The minimum effective dose (MED) is the smallest input that produces the desired outcome. Borrowed from medicine — the lowest dose of a drug that achieves the therapeutic effect — the model applies wherever you want to maximise result per unit of effort, time, or cost. Below the MED, you get nothing or too little. Above it, you often get diminishing returns or wasted resource. The discipline is finding the threshold and stopping there.
In building and scaling, MED thinking forces you to ask: what is the smallest viable version of this feature, campaign, or process that still works? Startups that ship the minimal product that validates demand are applying MED. Marketing teams that find the lowest ad spend that moves the needle are applying it. The mistake is doing more "to be safe" or "to do it right" when the extra work does not change the outcome. MED is the frontier of sufficiency.
In deciding and judging, MED helps you allocate effort to decisions. Not every choice deserves a deep analysis. The minimum effective dose of thinking for a low-stakes, reversible decision might be a quick heuristic. For a high-stakes, irreversible decision, the MED of analysis is higher — but it is still a dose. Beyond some point, more analysis does not improve the decision; it only delays it or creates false confidence. MED separates the amount of effort that actually changes the outcome from the amount that feels thorough but does not.
The model does not say "do the least possible." It says "do the least that is sufficient." Sufficiency is defined by the goal. If the goal is to learn whether customers will pay, the MED might be a landing page and five conversations. If the goal is to win a regulatory approval, the MED might be a full dossier. The same principle — find and hit the threshold — applies at every scale.
Section 2
How to See It
MED reveals itself when small changes produce the target result and additional effort does not. Look for the point where the curve flattens: one more unit of input yields little or no extra output. That flattening is the signal you are past the MED. When you see teams or individuals consistently doing more than is needed to hit the goal, MED is being ignored.
Business
You're seeing Minimum Effective Dose when a team ships a stripped-down MVP and gets the same product–market fit signal as a team that spent six months building a full product. The minimal version crossed the threshold for learning. The extra features were beyond the MED for that decision — they did not change the outcome, only the cost and time.
Technology
You're seeing Minimum Effective Dose when a single well-placed integration or API change fixes a performance issue that others tried to solve with large rewrites. The MED of the fix was small; the rest was over-engineering. The diagnostic is that the outcome (performance target met) was achieved with minimal intervention.
Investing
You're seeing Minimum Effective Dose when an investor reaches conviction after a focused set of reference calls and one key metric, while another spends months on due diligence and reaches the same conclusion. The MED of analysis was the smaller set; the extra work did not change the decision.
Markets
You're seeing Minimum Effective Dose when a brand finds that a narrow channel (e.g. one podcast, one community) drives most of its qualified leads. Doubling spend elsewhere does not double results — the MED for acquisition is concentrated. Scaling beyond MED in other channels wastes budget.
Section 3
How to Use It
Decision filter
"Before adding more scope, time, or resource, ask: have we already reached the minimum effective dose for this goal? If yes, stop. If no, what is the smallest next step that could get us there? Prefer that step over the 'complete' solution."
As a founder
Ship the minimum that validates or achieves the next milestone. For validation, the MED is often a prototype, a waitlist, or a manual process that proves demand or feasibility. For scaling, the MED is the smallest system that can handle the next order of magnitude without breaking. The trap is gold-plating: building more than the goal requires. Define the goal clearly, then find the smallest intervention that reaches it. Revisit the MED as goals change — what was sufficient at seed may be insufficient at scale.
As an investor
Apply MED to due diligence and to portfolio support. For each decision, identify the critical questions and the minimum evidence needed to answer them. Beyond that, more analysis often does not improve accuracy; it only adds cost and delay. For portfolio companies, the MED of your help is the smallest input that unblocks them — a single intro, one piece of advice, or a clear decision. More is not always better.
As a decision-maker
Match the dose of thinking to the stakes and reversibility of the decision. Low-stakes, reversible decisions get a low MED: quick rules, delegation, or a short checklist. High-stakes, irreversible decisions get a higher MED: more data, stress-testing, and explicit criteria. In both cases, define what "enough" looks like and stop when you reach it. Avoid the illusion that infinite analysis guarantees a better outcome.
Common misapplication: Treating MED as an excuse for sloppiness. MED is the minimum effective dose — effectiveness is required. If the small dose does not achieve the goal, it is not the MED; it is under-dosing. The model demands that you verify the outcome, not just reduce the input.
Second misapplication: Assuming the MED is static. As context changes — competition, regulation, customer expectations — the threshold can shift. What was sufficient last quarter may be insufficient now. Re-measure and update the MED when conditions change.
Hastings has emphasised simplicity and focus: do a few things well rather than many things poorly. Netflix's early strategy was the MED of content and product — enough to differentiate and retain subscribers without the bloat of a full studio or feature set. The company repeatedly stripped down to the minimum that served the next strategic goal.
