Twelve people saving twenty minutes a day is not a headcount

Here is a slide I have seen in some version at least fifty times. The tool saves twelve people twenty minutes a day. Twelve people, twenty minutes, two hundred and fifty working days, at a loaded rate. It comes out to something north of a hundred thousand dollars, it goes into the business case as savings, and everyone nods.

It is almost never a hundred thousand dollars. It is almost never any dollars at all.

The question that settles it is short: what will those twelve people do with the time, and who is accepting that in their plan? If there is a real answer, and I mean a named use with a named owner who has put it in their own objectives, then it is money. If the honest answer is that they will be less swamped, then it is capacity, and it goes in the document as capacity with no dollar sign attached.

That distinction is not pessimism. It is the difference between a case that survives contact with a CFO and one that gets quietly discounted to zero the moment you leave the room. Finance has seen the twenty-minutes slide before. They know that saved minutes distributed across a team do not show up anywhere, because nobody was ever going to fire a fifth of a person, and the work expands to fill the day. When you claim it as savings, you are telling a room full of people who already know better that you have not thought about it hard.

So we classify every benefit as exactly one of five types, and we say which one out loud. Cash out is a bill we stop paying, proven by an invoice and a renewal date. Cash avoided is a bill we never start paying, proven by an approved plan. Revenue or margin protected is a timing or win-rate change, or value we already earned and stopped leaking, with the attribution argument stated. Capacity released is hours returned. Risk reduced is expected loss avoided, stated as probability times impact, with the specific control that changes the probability.

Naming the type does two things. It forces the proof standard into the open, because each type is proven differently. And it stops the most common error in business cases, which is counting the same saving twice. Once every benefit is mapped to the specific baseline line it reduces, you can see immediately when one line is being reduced by more than a hundred percent across three different rows. That happens constantly, and it is almost always accidental.

Back to capacity. I am not arguing it is worthless. It is often the most valuable thing a build produces, because hours returned to a senior person are worth far more than their hourly rate suggests. I am arguing that it only becomes money at the moment somebody commits it. When a manager says the freed hours go to vendor renegotiation on the top forty accounts, and puts a target for that in their own objectives, it is money, and I will defend it in front of anybody. When nobody says that, it is a benefit to morale, and morale does not appear on a P&L.

This is the same problem as unowned work, wearing different clothes. Value that belongs to everybody belongs to nobody. If two names appear against a benefit, nobody owns it. If no name appears, it will not land, and twelve months later the review meeting will be an argument about whether it did.

Which is the other half of the discipline: somebody has to accept the numbers into their budget or operating plan, by name. Not the sponsor by default. The person whose plan actually changes if the benefit arrives. And there has to be a date when that person presents the actuals against what was promised. In most organisations nothing checks whether the return ever showed up, which is precisely why the same optimistic slide keeps getting made.

Ask the question. What will they do with the time, and who is accepting it? Then write down the answer, or write down that there wasn't one. Both are useful. Only one of them is money. Heroes don't scale, systems do, and neither of them run on hours nobody claimed.