The 3-Layer Model for Better Financial Decisions

How often does a financial model answer every question… except the one that actually matters?

It is the type of analysis that explains everything but decides nothing.

You enter the room: the slide count is impressive, the numbers are right, and the tables are clean and nicely formatted... And yet, when it comes the time to making a call, everyone looks at each other and says, “So, what’s next?”

Something failed; the uncomfortable silence after that tricky (but natural) question makes it obvious. Was it the structure of the analysis? Was it the conclusion? Or maybe the premises? It is clear that something broke between the spreadsheet and the room where decisions are made, and the answer could be more structural than idiosyncratic.

Most financial work fails because layers collapse

Most financial work fails when its layers collapse into each other, even when the inputs are right.

We make hasty conclusions based on partially processed data (lack of depth), or worse, we produce rigorous analyses that never translate into an actionable frame for the person who has to decide.

Both errors share the same root: a lack of discipline around where one layer ends and the next begins.

The 3-layer model is a corrective framework. It forces a simple but consequential distinction:

1) Data (structuring the chaos)

Raw information is rarely decision-ready. It’s fragmented, inconsistent, and often misleading. Structuring data typically means defining variables, cleaning noise, and establishing a coherent base, but there is a critical preliminary step often ignored. Ask yourself: what does this dataset actually tell me, and what assumptions are baked in before I even start? This is where ambiguity is detected and where the true context is set.

2) Analysis (processing with criteria)

This is where relationships are built, sensitivities are tested, and trade-offs are made explicit. Causality lives in this layer, and although the full picture is still blurred, you set the basis for understanding. Analysis is not about producing refined numbers or showing trends on a graph. The key is understanding how the system behaves when something moves.

“Revenue went up 12%” could be a factual observation, but an irrelevant one if the causes are not addressed: “why did it go up, what drove it, what happens if that driver weakens, and what does that mean for margins and cash flows?”

A large portion of financial work dies here, in the second layer; all the substance gets trapped in models, dashboards, and nicely formatted spreadsheets that explain everything and decide nothing.

3) Decision (translating into action)

This layer forces clarity. It asks: given everything we know, what do we actually do?

“Expand into this market.”
“Adjust pricing by 8%.”
“Delay the investment six months.”

The reality, however, is that many analyses never reach this point. They remain technically correct but strategically useless.

A decision is a commitment under uncertainty

Every financial analysis should lead to clear implications. Not a summary of what was found, but a frame for what needs to be decided, with the relevant constraints, trade-offs, and the cost of inaction made explicit.

Each layer has a distinct role in the structure, and blurring the limits often leads to failed financial work, for example, when someone presents Layer 1 outputs as if they were Layer 2 conclusions, or hands over a Layer 2 analysis without doing the Layer 3 translation for the people who have to act on it.

Better financial models and frameworks are always desirable; however, those should not be a goal to pursue in isolation. We need a coherent system, one where messy inputs flow through a disciplined structure and come out the other side as decisions that can actually be made.

Finance in action: evaluating a new product line

Your team gathers historical sales data, benchmarks competitors, and builds a detailed forecast model. The output includes scenarios, sensitivities, and beautifully structured dashboards.

But when the CEO asks, “Should we launch?”, the room hesitates.

Why?

Because the work stopped at analysis.

Nobody knows. Because everything presented was a blurred representation of Layer 1 and Layer 2: structured, even insightful, but ultimately inert.

No explicit translation was made from model outputs to decision criteria. No thresholds were defined. No risk tolerance was articulated.

A (proper) Layer 2 intervention would have asked: which variances are structural versus one-time? Which cost lines are behaving differently from the revenue trend, and why? What do the sensitivities look like under the base case versus a conservative macro scenario? That is analysis, processing data through relationships to set a system-level understanding, not just organizing it.

Now contrast this with a disciplined 3-layer approach:

• Layer 1: Clean segmentation of customer demand, cost structure, and pricing assumptions.

• Layer 2: Scenario modeling with clear drivers (volume elasticity, margin sensitivity, capital requirements).

• Layer 3: Defined rule: “We launch if expected IRR exceeds X% under base case, and downside scenario remains above Y%.”

Same data. Similar analysis. Different outcome.

Clarity emerges only when the third layer is explicitly designed into the process.

A Layer 3 framing would then take those findings and say: given this, guidance revision is warranted if X holds; if Y deteriorates, the more conservative range is defensible. Here are the two decision paths and what each one requires us to believe.

The 3-layer model expands beyond FP&A. In investment analysis, it is often observed that the different steps of the analytical process are condensed or overlapped: a financial model that outputs an IRR is Layer 1 work dressed as Layer 3. The analytical work is Layer 2: what drives that IRR, how sensitive it is to exit multiple assumptions, what the downside scenario implies for covenant headroom, etc. The decision frame is Layer 3: Does this return adequately compensate for the risk concentration, and what would change our view?

The financial model is not the answer; it is just another input in the process.

Beyond “doing the financials”

When we are hired to “do the financials,” the implicit ask is often Layer 1. What we should be delivering is Layer 3, with Layers 1 and 2 as the visible scaffolding. That reframe, from financial producer to decision architect, is, in my view, where the real value of analytical work sits.

The 3-layer model is not a tool I invented. It is a structure that emerged from repeatedly noticing where the work stopped being useful. The discipline it imposes brings clarity by forcing each layer to do only what it is supposed to do.

When data tries to answer questions, it becomes biased or erroneous.
When analysis tries to decide, it becomes overconfident.
When decisions ignore the previous layers, they become fragile.

Each layer has a role. Respecting that structure is what turns financial work into a system rather than a collection of outputs.

Better data won’t fix a broken process.
More analysis won’t guarantee better decisions.

What matters is the integrity of the chain: Data → Analysis → Decision.

The individual steps simply feed a system designed to transform ambiguity into action. Because at the end of the day, the goal was never a cleaner model. It was always a better decision.

Last time you delivered an analysis, where did your process break: data, analysis, or decision?