Direct-to-consumer (DTC) brands can feel a lot of pressure to launch new products into the ecommerce space. And while this pressure is largely unspoken, it’s a reminder that new products are what attract customers and keep brands competitive.
What’s often overlooked, however, is that new products come with their own set of challenges. Perhaps mainly, forecasting demand for something with no historical sales data.
Although this forecasting process can get complicated, we have tons of tips and best practices on how to increase demand for every new product you release.
Let’s just cut to the chase: The less accurate your forecasts are, the more problems your company will likely run into.
And when you don’t accurately forecast new product demand, it can lead to stockouts or overstocking, profit losses, and weakened relationships with your suppliers.
Forecasting errors can lead to an under- or oversupply of new product inventory. Both of which have consequences.
An undersupply of products puts you on the fast track toward a stockout. And if you end up going out of stock, it can chip away at customers’ confidence in your brand (and maybe even send them to your competitor’s site).
Meanwhile, oversupplying is no better. It increases your carrying costs since you have to keep paying for the excess inventory sitting at your warehouse.
There lies the risk “if companies buy too much inventory, and they can’t sell all of it.” As Modern Retail explains, “at that point, they’ve already sunk money into procuring inventory.”
This makes overstocking exceptionally complicated for budding DTC brands, who have to make every dollar stretch as far as they possibly can.
There’s a chance you remember Nike’s big mistake in 2001. That year, Nike installed a demand planning software without properly testing the system — which is a pretty necessary step for a global retailer with such complex operations.
Unfortunately, that demand planning software didn’t work — at least, not for Nike. And the famous athletic-wear retailer ended up using significantly inaccurate demand forecasts.
They wound up overstocking on low-selling shoes while understocking their bestselling Air Jordans.
The result? Nike lost a staggering $100M worth of sales.
Although this is an extreme example, it illustrates some important points:
The better you get to know your suppliers, the more likely you will benefit from their preferential pricing and dedicated service. And yet, inaccurate forecasts for new SKUs can really weaken your vendor relationships.
When your forecasts miss the mark, they cause a ripple effect throughout your supply chain and reflect poorly on your brand’s credibility.
Think about it — if suppliers don’t trust your supply chain management, they may start relying on their own predictions or pull back on offering you a discount.
Unfortunately, the supplier’s predictions could end up hurting your brand.
If they forecast conservatively (likely), but demand picks up (maybe even matching your original projection), it’ll be really tough to catch up on the fulfillment side of things.
Plus, if the supplier revokes their discount, purchasing stock will become all that much more costly. This increase in price will certainly impact your brand — especially considering that buying inventory is already the “most expensive part” of your growth.
Maybe you’ve never attempted demand forecasting, or you just want a fresh approach to the whole process. Either way, there are plenty of forecasting techniques you can choose from.
The following are the best methods to forecast customer demand for new products.
Using trend projection, retailers can anticipate patterns in demand and base their forecasts on these ebbs and flows. In other words, this method uses past sales metrics to predict future demand.
While trend projection is pretty straightforward, you’ll still need to adjust your calculations for seasonality or other anomalies in demand.
For example, you might’ve had a surge in sales because your product was featured on a television show, or you went viral on TikTok.
As great as this surge was, it’s not something you can bank on happening again soon — making it an outlier within your trend projections.
It’s important to note these unusual factors in your historical data, so don’t let them influence your forecasts for new products.
A quantitative analysis uses past data to forecast the future by emphasizing objective, mathematical forecasting models. So, the more numerical data your company collects, the more accurate your quantitative predictions will turn out.
At its core, the quantitative method is a historical analysis of similar products.
By tracking data on these products and looking for patterns in demand, you might notice sales dip in the winter and then surge in the spring. This knowledge can inform your forecasting by taking into account when your new product will be released.
While trend projection falls under the umbrella of quantitative analysis, it’s just one aspect of this whole strategy. Other examples of quantitative forecasting include seasonal index, naive method, straight-line method, moving average method, and more.
Qualitative analysis exists in contrast to the quantitative method. A qualitative approach is used when you don’t have any historical data to go off of.
Without this data, some companies seek the help of their customers and in-house sales teams. Both groups have a good gauge on actual demand — even though one is doing the buying and one is doing the selling.
Many times, qualitative analysis culminates in the use of customer and sales team surveys.
Customer surveys are a type of market research that collects information directly from your customers. These surveys can provide valuable insights you can’t get from static sales data.
On top of assessing satisfaction with your brand, these surveys can tell you whether customers are interested in your new product.
If interest is low, you may need to lower your forecasts (or drum up some excitement around the product launch). If interest is high, you’ll need to order enough inventory to avoid a stockout.
Mega beauty brand Glossier does a great job of collecting customer feedback. Their surveys not only gather ideas on the new products customers would like to buy next, but they shed light on how existing products can be improved.
Sales teams are the eyes and ears of your company. They’re the ones who interact with your customers on a regular basis.
