Please enable JavaScript.
Coggle requires JavaScript to display documents.
Forecasting - Coggle Diagram
Forecasting
Forecast links to other functional areas
Forecasting feeds into and influences many other functional areas within an organization.
Linkages
Unidirectional
Information flows in one direction, from the forecast to the functional area.
Ex: Sales forecasts drive production schedules, but production data does not influence the sales forecast.
Bidirectional
Information flows in both directions, allowing for feedback and adjustments.
Ex: Sales forecasts influence production schedules, and production data (like capacity constraints or inventory levels) feeds back into the sales forecast to refine it.
This approach creates a more dynamic and responsive planning process, improving accuracy and adaptability.
The links reflect the varied uses to which a forecast can be applied:
Revenue planning, resource allocation, project prioritization, partnering decisions, compensation plans, lobbying efforts, etc.
Forecasts require differing data requirement depending on its use and end-user
Forecasts may be done at a global, regional, country, or business unit level.
Each forecast has differing levels of data required, both at the input and output level. These forecasts need different structural constructs.
In the
sales and marketing areas
, forecasts are used to drive resource allocation decisions.
Forecasts used for research and development portfolio planning also require different constructs and outputs than the other functional areas.
In these types of forecasts, we need to adjust for risk and uncertainty in the forecast.
Finance function:
various financial ratios such as net present value, return on investment, breakeven dates, and so forthーneed to be calculated from the forecast revenues.
Although this flow, in theory, is
unidirectional,
the financial requirements of an organization (e.g. earnings per share) sometimes feed back into the forecast to drive a change in revenue projection.
The application of options valuation techniques are starting to find a place in pharmaceutical product forecasting.
Long used in assessing the equities markets
The many functional areas that interact with the forecast create tremendous pressure on the model construct, the analytics, the inputs and the outputs of the forecast.
Every function develops its own model, based on a different set of assumptions with little or no consistency across the board.
The key challenge to forecasting is to create a process where the needs of function can be met without compromising the integrity of the forecast approach.
Process of Forecasting
Defining the need for the forecast
Selecting the appropriate method
Analyzing the results
Presenting the forecast to the end-user
Pharmaceutical Forecasting
Forecasting has become a keystone in future planning functions such as strategic planning, business development, and portfolio optimization.
involves predicting future demand for pharmaceutical products.
Key aspects
New Product Forecasting
Before launching a new product, companies forecast its potential market performance. This involves analyzing patient populations, disease prevalence, and treatment rates to estimate the number of potential patients
In-market Product Forecasting
For products already on the market, companies use historical sales data and patient trends to predict future demand. This helps in adjusting production and marketing strategies
Methods
Various methods exists: quantitative and qualitative
Patient-based vs Prescription-based
Machine Learning and Statistical Methods
Challenges
complex forecasting due to regulatory changes, market competition, and innovation
Time horizons for forecast
A second key challenge in forecasting arises from the varied time horizons associated with a forecast.
Forecasts range from very long-term time horizons (e.g. a ten-year forecast for a product that will launch three years from now) to short-term focus (e.g. quarterly forecasts that will be used to set incentive compensation goals for the sales organization).
If a forecast is incorrect, it commits resources for the organization. It can spell out whether a company will grow or suffer a loss.
Analytic Tools in Forecasting
Risk and uncertainty
Simulations
Simulation methods are appropriate for most input variables that range in the forecast.
There are, however, some input variables that create strong dependencies on the other variables in the forecast.
When strong dependency exists simulation methods are not appropriate; in this case, the forecaster must use discrete scenarios.
a technique used to model and analyze the impact of uncertainty and variability in forecasting and decision-making
Scenarios
Decision analysis
From the forecaster’s perspective, the difference between risk and uncertainty is related to if and when the uncertainty is resolved.
Risk
is resolved throughout the planning process as a product progresses in its development;
uncertainty
remains even after the product launches to the market.
Sensitivity
Sensitivity analysis
Tornado diagram
Waterfall diagram
Productivity
Net present value
Return on investment
Break-even analysis
Risk v Return
Productivity multiple
A measure that is gaining increased use
Defined as the expected commercial value divided by the expected development cost; then weighted by the probability of launch.
