Q.Piyush is the Production Manager of a shoe company called 'Comforty'. He has 1000 workers working under him in his department. Piyush booked a special order for 300 pairs of gents shoes at an estimated cost price of ₹ 500 per pair plus 20% profit margin. The order is to be completed within 10 days. After completion of the order, Piyush found out that the production cost of one pair of shoes was ₹ 600. He tried to find out the reason, but he couldn't. He reported it to the Chief Executive Officer. The Chief Executive Officer appointed Abhay to find out the cause. Abhay was not only a specialized Cost Accountant, but also a qualified Human Resource Manager. After two days, Abhay reported that the set targets could be completed with the help of only 800 workers, which revealed that there was overmanning in the organization. The organization was not utilizing the human resources optimally, which had led to higher labour cost. The above case highlights one of the points of importance of an important function of management.

Controlling Definition — First Encounter

Imagine you're trying to prove that a function is continuous at a point. You've seen the epsilon-delta definition: for every $\epsilon > 0$, there exists a $\delta > 0$ such that if $|x - c| < \delta$, then $|f(x) - f(c)| < \epsilon$. That's the standard definition.

Now think about what you're actually doing when you use this definition to prove continuity. You're given an arbitrary $\epsilon$ — a tolerance on the output — and you must produce a $\delta$ — a window on the input — that controls the output error. The $\delta$ you choose depends on $\epsilon$, on the function $f$, and on the point $c$. The core idea: you control the input to control the output.

That's the intuition behind the term "controlling definition." It's a definition that tells you: to achieve a desired precision in the output, you must restrict the input within some range. The definition itself gives you the control — the $\delta$ — as a function of the desired precision $\epsilon$.

The Precise Statement

A controlling definition is a definition of a limit (or continuity, or derivative) that explicitly quantifies the relationship between an output tolerance and an input restriction. For the limit $\lim_{x \to c} f(x) = L$, the controlling definition is:

The word "controlling" highlights that $\delta$ is chosen after $\epsilon$ is given — you control the input window $\delta$ to force the output error below $\epsilon$. The $\delta$ is not arbitrary; it must be found (or shown to exist) for every possible $\epsilon$.

Why "Controlling" Matters

This phrasing is most common in analysis textbooks when contrasting with other equivalent definitions. For example, the "sequential definition" of a limit says: if every sequence $x_n \to c$ (with $x_n \neq c$) has $f(x_n) \to L$, then the limit exists. That definition doesn't give you a direct control mechanism — it's about checking sequences.

The controlling definition is the one you actually use in proofs. It's constructive: you pick $\delta$ as a function of $\epsilon$, often by solving inequalities. For instance, to prove $\lim_{x \to 2} (3x+1) = 7$, you set $\delta = \epsilon/3$. That's the control: if $|x-2| < \epsilon/3$, then $|(3x+1)-7| = 3|x-2| < \epsilon$. …