# Special Situations

{% hint style="info" %}
**Quick Answer:** Special situations — one-time events, new product launches, discontinuations, and manual overrides — require targeted forecast adjustments that don't fit standard category-level rules. Use Mo's surgical edits, product comparables, and forecast locking to handle these scenarios without disrupting your broader forecast.
{% endhint %}

## One-Time Events

One-time events are demand drivers that aren't part of your regular business rhythm. They create temporary spikes or dips that shouldn't be baked into your ongoing forecast.

### Types of One-Time Events

| Event Type                  | Demand Impact          | Example                                                   |
| --------------------------- | ---------------------- | --------------------------------------------------------- |
| **Flash sale or promotion** | Temporary spike        | 48-hour site-wide sale, influencer collaboration          |
| **Viral moment**            | Unpredictable spike    | Product featured on social media, celebrity endorsement   |
| **External disruption**     | Temporary dip or spike | Weather event, competitor recall, supply chain disruption |
| **Channel launch**          | New demand source      | Launching on a new marketplace or entering retail         |
| **PR or media event**       | Temporary lift         | Magazine feature, TV appearance, trade show               |

### How to Forecast One-Time Events

{% stepper %}
{% step %}

### Estimate the Demand Impact

Before the event, estimate the expected lift or decline:

* **Promotions:** Review past promotional performance for similar events. If your last site-wide sale drove a 2x lift, use that as a starting point.
* **Viral/media events:** These are harder to predict. If you have advance notice (e.g., a planned TV feature), estimate conservatively and monitor actuals.
* **External disruptions:** Estimate the duration and percentage impact on affected SKUs.
  {% endstep %}

{% step %}

### Apply a Surgical Edit with Mo

Use Mo to adjust the specific SKUs and time period:

**Example prompts:**

* "Increase SKU-1234 by 150% for the week of March 15 — we're running a flash sale"
* "Add 500 units to our protein bar forecast for April — we're being featured in a fitness magazine"
* "Reduce outdoor furniture forecast by 40% for the first two weeks of June — our main supplier has a shipping delay"
  {% endstep %}

{% step %}

### Isolate the Event Period

Make sure the adjustment applies **only** to the event window. One-time events should not bleed into subsequent months unless you expect a sustained tail effect.

{% hint style="warning" %}
After a major promotion, some categories see a **demand dip** in the following weeks as customers pulled forward their purchases. Factor in post-event normalization when planning.
{% endhint %}
{% endstep %}

{% step %}

### Lock the Event Period After It Passes

Once the event concludes and actuals are recorded, lock that period to prevent future guideline updates from overwriting the actual event data.
{% endstep %}
{% endstepper %}

### Preventing One-Time Events from Distorting Future Forecasts

When Mo generates your next directional forecast, it will include data from one-time events in its analysis. This can distort future projections if the event was truly non-recurring.

**How to handle this:**

* Note one-time events in your forecast guidelines: "The March 2026 flash sale was a one-time event — do not extrapolate the demand spike into March 2027"
* Use Mo to flag specific periods: "When forecasting Q1 2027, the March 2026 spike was a one-time promotion and should not be used as a baseline"

***

## New Product Launches

New products have no sales history, which means Mo can't generate a data-driven forecast for them. The key tool for this is **Item Comparables** — mapping a similar existing product so Mo can borrow its demand patterns as a starting point.

{% content-ref url="/pages/qHmiH0z8KHuvQI9BqrZh" %}
[Item Comparables](/planning-and-execution/forecasting/special-situations/item-comparables.md)
{% endcontent-ref %}

### Additional Launch Configuration

Beyond assigning a comparable, you can fine-tune the launch forecast:

**Set the Product Launch Date** — In the item details page, set the **Product Launch Date** to tell Mo not to forecast demand before that date.

**Define a Ramp Period** — New products rarely hit full velocity on day one. Use Mo to define a ramp-up:

* "New protein bar (SKU-2468) launches April 1 — ramp to full velocity over 8 weeks based on our chocolate bar launch trajectory"
* "New summer collection launches May 15 — start at 25% of comparable demand in week 1, reach 100% by week 6"

{% hint style="info" %}
A typical ramp period for consumer products is 4–8 weeks. Products with existing brand awareness may ramp faster; products in new categories may ramp slower.
{% endhint %}

**Monitor Post-Launch** — Review performance weekly during the ramp period. After 8–12 weeks, Mo will have enough actual data to start generating data-driven projections.

***

## Product Discontinuation

When you discontinue a product, you need to wind down its forecast gracefully to avoid over-ordering.

### How to Forecast a Discontinuation

{% stepper %}
{% step %}

### Determine the Exit Timeline

Decide when you want to sell through remaining inventory:

* **Hard stop** — Fixed date after which you stop selling (e.g., end of season)
* **Gradual wind-down** — Reduce forecast over several months as inventory depletes
* **Fire sale** — Short-term promotional burst to clear remaining stock
  {% endstep %}

{% step %}

### Apply the Appropriate Guideline

**Gradual wind-down:** "Reduce the forecast for SKU-9876 by 20% per month starting April. Forecast should reach zero by August."

