A MESA GROUP RESEARCH SUMMARY

The Model Is Ten Percent of the Problem

BCG's 10-20-70 finding, and why most AI budgets are built backward

Boston Consulting Group  ·  A Mesa Group summary

Ask a room of executives why their AI initiative stalled and most will describe a technology problem. BCG's work across enterprise AI engagements says the technology was never the variable.

The finding

10% algorithms  ·  20% tech and data  ·  70% people and process

BCG holds that successful AI at scale breaks down along a specific allocation. Ten percent of the effort belongs to the algorithms. Twenty percent belongs to technology and data. The remaining seventy percent belongs to people and process.

Stated as a budget, most organizations run it in reverse. They spend heavily on platforms, tools, and model selection, then treat workflow redesign and change management as something to handle after the technology is chosen.

Where it comes from

This is not a controlled experiment with a published sample and a confidence interval, and it should not be presented as one. It is an operating principle BCG derived from its work across a large volume of enterprise AI engagements, and the firm publishes it directly as guidance to clients.

What gives it weight is the source and the direction. A consultancy with every commercial reason to sell technology transformation is telling clients that ninety percent of the value sits outside the model.

Why the allocation gets inverted

The model is the visible part. It demos well, it has a name, it has a price, and it can be purchased on a timeline. The seventy percent has none of those qualities. It is the redesign of how work moves, who reviews what, and where a decision stops for a human.

That work is slower, harder to scope, and impossible to photograph. So it gets deferred, and the deferral is invisible until the pilot fails to scale and no one can say precisely why.

Why it matters

If seventy percent of the outcome depends on people and process, then the most consequential AI decisions in a company are not technology decisions at all. They are decisions about how the work is arranged and who is accountable for what.

Those decisions get made before a model is ever selected. Or they do not get made, which is also a decision.

THE ORIGINAL STUDY

Boston Consulting Group. The 10-20-70 approach to deploying AI at scale.

https://www.bcg.com/capabilities/artificial-intelligence

This page is a Mesa Group summary of independent research published by the organization named above. Mesa Group has no affiliation with it. Read the full study at the link.

ABOUT MESA GROUP

Mesa Group is a professional services firm in the Command Mapping category. The firm maps decision authority across an organization's operations and builds the working systems that authority calls for, delivered with full client ownership. Mesa Point leads the mapping practice. Mesa Built leads the building practice.

The firm's principle is constant: automation with authority.