Articles written based on real expert opinions Forecasting knowledge from practitioners who work with it daily
Each piece here comes from someone who has sat with the numbers, made the calls, and learned what the textbooks tend to leave out.
Data Analysis
What Business Forecasting Actually Gets Right (And How Often)
Accuracy rates, error ranges, and what the research actually shows
A data-driven look at forecasting accuracy rates across industries, written for readers who are tired of vague promises and want to understand what the numbers really say.
Expert Analysis
Statistical Models vs. Expert Judgment: Which One Loses Less Often
Evidence from 50 years of forecasting research, without the sales pitch
An analytical comparison of model-based and judgment-based forecasting, with specific evidence on where each approach breaks down and why combining them is more complicated than it sounds.
Statistical
Patterns
5 Data Patterns That Explain Most Forecasting Failures
Identifiable data problems behind the majority of significant forecast errors
A structured look at the statistical reasons business forecasts go wrong, aimed at analysts and planners who want to identify failure points before they show up in the actuals.
Scenario
Analysis
Scenario Planning Has a Mixed Evidence Base - Here Is What It Shows
What 44 organisations and a Rand review actually found about scenario planning outcomes
A measured look at what the research actually says about scenario planning as a forecasting tool, written for readers who are sceptical of both its advocates and its critics.
Machine Learning
Machine Learning in Business Forecasting: An Honest Assessment of the Numbers
M4 competition results and independent studies on ML forecasting accuracy
For analysts considering ML-based forecasting tools, this piece examines what the M-competition data and independent studies actually show about accuracy gains versus implementation costs.
Banking Access
Long Queues at Irish Bank Branches: What the Lines Actually Tell You
A resource guide for anyone trying to understand why Irish bank queues have become a fixture of daily life
Irish bank queues have become a daily ritual since early morning. Here is what most people miss about why this keeps happening and what it means for everyday banking.
What these pieces are actually about
Forecasting is one of those disciplines where theory and practice diverge sharply. A model that looks clean on paper can fall apart the moment it meets a real sales cycle or an unexpected supply delay.
The pieces collected here come from people who have worked through those gaps. A financial controller who spent three years rebuilding quarterly planning after an acquisition. A logistics analyst who learned which demand signals actually mattered after a year of chasing the wrong ones. Their accounts are specific because specifics are what make forecasting knowledge transferable.
83% of the contributors here have held operational roles - not advisory ones. That distinction shapes how they write about the subject.
Forecasting disciplines covered across this collection
Each area below represents a strand of forecasting that practitioners have written about from direct experience.
Demand planning
Reading signals in order data, seasonality patterns, and what happens when those signals contradict each other.
Scenario modelling and sensitivity
Building forecasts that account for multiple futures rather than committing to a single number - and communicating that uncertainty to decision-makers who prefer certainty.
Financial projection
Revenue and cost forecasting across planning cycles, including how assumptions get stress-tested before they reach the board.
Operational and capacity planning
Translating sales forecasts into staffing, inventory, and production decisions - where the cost of a wrong estimate is immediate.
Forecast error and accountability
How organisations measure forecast accuracy, where they tend to assign blame, and what a more useful post-mortem looks like.
Sharing your own forecasting experience
Practitioners who have worked through a specific forecasting problem - a methodology that failed, a tool that performed differently than expected, a planning cycle that needed rebuilding - are the kind of contributors this collection is built around.
Submissions do not need to be polished or comprehensive. A single well-documented experience is more useful than a broad overview.
- Specific industry or operational context
- A concrete problem or decision point
- What was tried and what the outcome was
- What you would do differently