Forecasting Inventory: Best Practices, Model Types, And More

No matter what size your ecommerce business is, inventory management is key to your success. If your business is small or you have a start-up ecommerce business, tracking inventory is not that difficult. For those of you that have larger companies, forecasting inventory can seem like a daunting task.

The truth of the matter, inventory is a key pillar of your business and one that demands accuracy and timing at every step of the process.

As your business grows and you need larger quantities of products to meet demand, proper inventory planning becomes extremely important. This is one of the main reasons larger ecommerce companies outsource order fulfillment to a third-party-logistics (3PL) provider.

Most companies don’t have the correct infrastructure in place to handle rising demand and sales growth. If you don’t have the resources to manage inventory, inventory turnover, shipping, and ordering, outsourcing to a fulfillment center is an immediate solution.

Forecasting Your Inventory

It’s near impossible to forecast your inventory without the right data. The key to knowing how much inventory to order is knowing how your inventory behaves historically. Having the right order management software to track all of your multichannel sales in one centralized location is key.

Here at Thill, our OMS allows you to opportunity to monitor all of your on hand inventory and units sold per day. You’ll have the ability to track all your SKUs, run reports and see which sales channels are performing the best.

Our OMS integrates with several top ecommerce platforms for your order fulfillment, such as Shopify, Magento, Amazon, Ebay and others.

3PL fulfillment companies also have software that can notify you when you need to recorder products, which is another benefit you get partnering with a 3PL company.

Besides historical inventory data, you’ll need to keep seasonal trends and the holidays in mind when forecasting inventory. Furthermore, you’ll need to adjust inventory based on marketing, seminars or television appearances. If you’re planning outsourcing 3PL fulfillment, you need to make sure you let your fulfillment partner know ahead of time of such events.

In total, there’s a lot of different things that can affect your inventory.

  • Seasonal Inventory
  • Supply Chain Factors
  • Inflation
  • Disasters
  • Competition
  • Markerting
  • Events, Seminars, Trade Shows, Sports
  • Law Changes
  • Technology Failure
  • Sales

What Are the Types of Inventory Forecasting?

For those that have been forecasting inventory for years, experience does play a role in the equation. While that may true, all forecasting is formulaic at some point. Due to this, you should be asking this specific question, “which inventory forecast method best fits what we need?”

The core formulaic methods for proper inventory forecasting would include;

  • Trend
  • Graphical
  • Qualitative
  • Quantitative

That may or may not be lingo you understand, but the experts that are tasked with forecasting should focus on the best method based on known issues and inputs, whether that be with stock, i customer input, sales, mathematical analysis or even market research.

Now that you know the core 4 inventory forecasting methods, let’s take a look at how each formula differs.

(1) Trend Forecasting

One of the most common and well known forecasting models is focused around trends. Trends are going to refer to changes in demand for a specific product over a period of time. This method would project outline potential trends, but it does not include seasonal trends (such as the holidays, winter vs summer). It also doesn’t include irregularities (such as a specific month last year bringing in 12 percent more sales). Instead, it concentrates on granular sales data because this specific forecasting technique would show how likely customers are to make purchases in the future. These forecast are valuable in penetrating new markets or helping build marketing campaigns around sales.

(2) Graphical Forecasting 

If you’re familiar with trading, you know how important charting is to your TA (technical analysis). A great TA can visually look at a chart and forecast where price action may go. Likewise, if you’re in sales, charts and graphs can give you a great idea of what direct sales is trending. This same data used in trend forecasting can be graphed in a chart to give you visual representation to show how forecast are growing, declining, or stabilizing. Many forecasters prefer using this graphical method because it is visual, they can see patterns easier versus just looking at data.

(3) Qualitative Forecasting 

If your company doesn’t have a lot of historical data to rely on, how in the world could you forecast inventory? Well, if that is the case, you can go one direction and that source is your customers. Qualitative forecasting involves extracting data from your customers (like market research, focus groups, interviews, reviews). Some forecasters will be able to collect this data and turn that data into forecasting models.

(4) Quantitative Forecasting

This last forecasting model is generally considered to be more reliable than qualitative research alone, the idea of quantitative forecasting is using past numerical data to create a forecast model. Of course, the more past data you have, the better the forecast is going to be. One common example of quantitative forecasting is time-series forecasting, this uses what is called temporal quantitative data to build a forecasting model. Companies that use this specific example are trying to model potential future trends.

How To Choose The Right Forecasting Method

Now that you know the core forecast methods that are used, how do you choose the right forecasting method for your company?

First things first, you have to ask yourself, “what type of data do we have available to use?” Every business owner or executive reading this is going to have a different answer to this. If your company is established and has been in business for 30 years, you’re going to be able to start with historical data, you can likely use the quantitative approach. If your company is only a year old, you won’t have much historical data to reply on. Even if you have some data, it’s likely not enough for an accurate model. Rather than the historical approach, you’d likely start by collecting qualitative market information about your customers.

