As Agile teams strive towards predictability in their planning, Monte Carlo simulation is becoming the most effective forecasting mechanism for Agile teams today. Previously Agile teams were relying on intuition or a set of static estimates or happy assumptions which resulted in mismanagement of time and disappointment at delivery.
Now using historical data to forecast delivery timelines based on probabilities provides confidence in forecasts, creates transparency in communication, and structures the management of uncertainty using data.
Once a perceived complex mathematical problem, Monte Carlo simulation is now an amazing tool for every size of Agile team to deliver accurate forecasting capabilities to all sizes of their teams.
Why Traditional Agile Forecasting Often Fails to Deliver Accuracy
Agile methodologies provide teams with the ability to have increased flexibility, yet one of the major challenges when deploying Business Agility is the marked difficulty in estimating how long it will take to do the work.
Estimating completion dates for projects with traditional techniques, like story point estimation, planning poker estimation, expert opinion, and velocity-based forecasting, has not typically proven to be an effective method of estimating completion dates because they rely heavily on assumptions that aren't usually true in an actual working environment.
Also, an optimistically biased human estimate is frequently lower than a true data-driven estimate.
Because of the continual problems associated with missed deadlines, there is a continual sense of frustration and let-down from both the teams and the stakeholders as the timescale for delivery is becoming increasingly unpredictable and could lead to missed deadlines.
Monte Carlo simulations can help to alleviate these ongoing problems by providing statistically sound predictions based on the experience of actually delivering projects from historical project data sources. Monte Carlo simulations also use historical delivery data as their basis rather than relying on assumption or optimistic estimates as a basis for forecast.
What Is Monte Carlo Simulation in Agile? A Beginner-Friendly Definition
In a Monte Carlo simulation, many statistically applied random experiments are calculated based on each team's historical performance (throughput). This provides a broad estimated range of completion dates rather than providing one absolute date that is typically unreliable.
The benefit is to make it easier for teams to say:
“Four sprints until completion.”
To be more accurate, teams now say:
“There is an 85% chance that we will complete in four sprints and a 95% chance of completing in five.”
Through probabilistic estimation, Monte Carlo simulations offer a way to estimate outcomes with a greater degree of accuracy than with simple date estimation, thus allowing teams to make better-informed decisions.
How Monte Carlo Simulation Works - Explained Step-by-Step
Despite its apparent mathematical nature, the workflow of this approach is significantly less complicated than it seems.
1. Gather Historical Throughput
The throughput refers to the total number of work items completed during each sprint or time interval, thus serving as the foundation of the simulation.
2. Execute Thousands of Random Simulations
While each individual simulation randomly draws from your historical throughput dataset to form a hypothetical future timeline, executing thousands will result in the formation of a distribution of probabilities for many different outcomes.
3. Create a Probabilistic Forecast
The generated output includes:
High-confidence timeframes (i.e., 95% confidence level)
Moderate-confidence timeframes (i.e., 70% to 80% confidence levels)
Low-end optimistic possibilities
Teams can select the timeframe they wish to use considering their level of risk tolerance and/or business needs.
Case Study: How an Agile Product Team Improved Delivery Confidence Using Monte Carlo Simulation
The Challenge
A mid-sized B2B SaaS product team (6 engineers, 1 product manager, 1 QA) struggled with unreliable sprint and release forecasts. Although the team followed Agile best practices and tracked velocity, delivery timelines were frequently missed.
Common issues included:
High variation in sprint velocity
Unplanned production fixes disrupting planned work
Stakeholder frustration due to changing delivery dates
Lack of confidence during quarterly planning
Leadership needed a way to forecast delivery that reflected real execution patterns rather than optimistic estimates.
The Approach
The team adopted Monte Carlo simulation using their historical throughput data from Jira. Instead of relying on average velocity, they analysed completed work items across the last 12 sprints and ran thousands of simulations to forecast future delivery.
Using probability-based forecasting, the team shifted from single-date commitments to confidence-based timelines.
For example, instead of committing to “delivery in 6 sprints,” they communicated:
70% confidence: delivery in 6 sprints
85% confidence: delivery in 7 sprints
95% confidence: delivery in 8 sprints
This allowed leadership to choose timelines based on business risk tolerance.
