Major improvements in supply planning

šŸš€ New Features

  1. Brand new daily average sales calculation

Since the introduction of our supply planning feature, we have always been discussing with our customers how we could improve the accuracy of the suggested restock count. One of the core aspects of the suggestion is the calculation of daily average sales per product. The initial implementation works most of the time as expected but it has some shortfalls under some circumstances.

Eg. if the product runs out of stock for a certain period of time, the daily average sales decrease since there are no sales for these days. But that could be misleading until the product will gain sales again. Therefore, out-of-stock days should be excluded when calculating the daily average sales of the product.

Another example is related to forecasting days. Our users can define the forecasting days for how many days our system should look back to calculate daily average sales. The problem is that if forecasting days are older than when the day product has a sales record, then the daily average calculation could be misleading as well. Therefore, the system should take into account both forecasting days and the first day of the product sales record and adjust the calculations based on these pieces of information.

After working on these circumstances carefully, our team has introduced a brand new daily average sales calculation engine which solves all the problems mentioned above.

How does the new daily average sales calculation work?

First of all, the first date of the product sold is taking into account to be able to mark the starting point of the history. Let’s say, the product’s forecasting days have been set 90 days, but the product has introduced to the market 25 days ago. To cover these issues, when collecting the sales and stockout history, only the last 25 days will be counted by the system. Of course the other day, it will be the last 26, 27, 28 days, and goes on. Thus, even though the forecasting days are 90 days, the system will adjust itself based on the facts.

The new algorithm starts calculating the daily avg sales from the beginning of product launch even though the “forecasting days” were passed before the launch. Similarly, it also ignores the out-of-stock days when calculating the sales average. In this way, we are able to offer more accurate results.

After marking the starting point, the system will collect sales history and the number of days between the starting point and today (it is not included since it has not finished yet). Stockout days are going to be deducted from the number of days since there are no sales for these dates and they should not be taken into account.* After calculation how many products are sold within the defined period and how many days should be used for calculation, daily average sales is calculated by sales count/days formula.

* If the end of date's stock count is zero or below, that day is marked as the stockout date for the product. But if there is a sale on date day, it's not deducted from the number of days product sold while calculating the daily average sales. Sometimes, the product could have sales and then stockout inside the same day. Therefore, that day should be counted. This is also applicable to the products which are sold as pre-order. Those products' stock counts are always below zero cause they don't exist in the warehouse yet.

Why is it important to have accurate daily average sales?

How well the product performs affects which products should be prioritized for restocking. Since there are always limited financial resources available at a time, deciding which products will be restocked and when it should be going to happen are crucial operations to have a sustainable business. The core issue is to compare the sales performances of the products in your inventory. For a fair comparison, the daily sales performance of the products should be calculated in their own terms. That’s why we believe the improvements in the calculation we made will help you to have a better picture of your next supply plan.

We have also updated the sales history popover to show you not only the sales history of the product but also how many days it’s stockout within different sales periods.

šŸ”Š Chores

 1. Listing filter items in alphabetic order

Day by day our filter items are growing and sometimes it’s time-consuming to find the right filter item to use. That’s why all filter items are listed alphabetically from now on.

2. Allow editing approved purchase orders

Previously, approved purchase orders are locked and not allowed to get any updates. This comes from the reason where any unwanted changes are prevented. But our users informed us that sometimes suppliers can not provide an agreed amount of products therefore they need to update even when they’ve agreed with suppliers before.

Because of this, the locking mechanism is removed. Instead, the warning is shown if the PO is about to change is approved.

3. Renamed menu items & additional info

We’re working hard to make every aspect of our solution to be easily understandable. For this reason, we’ve updated some of our menu items’ names, table column names, and their definitions.

4. Tax number limitation removed

To support different tax number storage, we have removed the validation of the tax number of the suppliers that we previously added.

šŸ› Fixes

1. Non-translated error messages

This week, some of our users have been faced with error messages which were not translated into English. Our team quickly fixed the non-translated messages and add all of their translations.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

You may also like