**Hard stop:** "Zero out the forecast for all winter-only SKUs after March 31."

**Fire sale:** "Increase SKU-9876 by 50% for the month of March (clearance sale), then zero out from April onward."
{% endstep %}

{% step %}

### Update the Product Catalog

Once the product is fully discontinued:

1. Archive the item in your product catalog
2. Verify that no future forecasts include the discontinued SKU
3. Confirm that no open purchase orders reference the item
   {% endstep %}
   {% endstepper %}

***

## Manual Forecast Locks

Locking sections of your forecast prevents any changes — including Mo's guidelines and surgical edits — from modifying the locked values.

### When to Use Manual Locks

| Scenario                            | Why Lock                                                 |
| ----------------------------------- | -------------------------------------------------------- |
| **Approved budget periods**         | Finance has signed off on these numbers                  |
| **Committed purchase orders**       | POs are already placed based on these projections        |
| **Completed months**                | Actuals are in; forecast is no longer relevant           |
| **Negotiated supplier commitments** | Changing the forecast would break contractual agreements |
| **Cross-team alignment**            | Multiple teams have planned around these numbers         |

### How Locks Work

* **Locked periods** are unaffected by mass updates via forecast guidelines
* **Mo will not modify** locked values during surgical edits unless you explicitly unlock first
* **You can still view** locked forecast data for reporting and comparison
* **Unlocking** is available to users with appropriate permissions

{% hint style="warning" %}
If you apply forecast guidelines and some SKUs don't update as expected, check whether those periods are locked. Locks take precedence over all other adjustments.
{% endhint %}

### Rolling Lock Strategy

Many teams use a rolling lock approach:

1. **Lock** the current month and all prior months (actuals are already recorded)
2. **Keep open** the next 2–3 months for active refinement
3. **Review and lock** each month as it transitions from planning to execution
4. **Extend** the forecast horizon as locked months roll off

***

## Handling Stockout Periods

When you experience a stockout, actual sales understate true demand — customers wanted to buy, but couldn't. This creates a problem when Mo uses that period to generate future forecasts.

### How to Account for Stockouts

**In your forecast guidelines:** "SKU-5678 was stocked out for 3 weeks in June 2025. Actual sales during that period understate demand by approximately 40%. Adjust June 2026 projections upward to reflect true demand."

**Using surgical edits:** "Increase SKU-5678 forecast for June by 40% to account for last year's stockout — actual demand was higher than recorded sales."

{% hint style="info" %}
If you consistently experience stockouts for certain products, this may indicate your safety stock levels need adjustment. Review your replenishment settings after updating your forecast.
{% endhint %}

***

## Handling Returns and Refunds

Significant return events can distort your sales data and, consequently, your forecast.

### When Returns Affect Your Forecast

* **Seasonal return spikes** — Post-holiday returns in January may create a temporary negative demand signal
* **Product quality issues** — A batch defect causing high returns understates net demand
* **Policy changes** — Extending your return window may shift when returns are recorded

### How to Address Return-Related Anomalies

1. Check if your integration nets returns against sales (most do)
2. If a specific period has abnormally high returns, note it in your guidelines so Mo doesn't understate demand for that period
3. For recurring return patterns (e.g., post-holiday), include them in your seasonal guidelines

***

## Frequently Asked Questions

### How far in advance should I set up a new product launch forecast?

Set up the product launch date and comparables as soon as the product is confirmed in your catalog. For the detailed ramp forecast, 4–6 weeks before launch gives you enough time to align with your replenishment plan.

### Can I apply one-time event adjustments retroactively?

You can adjust forecast numbers for past periods, but this primarily affects accuracy reporting rather than operational outcomes. It's more useful to note the event in your guidelines so future directional forecasts are not distorted.

### What happens if I lock a period and then need to change it?

You can unlock any locked period if you have the appropriate permissions. After unlocking, make your changes, then re-lock. Use this sparingly — frequent lock/unlock cycles can create confusion about which numbers are final.

### How do I forecast a product that's been out of stock for months?

Use product comparables to establish a demand baseline. If you have pre-stockout sales data, ask Mo to project based on that period rather than the stockout period. You can also manually set projections based on your expected demand at relaunch.

### Should I create a separate scenario for special situations?

For major events (like BFCM or a large product launch), creating a separate scenario can be helpful for comparison. For smaller adjustments, incorporate them directly into your active forecast to avoid scenario sprawl.

***

## Related Guides

{% content-ref url="/pages/KXh7MTxw84WmaJhhHp6g" %}
[How Mo's Forecasting Works](/planning-and-execution/forecasting/how-mos-forecasting-works.md)
{% endcontent-ref %}

{% content-ref url="/pages/a4VaHeybGOn75f976xpb" %}
[Setting Up Forecast Guidelines](/planning-and-execution/forecasting/setting-up-forecast-guidelines.md)
{% endcontent-ref %}

{% content-ref url="/pages/m7P5Z47OAkdJCWudsFQK" %}
[Customize a Forecast](/planning-and-execution/forecasting/customize-a-forecast.md)
{% endcontent-ref %}

{% content-ref url="/pages/D2BTtZoIEwFaGlVVWPXZ" %}
[Forecast Best Practices](/planning-and-execution/forecasting/forecast-best-practices.md)
{% endcontent-ref %}

{% content-ref url="/pages/ja1XQcZPneh77RB6KNv3" %}
[Build your forecast with Mo](/planning-and-execution/forecasting/build-your-forecast-with-mo.md)
{% endcontent-ref %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://learn.moselle.io/planning-and-execution/forecasting/special-situations.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.