Forecasting new products or markets can be challenging, especially when you have limited or no historical data to rely on. How can you make informed decisions and reduce uncertainty in such situations? In this article, we will explore some methods and techniques that can help you forecast with confidence and creativity.
One way to forecast new products or markets is to use analogies and scenarios. Analogies are comparisons between your new product or market and an existing one that has similar characteristics, features, or customers. Scenarios are stories that describe how your new product or market will perform under different conditions, such as changes in demand, competition, technology, or regulation. By using analogies and scenarios, you can leverage existing knowledge and data, as well as explore various possibilities and risks.
Forecasting requires a commitment. the decision makers must commit the materials to meet initial plan sales other wise why develop the product. Lead times must be part of the plan.
I agree with Mark, below, as forecasting is a business wide commitment. In my experience it is not an absolute requirement for a business to forecast, for example an entrepreneur will go off their knowledge. However, if a business does forecast it must be done well or the output will be compromised and resulting decisions poor.
Here are some steps you can take to use analogies and scenarios effectively: -Identify similar markets or products: Look for markets or products that are similar in some way to the one you are forecasting. -Create scenarios: Using the information you have gathered, create a range of scenarios for how the new market or product might perform. -Evaluate the scenarios: Evaluate the scenarios you have created based on their likelihood and potential impact. L -Refine your forecast: Use the insights gained from your analysis to refine your forecast for the new market or product. By using analogies and scenarios in this way, you can build a more informed forecast when you have limited or no historical data.
When it comes to creating a forecast for a new item using item-item collaborative filtering, the approach involves finding the most similar items to the new item based on their attributes, such as genre, features, and price. The system then uses the ratings and preferences of users who have rated these similar items to make a prediction of how users will rate the new item. To accomplish this, the system calculates a weighted average of the ratings for the most similar items, where the weights correspond to the similarities between the items. This method can provide valuable insights into how users are likely to respond to a new item, and can help businesses make informed decisions on product development and marketing strategies.
You have to get creative when building a forecast for new products or markets, especially since there is no historical data to leverage. Questions to ask & explore - are there similar products or markets, can you get insight from industry experts or your ecosystem of suppliers & consumers. As you build the forecast model - ensure that you have a comprehensive list of assumptions and have multiple modelling approaches. Continue to refine as your data gets better.
Another way to forecast new products or markets is to apply judgmental methods. Judgmental methods are based on the opinions, insights, and expertise of people who have relevant knowledge or experience, such as customers, suppliers, industry experts, or internal stakeholders. Judgmental methods can include surveys, interviews, focus groups, Delphi panels, or brainstorming sessions. By applying judgmental methods, you can gather qualitative and quantitative information, as well as uncover hidden needs, preferences, or trends.
There are 3 elements to gathering a judgemental forecast. 1 - Internal resources, such as operations and supply chain tend to be more fact based. 2 - External elements such as demand planning and pricing are more subjective. 3 - Analysis and judgement brings supply and demand together in an agreed and published plan.
A third way to forecast new products or markets is to build simulation models. Simulation models are mathematical representations of how your new product or market will behave and interact with other variables, such as costs, revenues, production, distribution, or customer behavior. Simulation models can include spreadsheet models, system dynamics models, agent-based models, or Monte Carlo models. By building simulation models, you can test different assumptions and scenarios, as well as quantify the uncertainty and variability of your forecasts.
Simulation models are great and spreadsheets are a highly flexible tool that many organisations use. My simple recommendation is to keep simulations as simple as possible and if using a spreadsheet separate it into 3 parts, inputs or levers, data, and output. This way to “de-risk” a forecast create a copy of the simulation with sales growth “lever” dialled down and see the impact with fixed overheads on net margins.
A fourth way to forecast new products or markets is to combine multiple methods. Combining multiple methods can help you improve the accuracy, reliability, and validity of your forecasts, as well as reduce the bias and error that may arise from using a single method. Combining multiple methods can include averaging, weighting, adjusting, or integrating the results of different methods, such as analogies, scenarios, judgmental methods, or simulation models. By combining multiple methods, you can create a more comprehensive and robust forecast that reflects the complexity and dynamism of your new product or market.
As a senior executive in data analytics, I fully understand how challenging generating forecasts with limited data can be. In these situations, we use both qualitative & comparative analysis for rapid dataset comparison. A mix of these methods is crucial for high fidelity forecasts. The value of spotting patterns cannot be overstated. It's the exact answer to the problem of insufficient data. If you have access, Artificial intelligence and mixed methods analysis can allow you to see patterns and connections that might otherwise go unnoticed. This is the only way to accurately build predictions without historical context.
A fifth way to forecast new products or markets is to validate and update your forecasts. Validating and updating your forecasts can help you monitor the performance and progress of your new product or market, as well as identify any gaps, deviations, or changes that may affect your forecasts. Validating and updating your forecasts can include comparing your forecasts with actual outcomes, collecting feedback from customers and stakeholders, conducting post-launch analysis, or revising your assumptions and scenarios. By validating and updating your forecasts, you can learn from your experience and improve your forecasting skills and processes.
An important point here is to ensure that the forecast is aligned to, and as detailed as the actual data that can be gathered. Through alignment the two elements can be compared and this enables for Variance Analysis. For example if a high level sales forecast is provided and actual data is detailed by customer and the target is missed how would the gap be understood and addressed.
A sixth way to forecast new products or markets is to seek external support. Seeking external support can help you access additional resources, expertise, or perspectives that can enhance your forecasting capabilities and outcomes. Seeking external support can include hiring consultants, analysts, or researchers, joining industry associations or networks, attending workshops or seminars, or subscribing to publications or databases. By seeking external support, you can benefit from the best practices, tools, and methods that are available and relevant for your new product or market.
Para pronosticar nuevos productos o mercados con datos históricos limitados o nulos, se pueden seguir recomendaciones como investigar el mercado, utilizar fuentes de información alternativas, utilizar métodos de pronóstico cualitativos o cuantitativos, realizar pruebas de mercado y actualizar y ajustar los pronósticos.
In my experience during the early years of the telecom market explosion 80s, 90s… there were several customers that were on the edge of innovation. Working hand-in-hand with these innovators enabled our company to be at the forefront of the technological innovation curve. Each market forecast of the early 80s proved to be under-called due to the explosive growth as new uses and increased bandwidth via innovation was introduced. In the late 90s the technology /innovation was ahead of demand so some of the early mover advantage was lost as it took the users (new apps adoption) a while to catch on. As stated, continuing to reassess where you are and your environment is essential.