A valid measure to apply in cross-product or cross-project comparisons.
Productivity measures
Defined as a set of financial outputs and ratios.
The final set of forecast analytics focus on this.
Standard financial measures
Net present value (NPV)
Return on investment
Break-even analysis
Risk versus return
Productivity multiple
Job of Forecaster
To predict the future accurately.
To create stories that paint an holistic picture of the future, not simply spreadsheets
Forecasting Methods
qualitative
quantitative
In market product forecasting
Trending Algorithms
Selecting the underlying data sets to trend
Trending Revenue
New Product Forecasting
New Product Forecasting Algorithm
Modeling the Market
Modeling the Epidemiology of the Disease
Prevalence-based model
Incidence-based epidemiology model
Patient Segmentation
Often this is not a valid assumption and the forecaster needs to divide the patient pool into relevant segments such as demographics, disease severity, treatment history, and other characteristics.
This segmentation helps in understanding different market dynamics and tailoring strategies to meet the specific needs of each segment. By doing so, companies can more accurately predict demand and optimize their marketing and production efforts
Determines the potential market size for new products. It is a measure of potential and represents the theoretical maximum use for a product.
Patient-based Model
Focus: Centers on the patient journey and disease progression.
Method: Uses epidemiological data to estimate the number of potential patients. This involves calculating the prevalence and incidence of the disease, diagnosis rates, and treatment rates.
Application: Ideal for new products or complex treatment patterns, such as oncology drugs. It provides a detailed understanding of how patients move through the healthcare system and how they are treated.
Advantages: Offers a deep, causal relationship between patient behavior and market outcomes. It helps in understanding the patient journey and decision-making process.
Challenges: Requires extensive data collection and research, which can be difficult to obtain and update frequentl
Prescription-based Approach
Focus: Relies on historical sales data and real-world evidence.
Method: Uses past prescription data to forecast future demand. This approach trends past performance into the future, assuming that historical patterns will continue.
Application: Suitable for products already on the market with established sales data. It provides timely insights into market dynamics and helps in adjusting strategies based on current trends.
Advantages: Frequently updated and provides a granular view of sales, making it responsive to changing market conditions.
Challenges: Offers a narrow view focused on current performance, without delving into the underlying causes of trends
components
Model the market
To forecast the product
To convert patients on product into revenue
forecasting patient adherence to the treatment and duration of use; consider the impact of pricing and reimbursement policies on the product's market performance
Forecasting the Product
Competitive mix
Key Aspects
Forecasting patient share
Forecasting the share of patients who will receive the product versus the competitor’s is the heart of the forecast model. As with many choices in forecasting, the modeler has the opportunity to balance simplicity and complexity.
User-entry share methodology
The simplest and most user-friendly method of share projection is user-entry.
Process: The forecaster enters the projected share for all products in the competitive grid. Shares may be entered on a monthly, quarterly, or annual basis or entered on any time basis aligned with the model design. The method is extremely simple - no complex calculation routines are needed and the dynamics of product uptake (how quickly the product gains patients and therefore share in the market) is embedded into the shares entered in by the user
Challenges: the challenge with this approach is that the logic and thought process behind the share projections may not be transparent or well-documented, making it difficult to judge the assumptions used
Estimate the share of patients who will use the new product compared to competitors. This involves understanding the dynamics of product uptake and how quickly the new product can gain market share
Peak share and time to peak share methodology
The user enters in two parameters - the peak share of the product being forecast and the amount of time required to attain this forecast peak share.
This suffers from the same weakness in terms of transparency and defensibility as the simple user-share method.
If the logic trail used by the user is not explicit or documented it becomes difficult – if not impossible – to judge the merits of the shared assumption.
The second set of share methodologies is referred to as peak share and time to peak.
Attribute Methodologies
Market Definition
ncludes all products against which the product competes and may include both the inter-class and intra-class competition.
analyzing the landscape of existing and potential competitors to understand how they might impact the market performance of the new product
Calculation of potential patients who are treated with the product being forecast. This is referred to as patient share, product share, or market share depending upon the definition of the share calculation.
Converting Patients to Revenue
Forecasting compliance and persistence
The ubiquitous effect of price and reimbursement