Law of Diminishing Returns

TL;DR

Law of Diminishing Returns: Each additional unit of input produces less additional output than the previous unit, after some point. The tenth engineer adds less than the first; the tenth ad impression matters less than the first. Knowing where diminishing returns begin is essential for resource allocation.

What Is the Law of Diminishing Returns?

The Law of Diminishing Marginal Returns is one of the oldest principles in economics, formalized in the early 19th century. In its canonical form: if you hold all other inputs constant and add successive units of one input, eventually each additional unit produces a smaller marginal increase in output than the unit before.

The classic example: a plot of land with one farmer produces X food. A second farmer, working the same land, produces more food — but not 2X, because they must work less productive land. A third produces even more, but the marginal gain is smaller. Eventually, additional farmers produce near-zero additional food — the land is the binding constraint.

In modern contexts:

Marketing: The first advertisement in a campaign reaches people who haven't heard your message. The tenth reaches people who have heard it nine times and are tuning it out. Marginal reach and recall fall with successive exposures.

Engineering teams: Adding engineers to a team below a certain size dramatically increases throughput. Adding engineers to a team above a certain size increases coordination overhead at roughly the same rate as it increases output — producing near-zero or negative marginal returns (Brooks's Law).

Feature development: Adding the first five features to a product typically produces large increases in utility. Adding features 50–60 typically produces complexity and cognitive load that may reduce utility for some users.

Exercise: Adding the first 30 minutes of exercise per week produces dramatic health benefits. Adding the 500th minute per week produces small additional benefits and potentially negative returns (overtraining).

How It Works

Total Output vs. Marginal Output:

Units of Input | Total Output | Marginal Output
      1        |     10       |      10
      2        |     18       |       8       ← still positive, falling
      3        |     24       |       6
      4        |     28       |       4
      5        |     30       |       2
      6        |     30       |       0       ← optimum
      7        |     28       |      -2       ← negative returns

Practical implications:
  • Stop adding input when marginal output equals the cost of the input
  • If marginal output > cost: add more
  • If marginal output < cost: reduce or redirect
  
  The shape of the curve matters:
  • Does diminishing return kick in early? → Diversify
  • Does it kick in late? → Concentrate

Three Real-World Examples

Engineering Headcount at Startups

A 3-person engineering team that grows to 10 typically sees productivity increase significantly. The team can work on more parallel tracks, cover more specializations, and maintain more of the codebase. Growth from 10 to 20 often shows smaller productivity gains per hire as coordination overhead increases. Growth from 20 to 50 may show near-zero gains per engineer added without restructuring (Brooks's Law).

This is why the best-run engineering organizations don't maximize headcount — they identify where their marginal return on engineering resources is highest and concentrate there. For most software companies, this is well below what an unlimited hiring budget would produce.

Marketing Channel Saturation

A brand launches on YouTube with excellent content. The first $100,000 in YouTube promotion reaches a large, engaged audience. The next $100,000 reaches a less targeted audience. The tenth $100,000 reaches an increasingly saturated audience with declining engagement. Eventually, the marginal return on YouTube spend falls below the return on alternative channels.

The optimal marketing strategy: identify diminishing returns on each channel, then reallocate to the marginal next-best channel. This explains why sophisticated marketers constantly test new channels rather than maximizing spend on proven ones — the proven ones are often past their diminishing-returns knee.

Feature Development and Product Complexity

Products begin with core value-producing features. Each early feature meaningfully increases the product's value to the target user. As the feature count grows, new features serve progressively more niche use cases, and the cognitive complexity they add may reduce value for the core user. Beyond a certain feature count, additional features reliably reduce satisfaction for the median user.

This is why the best product teams aggressively prioritize and often remove features — managing the diminishing returns curve of a product's complexity is as important as adding new capabilities.

