Entia Non Sunt Multiplicanda.

Seven hundred years ago, an English friar named William of Ockham wrote down a sentence that has outlived almost every theory it has been used to evaluate. Entia non sunt multiplicanda praeter necessitatem. Entities should not be multiplied beyond necessity. He was arguing about metaphysics. We use the same idea every week to argue about decks, dashboards, technology stacks, and engagements. The argument always lands the same place: the simpler answer almost always works better. Not because complexity is wrong, but because most of it is unnecessary, and unnecessary complexity is a tax the client never agreed to pay.

The Common Ground.

A child learns to add fractions before they learn to do anything useful with them. Half plus a third. The teacher writes it on the board. The first move is to find the least common multiple. Two and three share a common ground at six. Half becomes three sixths. A third becomes two sixths. The answer falls out. Five sixths.

The lesson sits underneath the arithmetic. To add unlike things, find the simplest common ground between them. Not the largest. Not the most clever. The simplest. That is the move the math actually rewards.

Replace fractions with people, departments, vendors, technologies, or campaign objectives, and the lesson holds. Most disagreement at the working level dissolves once the room finds the underlying common ground. Most strategy work that fails, fails because the team optimized something else instead — the breadth of the answer, the elegance of the framework, the fee tied to the size of the engagement.

Ockham's discipline, in one sentence.

Ockham was a Franciscan friar working in Oxford and Avignon in the early fourteenth century. He spent most of his career arguing with the church about poverty, papal authority, and the limits of human knowledge. His name became attached to the maxim only later, but the idea is his. Pluralitas non est ponenda sine necessitate. Plurality should not be posited without necessity.

Translated into how we work, the rule reads as a question: what would happen if we removed this? If the answer is "nothing meaningful," remove it. The agency, the product team, the marketing function, the tooling stack: each of them benefits from being asked that question once a quarter and answered honestly.

The razor, in practice.

Among competing explanations, take the one that requires the fewest assumptions. Among competing plans, take the one that adds the fewest moving parts. Among competing pitches, take the one that survives a stranger reading it cold. The razor is a heuristic, not a law. It will be wrong sometimes. The simpler answer is not always the correct answer. But it is almost always the right starting answer, because it is the cheapest one to be wrong about and the easiest one to revise.

When a senior team disagrees on a path, the simplest path that fits the evidence is usually the one with the highest expected value. Not because it is right, but because the cost of being wrong is lower, and the team learns faster from the result.

Inside Innovative Group we describe this as Occam the Slasher. The razor protects an idea from being multiplied. The Slasher takes the same blade to a body of work that has already been multiplied — to the deck that grew to eighty pages, the campaign that depends on six tools, the strategy that requires seven separate operating principles to defend itself. Cutting back to the working core is its own discipline.

The evidence is one-sided.

If the argument was only philosophical, this would be a shorter note. The empirical record on simplicity versus complexity in business forecasting is unusually clear. A 2015 review in the Journal of Business Research examined thirty-two published studies that pitted simpler forecasting models against more complex ones in head-to-head comparisons.

Zero. Across a body of research designed to find the case for sophistication, no paper made it. That does not mean every model should be a moving average. It does mean the working assumption inside any business forecasting room should be: the simpler model wins until proven otherwise. That is the opposite of how most organisations behave when they purchase forecasting tools.

How this shows up at Innovative Group.

The KISS principle (the U.S. Navy abbreviation, attributed to Lockheed engineer Kelly Johnson in 1960) sits behind a handful of operating choices we make differently than peers.

Engagements get scoped narrow first. We would rather start with one workflow, one funnel stage, one product surface, and prove the work, than sell a transformation. The transformation is sometimes the right end-state. It is almost never the right starting point.

Stacks get audited for subtraction. When we walk into a marketing operation, the first inventory is what to remove. Tools that overlap. Reports nobody reads. Meetings that exist because they used to exist. The savings fund the work that actually moves a number.

Models get sized to the job. Our AI Solutions practice runs small, tuned models inside the client's environment for the workflows that need them, instead of routing every query to the most expensive model in the market. Smaller models often win on accuracy in the narrow domain. They always win on cost and latency.

Decks get shorter. Strategy gets quicker. Decisions get clearer. None of that is glamorous. It is, mostly, the discipline of refusing to add.

What we refuse to add.

A short, opinionated list. These are the additions we have learned to decline, no matter how good they look on a roadmap.

• One more toolIf the existing stack can do it within two business days of configuration, the new tool waits. New tools earn their seat by doing what the existing ones cannot.
• One more dashboardIf a metric does not change a decision someone is going to make, it does not need a tile. Most dashboards are decoration the team has stopped reading.
• One more meetingRecurring status meetings without an open decision are a tax on calendars. We replace them with written updates and reserve live time for choices.
• One more frameworkThe room rarely needs another acronym. It usually needs the existing strategy executed well for two more quarters.
• One more dependencyEvery added integration is a future failure mode. Internal capabilities that already work earn priority over net-new vendors that promise to.