The Results
Within three months, the team observed measurable improvements:
Forecast accuracy improved by 30% compared to velocity-based planning
Stakeholder trust increased, as delivery expectations were set with transparent confidence levels
Planning discussions became faster and less emotional, shifting from opinion-based debates to data-backed decisions
Sprint commitments became more realistic, reducing carryover work and burnout
Most importantly, missed deadlines were no longer seen as failures—but as understood outcomes within a known probability range.
Why Agile Monte Carlo Forecasting Is So Powerful
1. It Utilizes Actual Data & Not Assumptions
Forecasting is based upon the proven patterns of delivery exhibited by the team over an extended period of time.
2. It Visualizes Uncertainty
The teams will not see just one unreliable forecast but will see a range of possibilities that help to illustrate variation.
3. It Enhances Communication with Stakeholders
Probability-based forecasting eliminates many of the subjective decisions and therefore reduces friction during conversations with stakeholders when compared to non-probability- based forecasting methods.
4. It Facilitates Improved Sprint and Release Planning
The team has a high level of assurance that they are making commitments based upon data and therefore has an increased likelihood of making realistic plans.
5. It Increases in Accuracy Over Time
As more data regarding throughput is collected and made available, the forecasts generated will better represent what is likely to occur.
6. It Takes Almost No Time or Effort
There is no requirement for extensive analytical background to use this; a simulation of all work items is run based on the data that has been collected thus far.
Why Monte Carlo Fits Perfectly into Modern Agile Delivery
To adapt to constant changes and fluctuations in requirements and dependencies, short- term static forecasts do not suffice. Instead, while we have seen fluctuating methodologies from year to year, they ultimately provide the same results.
A Monte Carlo Simulation provides dynamic data-driven information at every sprint as the data points must evolve with the iterative process and accommodate changes over time.
This means that Monte Carlo simulations are compatible with Agile development
methodologies; they allow teams to identify when change is a part of a final product approach but still provide agility for teams through constant innovation.
How Baseliner.ai Brings Monte Carlo Forecasting into Everyday Agile Workflows
Baseliner.ai provides Agile teams with total visibility into their Monte Carlo simulation process in automation mode and provides results based on the Agile cycle that are actionable. Through Baseliner, teams will no longer waste their time collecting work
completion data manually and attempting to build a proper forecasting spreadsheet for their work. Baseliner.ai creates continuous, real-time forecasting/planning updates based on their actual performance.
1. Automated Throughput Data Imports: Baseliner pulls throughput data from various tools including Jira, providing teams with the highest quality and most accurate data for use in their Monte Carlo simulations.
2. One-Click Probability-Based Forecasts: Teams now can view probability-based forecasts including delivery dates, risk areas and confidence intervals using Baseliner's service without needing to be a technical person to set up.
3. Automatic, Real-time Updates during Sprints: Each time an Agile team completes a sprint, their forecasts automatically refresh according to the work they just completed.
4. Simple to Read Dashboards: Users have access to district types of probability curves (density plots), uncertainty ranges, and likelihood of delivery using formats that user- friendly, not just to technical persons, but to all stakeholders and users of Baseliner's
software.
5. More Effective Decision-making Across Teams:
Agile Leaders are able to:
Specify deadlines that are realistic.
When committing to sprints the Agile leader does so with confidence.
Communicate to the team transparently marking all timelines.
Manage expectations more efficiently than previous behaviour.
6. Ability to Scale Across Multiple Teams and Enterprise-level Projects:
Baseliner allows for continued forecast across squads, programs and entire company, and is essential to successful enterprise-level Agile transformation.
Conclusion
Utilizing Monte Carlo simulation as part of agile provides more clarity, accuracy and certainty when making forecasts because it allows organisations to better utilise data-driven probabilistic models as opposed to relying solely on a person’s perception (i.e., guesswork).
Using Monte Carlo Simulation enables better visibility into the level of uncertainty so teams can effectively communicate this with their stakeholders when developing their planning and process improvement efforts based on accurate knowledge as opposed to simply guessing or taking opinions into consideration.
Through the integration of Monte Carlo Simulation into Baseliner.ai’s platform, it will allow agile organiZations to become smarter forecasters, better planners and able to manage uncertainty with a greater degree of confidence in less time converting the process of forecasting from a hindrance to a competitive advantage!