So, surveying your sales staff might involve asking about the buyer journey. For example, how long does this journey usually last? Or, what drives someone to add items to their cart?
With this deeper understanding of the customer journey, you can anticipate how your new product will sell (and make forecasts to reflect that).
It’s important to note that not every DTC brand has a sales team they can survey. Only DTC sellers who also have B2B or wholesale business — and thus, a sales team — will be able to take advantage of this method.
The Delphi method is where you engage with expert opinions to guide your forecasting. If you go this route, you’ll send out several rounds of questionnaires to a panel of inventory experts.
For a sustainable footwear brand like Allbirds, this might mean sending questionnaires to advisors from Everlane, Adidas, or Cariuma (who all work within a similar industry).
After each round of questions, a 3rd-party facilitator gathers the answers, creates a summary report, and then distributes this report to each expert. The experts then read through that report and either agree or disagree with one another’s answers.
The Delphi method is complete once the whole group has reached a consensus.
While this process is lengthy, it lets you draw insights from very knowledgeable people. And because there’s no in-person discussion, your experts can be located anywhere in the world.
Still, the questionnaires should be filled out within the same time frame — so keep that in mind if you opt for this method.
The econometric method looks at the relationship between different economic factors.
This technique examines your sales data along with outside forces that can affect customer demand (like competing brands and fluctuating preferences).
For instance, early on in the pandemic, most people avoided the gym in favor of working out at home. In turn, the home fitness industry saw an incredible spike in sales.
Peloton is a terrific example of this. By September 2020, Peloton’s sales had increased by a whopping 172%. In that way, Peloton benefitted from the outside forces of the pandemic.
Econometric forecasting takes exactly these sorts of situations into account. It weighs your previous sales against economic variables and predicts the future demand for your products.
Time series analysis is when you make forecasts based on time-stamped data. This data is collected at consistent intervals within a set time period (rather than being recorded randomly).
When your team analyzes the data, it provides the “why” behind certain inventory outcomes — like why the sales of your core product took a dive in Q2.
With that said, times series analysis will be less accurate if you have a lot of variables within your data (again, like those caused by the pandemic). If you go with this method, just be sure to keep an eye on any variables so they don’t wind up affecting your forecasts.
If you’ve got a new product launch coming up, you might need help with forecasting demand. Below are 5 actionable steps to forecast demand for new products.
As you’re forecasting demand for a new SKU, you’ll need to assess the capacity of your suppliers and manufacturers.
Make sure you and your supplier are on the same page about order lead time and delivery. Similarly, confirm that your manufacturer has all the raw materials needed for production.
(For added peace of mind, check that your manufacturer is up to date on product packaging, too).
These conversations will ensure your suppliers and manufacturers know what they’re each responsible for, and it’ll verify whether they can meet your projected demand in full.
You can’t achieve forecast accuracy unless you have a clear plan for product distribution — such as the process of delivering products from the manufacturer to the purchasing customer.
Your new product will probably follow the distribution patterns of similar products in the same category. Still, it’s in your best interest to confirm all this with your distribution partners.
By discussing specific routes and methods for shipment, you can guarantee your distributors are ready for the volume of product you’ll have.
In other words, you should finalize your transportation plan in light of your demand predictions.
If your distributors are confident they can deliver everything on time and on budget, feel free to turn your forecasts into final purchasing decisions.
Product cannibalization happens when a new product displaces an existing one. That is, your new release affects the product sales of an older item (and winds up hurting your cash flow).
Unless you’re introducing a totally new category, new products will likely impact your existing items to some degree.
Still, there are ways to brace yourself for product cannibalization and even lessen its impact.
One way to do this is by communicating the upgrades that come with your newest SKU and how it differs from your older products. That way, customers will see the value in buying both items.
For instance, your new chocolate chips might have a superior taste thanks to their use of organic ingredients. But the old product is better for baking.
This step is potentially the most challenging, seeing as new products don’t have a sales history you can base your forecasts around.
Fortunately, there is a silver lining here. You can reference a few different data sources to estimate what your sales will look like right out of the gate. These sources might include:
By evaluating each data source, you’ll have an easier time forecasting new product demand for the first few weeks of sales. From there, you can make additional adjustments to your forecasts based on how customers responded to your initial launch.
The first few weeks after a product rollout are critical to your success — which is why you need to pay attention to your sales and customer feedback during this time.
One of the best things you can do is interact with customers directly by sending a product review email or engaging with them on social media. This will give you a peek at how they like the product or how they’re using it.
By monitoring sales and evaluating this feedback, you can resolve issues that pop up early on and make more informed forecasts moving forward.
Apple is a great example of prioritizing customer feedback. Even though Apple continually releases new versions of their watches, phones, and more, customers keep buying their products in droves — despite having the previous model.
That’s because the company listens to what its customers have to say via customer feedback surveys. With these surveys, Apple can gauge satisfaction and how likely people are to purchase again — and then use these insights to drive product development and forecasting.
