> For the complete documentation index, see [llms.txt](https://help.modelreef.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.modelreef.io/how-tos/scenarios-and-planning/build-a-sensitivity-analysis-pack.md).

# Build a Sensitivity Analysis Pack

This guide explains how to build a sensitivity analysis pack in Model Reef by creating dedicated models that vary specific assumptions and then summarising their impact on key outputs.

Model Reef does not have a one-click data table feature. Instead, you produce a set of carefully designed models that each represent a point on the sensitivity curve.

Before you start

You should have:

* A solid Base Case model.
* A clear set of assumptions you want to test, for example:
  * Revenue growth.
  * Gross margin.
  * Capex levels.
  * Discount rate.
* Valuation metrics configured (NPV, IRR, Money Multiple) if you want value sensitivities.

If needed, review:

* Build a DCF Model (FCFF)
* Build a Valuation Sensitivity Model

What you will build

* A small family of models where only one or two assumptions change.
* A summary of how outputs respond as each key assumption is moved.
* A pack that can be shared with stakeholders showing risk and upside.

{% stepper %}
{% step %}

### Choose sensitivity dimensions and ranges

Select a small number of variables to test, for example:

* Annual revenue growth rate.
* Gross margin percentage.
* Capex per year.
* WACC or equity discount rate.

For each, define a range of values that you want to test, such as:

* Base growth plus or minus 5 percentage points.
* Margin increasing from 50 percent to 60 percent.
* WACC between 8 percent and 12 percent.

These ranges should be realistic and focused on the questions people actually ask.
{% endstep %}

{% step %}

### Create sensitivity models from the Base Case

For each sensitivity dimension, create separate models.

* Duplicate the Base Case model for each point you want to test, for example:
  * `Model - Sens - Growth 15pc`
  * `Model - Sens - Growth 20pc`
  * `Model - Sens - Growth 25pc`
* In each model:
  * Adjust only the targeted assumption, for example revenue growth drivers.
  * Leave everything else unchanged.

Repeat this process for each dimension, keeping changes isolated to one assumption per group of models where possible.
{% endstep %}

{% step %}

### Update valuation and outputs in each model

In each sensitivity model:

* Ensure valuation settings are appropriate.
* Record key outputs for that point, for example:
  * Revenue in key years.
  * EBITDA in key years.
  * Project NPV and IRR.
  * Equity IRR and Money Multiple.
  * Minimum cash balance.

You can export these values or copy them into an external summary table.
{% endstep %}

{% step %}

### Build sensitivity tables and charts outside the models

Once you have collected outputs from each sensitivity model, arrange them into tables or charts, for example:

* Growth rate along the horizontal axis and NPV on the vertical axis.
* Margin assumptions versus IRR.
* Capex levels versus minimum cash balance.

This step is typically done in an external tool or documentation, using Model Reef as the engine that generates the underlying numbers.
{% endstep %}

{% step %}

### Create a narrative sensitivity pack

For communication purposes, assemble a pack (for example a deck or report) that includes:

* A short description of each sensitivity dimension and why it matters.
* The tables and charts built from your models.
* Key non-technical messages, for example:
  * Value is more sensitive to margin than to growth beyond a certain point.
  * Small changes in WACC have a meaningful effect on valuation.
  * High capex strategies significantly increase funding risk.

This pack gives stakeholders a structured view of risk and leverage points in the model.
{% endstep %}

{% step %}

### Keep the sensitivity family in sync with model updates

When the Base Case changes, the sensitivity models may become stale.

* Decide how often to refresh the sensitivity pack, for example:
  * After major assumption updates.
  * After board meetings or financing events.
* When refreshing:
  * Start from the updated Base Case.
  * Recreate only the sensitivity models that are still relevant.

Keeping a small, focused set of sensitivities is easier than maintaining a very large grid of cases.
{% endstep %}
{% endstepper %}

Check your work

* Sensitivity models are identical to the Base Case except for the targeted assumptions.
* Outputs are consistently measured across all models.
* Tables and charts clearly show how outputs respond to assumption changes.
* Stakeholders can understand both the range of outcomes and which assumptions matter most.

Troubleshooting

<details>

<summary>Too many models become hard to manage</summary>

Reduce the number of sensitivity points and focus only on the most informative ones.

</details>

<details>

<summary>Sensitivity results look erratic</summary>

Check that assumptions are being changed consistently in each model and that there are no unintended structural differences.

</details>

<details>

<summary>Stakeholders are overwhelmed by detail</summary>

Highlight the top few sensitivities and summarise the rest in an appendix or technical note.

</details>

## Related articles

* [Promotions & Margin Sensitivity](/use-cases/hospitality-groups-multi-venue/promotions-and-margin-sensitivity.md)
* [Actuals Import](/help/xero-integration/actuals-import.md)
* [What Is Model Reef?](/help/getting-started/what-is-model-reef.md)
* [Chart & Table Syntax](/syntax/chart-and-table-syntax.md)