The Forecasting function in Toolio's Item Planning module streamlines the sales projection process at the item level. Forecasting is primarily used for creating a demand plan for core / existing products but with Toolio's feature rich forecast function we are able to create demand plans for all types of products. Here is a breakdown of this article.

Please see below for a walkthrough of using the forecasting feature in Item Plan.

1. Navigate to your Toolio - In-Season Planning view in Item Plan. This view will be preconfigured out of the box in your Item Planning module and is designed to help visualize your forecasted sales and inventory values.

2. Using the filter, group, and sort functions, you can easily configure the view to display the products you are interested in forecasting.

3. By clicking on any of the checkboxes, you can select the item(s) you would like to forecast. You can then right click and select the Forecast feature which will take you to the Forecast menu.

Please note that you can select and forecast multiple items at once. This may be particularly useful if you want to forecast your products bulk.

4. Now that the Forecast menu is open, you will see the number of items selected, the time frame selected to forecast and the scenario to forecast to. Below that the Forecast Profile can be set to Use Defaults or Override Defaults. Please read more about Forecast Profiles here.

5. Let’s click on the Override Defaults to expose the full menu.

Please see a full breakdown of each selectable field below:

This is where you will define the sales curve of your forecast.

Please see below for more information on the forecasting algorithm for both of these methods.

Choosing Merchandise Plan as your Seasonality Method:

Choosing Sales History as your Seasonality Method:

Check this box if you want to drive your seasonality from a category or item that has a comparable selling trend but may have more history. This is also great for helping plan our new items. Review these articles (Like for Like and Forecasting New Choices) to learn more.

This is where you can specify attributes to define the scope of your seasonality curve. If you choose to leave attributes blank, your forecast will follow the highest level sales curve from your source.

As an example, if you were to add the Division as an attribute, Toolio will then specifically model the forecast at the Division level from the Seasonality source defined in the previous step. You can choose to add as many or as few attributes as you see fit, depending on the granularity you want the Seasonality curve to match.

When you have selected Merchandise Plan for your Seasonality Method the Seasonality Groups will list any attributes you plan in the Merchandise Plan module.

When you have selected Sales History for your Seasonality Method the Seasonality Groups will list every attribute assigned to your item from Division to SKU.

Now that we have your Seasonality curve fully defined, this is where we will define the input in terms of sales volume for your forecast. This dropdown allows you to select what time frame the sales performance is sourced from. If you are selecting a multiple week period, the average of those weeks will be the Volume input of your forecast.

You can either select the trailing few weeks, or you can select a Custom Time Frame that is best representative for the group of products you are forecasting.

The Volume Factor % will change the forecast results by the selected factor. This means if you have 100% it will give the results as intended by all of the parameter you entered in to the previous fields. Let's say you are forecasting a style using a like for like but this style is a bit more expensive, you may not want to use the the like for like verbatim. Based on your knowledge this new style should be about 80% of the like for like, this is where the Volume Factor % will come into play. The default is 100% but you can go either direction from this. If you want to double the forecast; 200% and if you want to halve the forecast 50%.

6. Once you have configured the above steps, you can simply press Forecast and your sales projections (including Sales Cost, Sales Retail, Sales Units and Allowances, i.e. Discount & Markdown) will be generated accordingly. You can always edit these values further and fine tune your forecast as you see fit.

When using the Sales History, your Gross Sales Ticket is derived from the Gross Sales Units output from the forecast function at the Ticket Price in the variant table feed or from a pricing table feed. Gross Sales Retail is derived from the Gross Sales Retail ratio to the Gross Sales Ticket in The Volume Trend and applied to the Ticket Price to generate a Retail Price which is then multiplied by the forecasted Gross Sales Units. This will also output a Discount and Discount %.

This gives you the opportunity to have the size penetrations be based on a size spread that could be more accurate than the current trend or the history. This is good for items that may not sell every size every week and for time when you had a stock out and customers may have sized up or down. These are the same size spreads you use in the Assortment Plan module when setting up newness. Check out this article for a deeper dive into this option.

Flow On Order to Receipt

Use Case: Primarily be used for non-replenishment, i.e. seasonal items, and is relevant only if you select `Constrain sales by inventory` as well.

Check this box, if you want to assume on order items as the receipts plan, so that on order will be put into consideration while forecasting sales.

Use Case: Primarily be used for non-replenishment, i.e. seasonal items.

This option will suppress your sales forecast based on the inventory and receipts plan.

Toolio provides multiple options for forecasting returns. Below are the options:

The forecasting algorithm for Merchandise Plan Method leverages two core concepts, Volume & Seasonality. This approach looks at the penetration of the items you are forecasting to the Seasonality Groups you selected based on your forward plan in your Merchandise Plan module.

For Example:

This approach looks at the penetration of the items you are forecasting to the Seasonality Groups you selected and how they performed in the historical time frame you selected.

For Example:

This approach looks at the penetration of the items you are forecasting to the Seasonality Groups you selected and how they performed in the historical time frame you selected. Historical performance is assessed by looking at a Rolling Average for the time frame, using the trailing X weeks for each weekly forecast in order to smooth out any spikes or dips that may have occurred.

For Example:

In this approach, you are able to maintain the original shape of your forecast but adjust it up or down depending on the a rolling trend defined by the number of weeks to evaluate. The main business reason for this method is to prevent the planned seasonality from shifting as a result of forecasting, this is likely the case where users have adjusted their forward seasonality index/builds to account for upcoming seasonal shifts that may not appear in history.
​

To apply the Item Plan seasonality to a forecast

In the example below a rolling trend for the last four weeks is used against the plan for the last week to adjust the volume for the future forecast based on the latest trends but note the shape of the future forecast remains the same and the builds remain intact.

​

Forecast Profiles

Choosing the Right Forecast Method

Item Plan Reporting

Generating Receipts in Item Plan

Like for Like

Forecasting New Choices

Size Spread

Forecasting pulls only items that are live in Item Plan on the time-frame that you're currently looking at. To be able to add items that are in 2022 into your 2023 forecast, make sure that you do the forecasting while looking into 2022. Steps are:

Please see Choosing the Right Forecast Method

Forecasting first forecasts Sales Units, and then the other metrics are derived either off of Variant level pricing information or Inventory cost. You can see how other metrics are calculated in the forecasting process here.