a shortcut that works when the structure is recognisable, not when it's genuinely novel. |
| Decision stakes | High-stakes strategic decisions where getting the intervention point wrong means wasting quarters of effort. The archetypes' primary value is preventing the wrong intervention — the one that feels right but reinforces the dysfunctional structure. That prevention is worth the diagnostic investment only when the cost of the wrong intervention is significant. |
Section 4
When It Breaks Down
⚠
Failure Modes
| Failure pattern | What goes wrong | What to use instead |
|---|---|---|
| Forced pattern matching | Teams learn the archetypes and start seeing them everywhere. Every problem becomes "Shifting the Burden" or "Limits to Growth" regardless of whether the structural fit is genuine. The archetype becomes a Procrustean bed — the situation is stretched or trimmed to match the template. This produces confident-sounding diagnoses that are structurally wrong. | Start with Causal Loop Diagrams built from scratch; only match to an archetype if the loop structure genuinely corresponds |
| Abstraction without action | The team produces an elegant systems diagram, nods sagely about feedback loops and delays, and then... nothing changes. The archetype becomes an intellectual exercise — a way to sound sophisticated about why things are broken without actually committing to a structural intervention. Systems thinking without systems doing. | Pair with Impact-Effort Matrix to force concrete intervention design immediately after diagnosis |
| Missing the delay structure | The archetype is correctly identified but the team underestimates the delays involved. They implement the structural fix, see no results in six weeks, panic, and revert to the symptomatic solution. The archetype predicted this would happen — delays are part of the structure — but knowing about delays intellectually and tolerating them emotionally are different things. | Define leading indicators that will show structural change before lagging metrics respond; use Pre-Mortem to anticipate the reversion temptation |
| Single-archetype tunnel vision | Real situations often involve multiple overlapping archetypes. A company experiencing Limits to Growth may simultaneously be caught in Eroding Goals (lowering its standards each quarter to make the numbers look acceptable). Diagnosing only one archetype produces a partial intervention that leaves the other dynamic untouched. | Map the full causal loop structure first, then check for multiple archetype signatures within the same system |
| Insufficient system boundary | The team draws the archetype but excludes critical variables that sit outside their department or organisation. A SaaS company maps its growth dynamics but ignores the competitive response loop — a competitor dropping prices in response to their market share gains. The archetype looks complete but the system boundary is too narrow. | Connection Circles to identify all relevant variables before drawing loop structures; explicitly ask "what's outside our diagram that could affect these loops?" |
| Narrative seduction | The archetypes are stories — compelling, memorable, emotionally resonant. That's their strength and their danger. A well-told "Tragedy of the Commons" narrative can convince a room without any quantitative evidence that the dynamic actually exists in their specific context. The story substitutes for analysis. | Require quantitative validation: show the data that confirms the feedback loop exists and operates at the magnitude the archetype implies |
The most dangerous failure mode is abstraction without action — and it's endemic to systems thinking in general. The archetypes are seductive precisely because they're intellectually satisfying. Naming the pattern feels like solving the problem. It isn't. A team that can eloquently describe why they're caught in a "Growth and Underinvestment" trap but won't reallocate budget to the underinvested capacity has gained nothing except vocabulary. The protection is procedural: every archetype diagnosis must end with a specific structural intervention, an owner, a timeline, and a leading indicator. If the session produces a diagram but not a decision, it failed.
The second risk worth flagging: systems thinking can become a form of learned helplessness. "Everything is connected, the delays are long, the leverage points are counterintuitive" — all true, and all potentially paralysing. The archetypes should narrow the action space, not expand it into infinite complexity. If the diagnosis makes the team feel more confused than before, the tool has been misapplied.
Section 5
Visual Explanation
System Archetypes applied to CloudSync — two overlapping archetypes (Limits to Growth and Shifting the Burden) showing why increased marketing spend reinforces the plateau rather than breaking through it.
Section 6
Pairs With
System archetypes are a diagnostic lens, not a complete intervention toolkit. They tell you what pattern you're trapped in and where the leverage point sits. The tools around them handle the upstream framing and the downstream execution.
Use before
Iceberg Model
The Iceberg Model moves analysis from events to patterns to structures to mental models. System archetypes live at the "structure" level. Use the Iceberg first to confirm you're dealing with a structural problem — not a one-time event or a trend that will self-correct — before investing in archetype diagnosis.
Use before
Connection Circles
Before matching to an archetype, map all the relevant variables and their causal connections using Connection Circles. This bottom-up approach prevents forced pattern matching — you build the structure from observed relationships, then check whether it resembles a known archetype, rather than imposing an archetype template on ambiguous data.
Use after
Causal Loop Diagrams
Once you've identified the archetype, draw the full causal loop diagram specific to your situation. The archetype is the template; the CLD is the detailed, context-specific map. This is where you label every variable, mark every delay, and identify exactly which link in the loop your intervention will target.
