Skip to main content
Holist-IQ Logo

Stock-and-Flow vs Causal Loops: When You Need Which

8 min read
stock-and-flow causal-loop-diagram systems-thinking
HT
Holist-IQ Team Author

Stock-and-flow vs causal loops

Teams modeling dynamic problems choose between stock-and-flow diagrams and causal loop diagrams (CLDs). Both can show feedback. They answer different questions. Stock and flow vs causal loops is a fit decision about accumulation, units, and audience.

Use stock-and-flow when you must say how much of something is on hand, in which unit, and which rates change it. Use a causal loop diagram when the decision is about signed influence, including factors you cannot yet put in a conserved unit, and when stakeholders need one story before anyone writes an equation.

Prefer stock-and-flow when

  • Accumulation and depletion need numeric rates
  • You will simulate levels over time
  • Conservation or unit consistency is the point of the argument

Prefer causal loops when

  • Qualitative feedback structure is the decision object
  • Stakeholders need a shared map before equations
  • Important variables are beliefs, rules, or perceived risk

What stock-and-flow emphasizes

A stock is a quantity that accumulates. A flow is a rate that fills or drains it. The stock is the integral of the net flow. Water in a bathtub, cash in an account, people in a cohort, tonnes of material in a process: the level only changes when a flow runs.

Flows carry units per time. Stocks carry the unit itself. If you cannot name the unit, you do not have a stock yet. You have a topic.

The question the diagram is for: how much is on hand, and what rates change it? That matters for inventory, treasury, population cohorts, carbon and energy balances, and any policy where “more” has to survive contact with conservation. A subsidy cannot create skilled installers in the same month if the stock of trained people only grows through a slow flow.

Stock-and-flow diagrams also make delays physical. A pipeline delay is stuff in transit, which is itself a stock, not a wavy arrow.

What causal loops emphasize

A CLD shows signed influence among factors. An arrow means “if this rises, that rises or falls,” not “this many units per week move from A to B.” Positive and negative marks, loop labels (reinforcing or balancing), and delay marks are the notation that matters.

CLDs can include variables that are not physical stocks: trust in a forecast, perceived enforcement risk, political patience, willingness to report. Those factors steer behavior, and they do not conserve like cash or carbon.

The question the diagram is for: what influences what, including the closed loops, on which clocks? That matters in alignment workshops and in policy narratives before anyone is ready to quantify. A room can agree the structure of a backlash loop without agreeing the coefficient.

CLDs do not, by themselves, tell you the level of a stock next quarter. Adding numbers onto arrows does not turn them into flows. It hides the missing units.

What each view cannot do

A stock-and-flow model without belief and incentive variables will simulate a clean plant and miss the loop that actually moves the decision. A CLD that treats “inventory” as just another phrase, with no inflow and outflow, cannot tell you whether a target is feasible. It can tell you that pressure, expediting, fatigue, and defects close on each other.

If your argument depends on a level crossing a threshold, you will eventually need stocks. If it depends on whose story of influence is accepted, a simulation with hidden structure will not settle the room.

Honest tool alternatives

NeedOften fitsTrade-off
Quantified accumulationStella, Vensim, AnyLogicLearning curve, data demand, someone must own the units
Executive alignment CLDHolist-IQ, kumu, dedicated CLD toolsOften starts qualitative; not a substitute for conservation math
Quick numbersSpreadsheetsFeedback structure is easy to hide in tabs
Whiteboard speedMiro, Lucid, pen and paperLoop semantics drift unless you enforce them

Holist-IQ focuses on decision-ready causal loop diagrams from plain language. Stella-class tools are the right class when equations, units, and time integration are non-negotiable. This page does not rank prices or licenses. Check the vendor when you buy. Spreadsheets win one-off budgets and cohort tables. They obscure loops unless someone designs the circularity on purpose and labels it.

Hybrid workflow

Many teams use both, in that order. Sketch the CLD while people still disagree about signs and delays. Name the loops. Then promote one slice into stock-and-flow where the variables are physical or financial and the data can support rates.

Keep the names aligned. If the CLD says “perceived stockout risk” and the simulation says “inventory units,” you have two models that cannot correct each other. Write a short glossary: which CLD variables are stocks, which are information that only influences a flow, and which you will leave qualitative.

Do not quantify every arrow. Quantify the slice that would change the decision if the level were wrong. Leave the rest as a loop story so the simulation does not pretend to know trust.

A teaching sketch: goods on hand, and the story around them

This is a structure sketch, not a client result.

Stock-and-flow slice. Finished goods are a stock. Production is an inflow (units per week). Shipments are an outflow. A quality hold can be a second stock, fed by a defect flow, drained by a rework flow. You can ask whether a production increase raises the stock or only fills the hold. Units have to balance.

Causal loop slice on the same situation. Perceived stockout risk rises, expediting rises, overtime rises, fatigue rises, defect rate rises, shipments of good units fall, perceived stockout risk rises further. That is a reinforcing loop. A balancing loop may sit on hiring or on a tighter release rule, both with delays. None of those perception links is a flow of units. They change the rates in the stock-and-flow slice.

Use the CLD to decide which behavioral link you believe. Use the stock-and-flow to see whether the warehouse level can carry the promise on the slide. If you only build one, say which question you are refusing to answer.

When not to force stock-and-flow

Skip a full stock-and-flow model when the data for rates is sparse, the audience needs a shared story this month, or the dominating loops are beliefs and incentives with no reliable measured flow. Building a simulation to look rigorous will invent coefficients. Document the CLD, name the missing rates, and quantify later if a decision actually depends on a level.

Also skip a CLD-only process when a regulator, a finance partner, or a physical balance will reject anything that does not conserve. In that room the stock-and-flow is the decision object. The loop story can be an appendix.

Who should pass on each

Pass on stock-and-flow software if nobody on the team will maintain units after the consultant leaves. Pass on a CLD practice if the group wants a mural of topics and will not mark polarity, loop names, and delays. In that case a mind map is the honest artifact.

Label the model you already have

Look at last week’s deck. Was it a CLD, a stock-and-flow, or a spreadsheet table with boxes drawn around it? Write the intended question in the footer: “how much on hand” or “what influences what.” If the footer and the picture disagree, the audience will answer the wrong question and feel aligned.

Frequently asked questions

Can a causal loop diagram become a stock-and-flow model?

Yes, in part. Promote variables that accumulate into stocks, define the flows between them, and assign units. Leave belief and rule variables as influences on those flows unless you have a real measurement for them.

Which is better for climate and transition work?

Often both. Causal loops carry lock-in stories that include politics and expectations. Stock-and-flow carries carbon, energy, and capital stocks where conservation is the claim. A single view will drop one of those questions.

Are causal loop diagrams less rigorous?

They are rigorous about structure and weak about levels. A stock-and-flow model can be numerically precise and still omit the loop that changes the decision. Rigor is matching the artifact to the question.

What to do next

Read causal loop diagram software and the systems thinking tool checklist before you buy simulation capacity nobody will staff. For policy and rate stories that mix both views, see economic systems modeling.

Ready to see the whole picture? Start Mapping
HT

Written by

Holist-IQ Team

Helping teams see the whole picture through systems thinking and feedback loop mapping.

Customize your cookie preferences. Essential cookies cannot be disabled as they are required for the website to function properly.

Essential Cookies

Required for basic website functionality, security, user authentication, and error tracking.

Always active

Analytics Cookies

Help us understand how visitors interact with our website to improve user experience. Includes Google Analytics and Microsoft Clarity session recordings.

Marketing Cookies

Used to track visitors across websites to display relevant and engaging advertisements.