The good news, you don’t have to rely on just one model to get a forecast. In fact, the best forecasting uses multiple methods and data types for modeling and forecasting.

While this is true, you still want to make sure your models reflects wild cards, you always want to account for unpredictable trends and factors that can shift demand immediately. Those events are going to happen, like a black swan event, so you always want to be prepared for those.

How To Forecast Inventory

In our example below, we’ll be using a SKU example since nearly all of you have this type of system in place.

Before we jump into that, here’s a few things you need to remember.

  • Make sure you’re forecasting specific periods, such as 30 days, 60, days, 90, days, 6 months, year, etc.
  • Make sure you’re always using a base demand for your forecast. For example, if your company sold 1,000 units during the last period, 1,000 would be your base.
  • Make sure you always consider marketing activities as marketing campaigns can boost sales.
  • Make sure you always review sales velocity – this refers to how fast sales are moving through your company. It should be based on the number of leads, average deal value, conversion rate and sales cycle length.

Each SKU has a specific workflow. You can forecast inventory if you know each level per each SKU you have.

  • Maximum Stock Level – This is the max stock level for any specific SKU you have.
  • Reorder Point – Your reorder point is when you place a new order for inventory. You have to keep in mind how long it takes for you to get new orders to the warehouse and stocked.
  • Low Stock Warning Level – This level is your warning for a SKU that is going out of stock.
  • Restocking Level – This is the level after reordered product is delivered to your warehouse.

Your SKU catalog would have each level per SKU. In your supply chain, keeping track of your inventory levels is vital. The more SKUs you have, the bigger the challenge. It’s easy to see why having inventory management software is important but necessary.

(1) Reorder Points

Let’s go a little deeper on reorder points as they’re vital to all ecommerce businesses. Your reorder point is the answer to “when you need to reorder a product.” This is the level that triggers an action to replenish your stock.

Now, the lead time refers to the time it takes products to be delivered. You have to take this into account. We’ll talk about this more in just a minute. You can use a basic reordering formula as follows;

Reorder Level = Average Daily Usage Rate X Lead Time (In Days)

Some ecommerce companies will keep safety stock on hand. This gives you extra stock on hand just in case there’s a rise in demand.

(2) Lead Time

We touched on this just a second ago, but when you place an order to your supplier, it can take time for that inventory to reach your warehouse. The time that it takes in between is your lead time for that product. It’s important to keep record of this per SKU so you know in advanced of ordering.

If your suppliers are local, your lead time is going to be less than that of a supplier that’s far away. If your supplier is overseas, it can take weeks to get your order. The rule of thumb is making sure you have enough inventory on hand to last during the lead time.

Another word you need to become familiar of is lead time demand.

Lead Time Demand – This refers to what you expect to sell during the lead time period.

(3) Stock-Outs

Stock-outs refer to when customer demand for a specific product exceeds the amount of inventory a business has on-site. The reason stock-outs are bad is because you can lose both sales and customers due to them. In order to prevent them, safety stock is a good start.

(4) Safety Stock

Safety stock refers to having extra product inventory in storage just in case demand picks up. You can calculate safety stock by subtracting historical data on maximum usage and average daily usage.

Forecasting Boundaries

There’s 3 more terms we want you to become familiar with.

  • Base Demand – Your base demand is the starting point for forecasting inventory, usually based off your current demand for any given SKU.
  • Trending – Trending refers to increases and decreases in demand over a period of time. If you can identify trends, your forecasting will be more accurate.
  • Forecasting Duration – This refers to your forecast period, which can differ from one company to the next.

Monitoring Inventory Levels

Inventory control is extremely important, you always want to make sure you’re monitoring your inventory levels. Once again, relying on your inventory management software is essential.

Monitoring your SKUs will help you identify what products are selling the best, which products are selling slower and which products are not selling at all so you can make the proper adjustments to inventory.

This is why real-time inventory software plays a big role.

We Can Help You Better Manage Your Inventory

Here at Thill Inc., we have over 50+ years of fulfillment, shipping, and inventory experience.

Our state-of-the-art technology makes tracking and forecasting inventory simple.

If you want to learn more, feel free to reach out and get in touch with our team. You can reach us anytime by calling 1-920-967-9201.

Need a Fulfillment Partner?

More Posts

Ecommerce Logistics

What Is Ecommerce Logistics?

Ecommerce logistics, occasionally known as elogistics, encompasses the procedures employed by ecommerce businesses to fulfill orders and deliver products to

Understanding Average Order Value

What Is Average Order Value?

In the dynamic landscape of e-commerce, businesses continually strive to enhance their understanding of customer behavior and optimize their strategies

Increase conversions and sales!

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.