Use after
Pre-Mortem
Structural interventions take time and face organisational resistance. Run a Pre-Mortem on the intervention itself: "It's six months from now and we've reverted to the symptomatic fix. What happened?" This surfaces the political, budgetary, and psychological forces that will pull the team back into the dysfunctional loop.
Use after
[Second-Order Thinking](/mental-models/second-order-thinking)
After identifying the leverage point, use Second-Order Thinking to trace the consequences of your proposed intervention two and three steps out. Structural interventions can trigger new feedback loops. The fix for Limits to Growth (investing in capacity) can create a new Tragedy of the Commons if multiple teams compete for the same expanded resource.
Mental model
Reinforcing [Feedback](/mental-models/feedback) Loop
Every system archetype contains at least one reinforcing loop and one balancing loop. Understanding these building blocks — how reinforcing loops amplify and balancing loops stabilise — is prerequisite knowledge. Without it, the archetype diagrams are hieroglyphics.
Section 7
Real-World Application
People Express Airlines — Growth and Underinvestment in real time
The scenario
People Express Airlines launched in 1981 with a radical model: ultra-low fares, no frills, minimal staff per aircraft, and a flat organisational structure where everyone from pilots to managers handled baggage. It worked spectacularly. By 1984, People Express was the fifth-largest US airline by passenger count, growing revenue from $38 million to over $1 billion in three years. Then it collapsed. By 1986, the airline was acquired by Texas Air Corporation at a fraction of its peak valuation. The speed of the rise and fall made it a case study at MIT's System Dynamics Group — and the diagnosis was a textbook instance of the Growth and Underinvestment archetype.
How the archetype appliedThe reinforcing loop was clear: low fares attracted passengers, high load factors generated revenue, revenue funded route expansion, more routes attracted more passengers. The balancing loop — the constraint — was service quality. As passenger volume grew, the airline's lean staffing model couldn't maintain service standards. Flights were delayed. Baggage was lost. The booking system couldn't handle the volume. Customer complaints rose. The fundamental solution was investment in infrastructure: better reservation systems, more ground staff, training programmes, maintenance capacity. But each of these investments would have reduced the short-term cost advantage that made the model work. So the leadership chose the symptomatic fix: add more routes and cut fares further to maintain volume growth, compensating for declining repeat-customer rates with new customers from new markets.
What it surfacedThe archetype makes visible what was invisible in real time: the delay between underinvestment and its consequences. Service quality didn't collapse overnight. It eroded over 18 months — slowly enough that each quarter's numbers could be explained away. "We're growing too fast" was treated as a good problem to have. By the time the service degradation was severe enough to show up in the revenue numbers, the investment gap was too large to close. Competitors — particularly American Airlines with its Sabre reservation system and yield management capabilities — had invested in exactly the infrastructure People Express had deferred, and they used it to selectively match People Express fares on high-volume routes while maintaining service quality.
The non-obvious factorJohn Sterman at MIT later built a system dynamics simulation of People Express that demonstrated something the management team couldn't see from inside the system: the airline's optimal strategy was to slow down growth in 1983 — to deliberately constrain route expansion and invest the margin in service infrastructure. The simulation showed that this slower-growth path would have produced a larger, more profitable airline by 1986 than the actual high-growth path did. The archetype's deepest lesson isn't that People Express made a mistake. It's that the mistake was structurally invisible from inside the reinforcing loop. Growth felt like success. The constraint was a lagging indicator. By the time it became visible, the window for the fundamental investment had closed. This is why the archetype exists as a diagnostic tool — to make the invisible structure visible before the delay runs out.
Section 8
Analyst's Take
Faster Than Normal — Editorial View
System archetypes occupy a strange position in the strategy toolkit. They're simultaneously one of the most powerful diagnostic frameworks available and one of the most frequently misused. The power is real: once you internalise even four or five archetypes, you develop a structural intuition that changes how you read business situations. You stop being surprised by the same patterns. You start seeing the balancing loop before it binds, the symptomatic fix before it atrophies the fundamental solution, the commons before it's depleted. That pattern recognition, applied early enough, is worth more than most quantitative analysis — because it tells you which interventions will backfire, and that knowledge prevents the most expensive category of strategic error.
The misuse I see most often is what I'd call archetype tourism. A leadership team reads The Fifth Discipline, gets excited, runs a workshop, produces beautiful loop diagrams on whiteboards, and then returns to making decisions exactly as before. The archetypes become a shared vocabulary for describing dysfunction rather than a tool for changing it. "We're in a Shifting the Burden pattern" becomes a thing people say in meetings the way they say "we need to be more strategic" — a sophisticated-sounding observation that creates no obligation to act differently. The tell: if the archetype diagnosis doesn't produce a specific, uncomfortable reallocation of resources or attention, it hasn't been used. It's been performed. Structural interventions are almost always politically difficult because they require redirecting resources from the thing that's producing visible short-term results (the symptomatic fix) toward something whose payoff is delayed and uncertain (the fundamental solution). If the archetype diagnosis doesn't make someone in the room uncomfortable, you haven't reached the leverage point.