Graham's advice to founders — "do things that don't scale," "make something people want," and ship fast — is MED applied to startup building. The minimum effective dose for learning is often a manual, ugly version that gets real usage and feedback. YC's batch structure is itself a compressed MED for accelerating startups: the smallest program that reliably improves outcomes.
Section 6
Visual Explanation
Minimum Effective Dose — Below the threshold, outcome is insufficient. At and past the MED, outcome is achieved; beyond that, diminishing returns. The goal is to operate at the MED, not below or far above.
Section 7
Connected Models
Minimum effective dose sits with models about sufficiency, efficiency, and prioritisation. The connections below either explain why MED works (Pareto, diminishing returns), extend it to product and strategy (MVP, essentialism), or create tension with optimisation (marginal gains, efficiency vs effectiveness).
Reinforces
80/20 Rule (Pareto)
Pareto says a small share of inputs often produces most of the outputs. MED is the discipline of finding that small share and using only that. The 20% that drives 80% of results is often close to the minimum effective dose — the smallest set of actions that gets you most of the way there.
Reinforces
Law of Diminishing Returns
Diminishing returns describe why there is an MED: beyond some point, each extra unit of input adds less and less. The MED is the point where you have "enough" — just past where the curve flattens. The law explains why more is not always better.
Tension
Marginal Gains
Marginal gains focus on small improvements that compound. MED focuses on the minimum to reach a threshold. The tension: after the MED, marginal gains say "improve a little more"; MED says "stop unless the goal changes." Use marginal gains when you are optimising past the threshold; use MED to find the threshold.
Tension
Efficiency vs Effectiveness
MED is about effectiveness first: hit the goal. Efficiency is about doing so with less waste. The tension: effectiveness can push you to do "enough to be sure," while MED says "the least that works." Pair them by defining effectiveness clearly, then finding the efficient (minimum) dose that achieves it.
Section 8
One Key Quote
"Perfection is achieved not when there is nothing more to add, but when there is nothing left to take away."
— Antoine de Saint-Exupéry
The quote inverts the default: instead of "what else should we add?", ask "what can we remove and still succeed?" That is the MED mindset. The minimum effective dose is the point where taking away more would break the outcome. Everything beyond that is optional.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Most teams over-dose. The default is to do more — more features, more analysis, more process — because it feels safer or more thorough. MED forces the question: did the extra work change the outcome? If not, it was waste. The discipline is defining the outcome, finding the minimum that achieves it, and stopping.
MED is not laziness. The minimum effective dose requires that the outcome is actually achieved. Under-dosing is failure. The model is about sufficiency, not minimalism for its own sake. Test that the dose works before you commit to it.
Revisit MED when the goal or context changes. What was sufficient for validation may be insufficient for scale. What was sufficient in a benign market may be insufficient when competition intensifies. Treat MED as a moving target and re-estimate when conditions shift.
Use MED to prioritise. When resources are limited, MED tells you the order of operations: hit the minimum effective dose for the highest-leverage goal first, then the next. Avoid spreading effort below the MED across many goals — you get no result anywhere. Concentrate to cross the threshold on one thing at a time.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A founder builds a simple landing page and runs one ad campaign. She gets 200 sign-ups and five paying customers. She stops and uses that signal to raise a seed round instead of building the full product first.
Scenario 2
A team spends three weeks on due diligence for a small acquisition. Another team spends six months on a similar deal and reaches the same go/no-go conclusion.
Scenario 3
A company ships a feature with minimal testing and it causes a major outage. The CEO says they were applying minimum effective dose.
Section 11
Summary & Further Reading
Summary: Minimum effective dose is the smallest input that produces the desired outcome. Use it in building and scaling to ship the minimum that validates or achieves the next milestone, and in deciding and judging to match the dose of analysis to the stakes of the decision. Find the threshold, hit it, and stop — unless the goal or context changes. Pair with 80/20, MVP, law of diminishing returns, essentialism, and efficiency vs effectiveness.
Ferriss popularised MED in fitness and nutrition: the minimum effective dose of exercise, sleep, and diet for specific outcomes. The framing — smallest dose for the result — transfers to productivity and business.
Ries's build–measure–learn loop and MVP concept are MED applied to startups: the minimum product that tests the hypothesis. The book grounds MED in product and company building.
McKeown argues for doing only what is essential. MED identifies what is essential for a given goal; Essentialism extends that to choosing goals and eliminating the non-essential.
Leads-to
MVP
MVP is MED applied to product: the minimum viable product is the smallest version that validates the hypothesis or serves the user. Both ask "what is the least we can do and still succeed?" MVP is MED for building and learning.
Leads-to
Essentialism
Essentialism is the practice of doing only what is essential. MED identifies what is essential for a given goal — the minimum that works. Essentialism extends that to how you choose goals and allocate life and work; MED is the dose, essentialism is the filter.