Estimating new product demand is something of an art form. But if you abide by these best practices, you can wildly improve your forecasting efforts for every new release.
As you’re making estimates for new product demand, you’ll need to consider the availability of your raw materials suppliers.
If there’s a shortage of raw materials (or they aren’t ready to be manufactured), you might need to pivot the plan. For instance, you could enable preorders of your new product.
This way, you can guarantee a set amount of sales and revenue during your launch and offer customers the satisfaction of completing their purchase.
For your new product to be successful, you’ll need a solid plan for production and distribution.
To determine the initial production quantity, you can follow the same guidelines listed in our “how-to” section above (like referencing historical data for existing products in the same category).
This data will give you an idea of how much new product you might sell, so you can make better estimates for your ensuing production schedules.
Likewise, be sure you have a good distribution plan. Discuss routes and shipping methods with your distribution partners, and then finalize these plans (according to your demand predictions) before you begin manufacturing.
If your new product has periods of high and low demand throughout the year, you’ll need to consider these seasonal trends (like more demand for your products around the holidays) in your forecasting estimates.
Because your new product doesn’t have any historical data just yet, you can observe the seasonal patterns of other items in the same product category. This will improve your forecasting estimates until your new product has viable data of its own.
When estimating demand for a new product, be sure to factor in the effects of your advertising and promotional tactics.
For instance, say you’ve marketed to a wide demographic (or reached out on multiple digital platforms). Chances are that your hard work will pay off.
This means anticipating the positive impact of your marketing efforts and then increasing your forecasts accordingly so you can fulfill all the sales coming your way.
If your DTC brand is launching its first product ever (congrats!), there’s a chance you could run into a stockout. Even with a trusted forecasting method, it’s difficult to estimate demand when you don’t have any foundational data to lean on.
On the surface, selling out might not seem like such a bad thing. But when you dig deeper, you’ll realize that a stockout limits your out-the-gate growth because you can’t collect more data on actual demand (or revenue).
Without this data, you could face even more stockouts since you don’t have the information needed to make accurate sales forecasts. But that’s where selling on backorder comes to the rescue.
Backorders help you gather the data you need to forecast a lot faster (and more accurately). Plus, backorders let you keep selling even when your new products aren’t in stock. And they maintain conversion rates comparable to selling that same product when it’s in stock.
The best operations software comes with forecasting features that integrate previous sales data, purchasing trends, seasonal analysis, and other inventory analytics into one convenient platform.
With this information at your fingertips, you’ll have incredible inventory visibility into how much your new product might sell. This way, you can make more accurate forecasts right from the start.
In addition, operations software is a lot easier to navigate than static spreadsheets. With proficient software, you can quickly pivot whenever new data is introduced. This flexibility leaves less room for error, whereas manual spreadsheets are more susceptible to forecasting mistakes.
If you need help with forecasting, Cogsy’s got you covered. Cogsy makes forecasting a breeze with optimized (new) product planning, a proficient operations platform, and trusty backorder support.
The latest and greatest development from Cogsy is our new product planning feature. With this feature, you can select products with similar characteristics. Then, let Cogsy handle the calculations and figure out how much new product you’ll need.
What’s more, Cogsy will factor in lead time, so you place your order at exactly the right time. And if that wasn’t enough, you can even plan product launches for a single SKU or SKUs with multiple variants (like items you offer in different colors or sizes). It doesn’t get much better than that!
Cogsy’s operations platform allows you to monitor all your inventory data in one convenient place. Yep, you read that right — Cogsy stores your real-time and historical data within the same inventory management platform, so it’s accessible whenever you need it.
This clarity around your sales data plays a big part in new product forecasting. You can check on previous forecasts for similar products and then use this information to determine the new product volume you expect to sell.
While Cogsy is designed to help you forecast with confidence, there might be times when demand spikes and your actual sales overtake your previous projections. The good news is that Cogsy has your back in these situations, too.
With it, you can sell on backorder. That way, you can keep converting the demand you’ve generated into revenue while waiting for your stock to replenish. With backordering, customers can still buy your product instead of looking elsewhere to make a purchase.
But the best part of backordering? (I hinted at this earlier.) It converts customers at almost the same rate as selling the same product in stock, so you’re not missing out on revenue opportunities.
Want your next launch to be your best launch yet? Try Cogsy free for 14 days.
The best method to forecast demand is trend projection. Essentially, trends are the changes in product demand over a set time period. So, using trend projection, retailers can anticipate these patterns in that demand and base their forecasts on the ebbs and flows.
To forecast sales of a new product, you can reference a few different data sources. This might include referencing historical data for existing products in the same category as your new product or checking on current data for comparable products sold by your competitors.
To create demand for a new product, you can hone in on your promotional efforts (like email marketing or social media campaigns). Additionally, you can ensure your new product has unique features that set it apart and demonstrate value to your customers.