When to Use It

✅ Use Diminishing Returns analysis when:

Allocating resources across channels, teams, or initiatives
Deciding when to diversify vs. concentrate investment
Evaluating marginal hiring decisions
Designing workout programs, study schedules, or any activity with an optimal volume

Pairs well with	Why
Theory of Constraints	The constraint is where diminishing returns from other inputs goes to zero first
Pareto Principle	80/20 rule reflects diminishing returns: the first 20% of inputs produce 80% of output
Brooks's Law	Brooks's Law is diminishing returns applied specifically to software team size

Common Misuses and Limitations

Confusing with negative returns. Diminishing returns mean each additional unit adds less than the previous one — not that it adds nothing or makes things worse. The tenth hour of studying still improves performance; it just improves it less than the second hour. Negative returns (the eleventh hour causes fatigue that hurts performance) are a further stage that may or may not follow.

Assuming the same curve applies across contexts. The shape and location of the diminishing returns curve varies enormously by task, domain, and individual. Highly complex, open-ended problems (research, creative work) often have much later diminishing returns than simple, well-defined tasks. Assuming a single universal curve for "effort" is usually wrong.

Ignoring returns to different inputs. Diminishing returns typically apply to a single input held constant while others are fixed. Adding more labour to a fixed amount of capital shows diminishing returns. But if you increase both labour and capital together, you may get constant or even increasing returns (returns to scale). The interaction between inputs matters.

Using it to justify premature stopping. "Diminishing returns" is sometimes invoked to justify stopping work before quality is achieved — "we're well into diminishing returns territory." This may be correct, but it ignores that some outcomes have quality thresholds: a bridge that's 90% as structurally sound as optimal is dangerous, not "good enough."

Model	Relationship
Pareto Principle	The 80/20 rule describes where diminishing returns set in — the first 20% of effort often yields 80% of results
Theory of Constraints	Identifying constraints allows you to direct effort where diminishing returns haven't yet set in
Satisficing	Satisficing is the rational response to diminishing returns — stop when good enough rather than when optimal
Carrying Capacity	Carrying capacity is where diminishing returns become zero returns

Frequently Asked Questions

Where do diminishing returns typically set in for common tasks?

Research on productivity suggests: for focused cognitive work, quality typically peaks at 4–5 hours per day of deep work (Newport) — additional hours show strong diminishing returns due to mental fatigue. For training, most fitness gains occur in the first 80% of sessions, with the final 20% yielding marginal improvement while significantly increasing injury risk. For marketing, diminishing returns on advertising spend typically set in well before market saturation — 5–7 exposures is roughly optimal for brand messaging, with additional exposures adding negligible recall.

How does diminishing returns relate to diversification?

In portfolio theory, diversification follows a diminishing returns curve. The risk-reduction benefit of adding the 2nd stock to a portfolio is massive; adding the 20th is minimal; adding the 50th is nearly zero. Studies suggest ~20–30 holdings achieve ~90% of the maximum diversification benefit available in an asset class. Beyond that, incremental diversification adds administrative complexity with negligible additional risk reduction. This is why index funds achieving "market-level diversification" don't need to hold all 500 S&P companies proportionally for most purposes.

What's the relationship between diminishing returns and marginal analysis?

Marginal analysis is the economic framework built on diminishing returns. The optimal point for any variable input is where its marginal product equals its marginal cost — where the last unit added produces exactly as much value as it costs. Below this point, adding more is profitable; above it, each additional unit costs more than it produces. All of microeconomic price and production theory is built on this foundation. In practice, finding the marginal product curve requires experimentation — which is why A/B testing, incremental budget allocation, and pilot programmes are valuable.

Apply with AI

🚀 Find your diminishing returns curve with MindMax →

This page is part of the MindMax Mental Models Knowledge Base.

What Is the Law of Diminishing Returns?​

How It Works​

Three Real-World Examples​

Engineering Headcount at Startups​

Marketing Channel Saturation​

Feature Development and Product Complexity​

When to Use It​

Common Misuses and Limitations​

Related Models​

Frequently Asked Questions​

Further Reading​

Apply with AI​