The highest-leverage modification: quantify the delay. Most archetype discussions stay qualitative — "there's a delay between our investment in onboarding and its effect on conversion." That's useful but insufficient. Force the team to estimate the delay in weeks or months. "If we invest in onboarding redesign today, we expect to see conversion improvements in 8–12 weeks, and MRR impact in 16–20 weeks." That number — 16 to 20 weeks — is the team's commitment horizon. It's how long they need to resist the pull of the symptomatic fix. Without a number, "the delay is long" becomes an excuse to revert at the first sign of impatience. With a number, the team has a concrete agreement: we will not evaluate this intervention until week 16. That single quantification transforms the archetype from an intellectual framework into an operational commitment.
Section 9
Top Resources
01
The Fifth Discipline — Peter Senge (1990)
Primary sourceThe origin text. Senge introduces the eleven system archetypes within a broader argument about organisational learning, and the archetypes chapter remains the most accessible entry point to structural pattern recognition. The examples are dated — GE, Royal Dutch Shell, the beer distribution game — but the structural logic is timeless. Read chapters 5 through 8 for the archetypes specifically; read the whole book to understand the learning organisation context that makes them actionable rather than merely intellectual.
02
Thinking in Systems: A Primer — Donella Meadows (2008)
BookMeadows was Senge's intellectual predecessor and, arguably, the clearer writer. Published posthumously, this book covers stocks, flows, feedback loops, delays, and system behaviour with a precision and elegance that The Fifth Discipline doesn't quite achieve. It doesn't use the "archetype" framing explicitly, but every archetype is a specific configuration of the building blocks Meadows explains here. Read this first if you want to understand why the archetypes work, not just what they are.
03
The Fifth Discipline Fieldbook — Senge, Kleiner, Roberts, Ross & Smith (1994)
BookWhere The Fifth Discipline is theory, the Fieldbook is practice. It contains facilitation guides, workshop designs, and worked examples for each archetype — the operational manual that the original book lacks. The "archetype template" pages, which show each pattern's generic structure alongside diagnostic questions, are the single most useful reference for teams applying the tool in real meetings. Worth owning in physical form for the diagrams alone.
04
Book
Grove never uses the phrase "system archetype," but his account of Intel's strategic inflection points is the best real-world illustration of multiple archetypes operating simultaneously. The shift from memory chips to microprocessors involved Limits to Growth (the DRAM business hitting commoditisation), Shifting the Burden (incremental process improvements substituting for the fundamental strategic pivot), and Eroding Goals (gradually accepting lower margins as "the new normal"). Read it as an archetype case study, even though Grove didn't frame it that way.
05
Business Dynamics: [Systems Thinking](/mental-models/systems-thinking) and Modeling for a Complex World — John Sterman (2000)
BookThe definitive academic treatment. Sterman — who built the People Express simulation referenced in Section 7 — covers every archetype with mathematical rigour, simulation models, and extensive case studies. At 982 pages, it's not casual reading. But if you want to move beyond qualitative archetype matching to quantitative system dynamics modelling — to actually simulate the delays and test intervention strategies before committing resources — this is the only book you need. Chapter 5 on causal loop diagramming and Chapters 10–11 on generic structures are the most directly relevant sections.
Decision Tools Library — Browse by phase
Why this matters next
mental modelsIncentives
Iceberg Model applied the Incentives mental model
mental modelsSecond-Order Thinking
Iceberg Model applied the Second-Order Thinking mental model
mental modelsLeverage
Iceberg Model applied the Leverage mental model
mental modelsTragedy of the Commons
Iceberg Model applied the Tragedy of the Commons mental model
mental modelsSystems Thinking
Iceberg Model applied the Systems Thinking mental model
mental modelsNarrative
Iceberg Model applied the Narrative mental model
Continue exploring
CCCDRLBLSDCM
Decision tool
Connection Circles
Map the elements of a system and the relationships between them — the simplest s
Decision tool
Causal Loop Diagrams
Formally map cause-and-effect relationships with polarity to show how variables
Decision tool
Reinforcing Feedback Loop
Understand the mechanism behind exponential growth and vicious/virtuous cycles
Decision tool
Balancing Feedback Loop
Understand the mechanism that pushes back against change to create stability or
Decision tool
Stock and Flow Diagrams
Model how things accumulate (stocks) and change over time (flows) — the bathtub
Decision tool
Concept Map
Visualise relationships between entities in a concept or domain to build shared
More like this, in your inbox
I send a newsletter every week — free, no spam, unsubscribe anytime.