In the world of agile project management, accuracy in estimating time and effort is an important input factor for achieving efficiency in planning and delivery. If multiple departments are involved, like development, design, QA, and operations, synchronizing their inputs can be difficult. Forcing a process of estimating across functions captures the collective input of each discipline's effort, dependency, and timeline.
Teams can collaboratively create estimates in sync with each other through their functions, limit bottlenecks, and increase predictability during sprint planning.
Baseliner AI makes this process more accurate and data-driven. Baseliner AI integrates analytics, predictive modelling and visual dashboards to automate the structures and collaboratively manual estimating process.
Teams can take an input on velocity, planned versus actual effort, to be more accurate with their estimating process over time. All teams work collectively from the same origin and via a group interface. This supports collaboration across teams, is fully transparent, and adaptable to actual scenarios.
What is Cross-Functional Team Estimation?
Estimating in cross-functional teams is a shared process where people from various functions estimate effort, complexity and resources needed for work. Each team has their unique view; the developers gauge the work effort of coding, the QA team gauges the scope of testing and the designers assess what the user experience will mean to the dependencies. This visibility between teams reduces risks and gives a realistic estimate of the overall effort.
Techniques for Cross-Functional Team Estimation
Planning Poker:
Planning Poker is an agile estimation practice that couple’s deliberation and consensus. Each team member has numbered cards and picks their estimate for a user story—all privately. After they reveal their estimates, the team can deliberate on any outliers until the team reaches an agreement on the estimate of the user story. It is a means for a team to drive membership, remove bias, and think about all of the functional aspects of an estimate before finally settling on a number.
T-Shirt Sizing:
T-Shirt Sizing provides teams with a fast and relative way to size tasks as extra-small, small, medium, large, or extra-large based on complexity or amount of effort. T-Shirt Sizing works very well for the early backlog grooming and roadmap planning, when teams only know limited details. The team has a quick way to size tasks, and can get the discussion going for capacity, and alignment between functions before returning to detailed estimation.
Bucket System
The Bucket System framework is a powerful way of addressing and estimating various backlog items simultaneously. The team agrees on common "buckets" that represent a value range of effort (e.g. 1, 2, 3, 5, 8 points). The team will carefully allocate each proposed task into a bucket of millionaires after considering dependencies and impacts into account. This framework speeds estimating and provides common assessment mechanisms across estimating, stories, and teams.
Three-Point Estimation:
This estimation technique acknowledges uncertainty by estimating three values:
Optimistic (O) – the best case.
Most Likely (M) – the realistic estimate.
Pessimistic (P) – the worst case.
Teams arrive at a weighted average using the formula (O + 4M + P) / 6. It ultimately gives teams an average, balanced assessment of all possible scenarios. This technique is especially helpful when several teams are involved in a project with differing degrees of risk and dependencies.
Estimation Dashboards and Analytics
Current Agile estimation dashboards can provide instantaneous information on team performance, trends in deviation, and comparisons in velocity. They centralize the estimation data, differentiate actual from projected time/effort, and give actionable intelligence for future planning. With integrated or connected tools like an AI-based Baseliner, the estimation data can turn into predictive cognition and allow the team to proactively manage the workload.
Frequent Pitfalls in Cross-Functional Estimation
Ignoring Dependencies: Not realizing how completion of one team’s effort can impact another’s delays and rework.
Unequal Engagement: Letting the "loudest voices" craft estimates can influence the process.
Overly Specific: Spending too much time striving to gain a specific number often means the team feels they know the specific number.
Lack of Learning from Historical Data: Inability to compare the process to our previous estimations prevents us from getting better over time.
Best Practices for Accurate Estimation
Cultivate a climate of authenticity and transparency amongst all functions.
Employ various estimating methodologies that optimize both speed and accuracy.
Track progress toward ongoing improvement using the estimation dashboard.
Read and document assumptions and dependencies for future reference.
Look back to the estimates after each sprint, and keep progressing toward greater accuracy.
Conclusion
Estimation across a cross-functional team connects strategy to execution. By taking advantage of structured agile planning processes and establishing collaboration efficiencies across teams, organizations increase accuracy and predictability in estimating work. Baseliner AI is a tool that brings these together on a single intelligent platform, and intentionally enhances calculating effort, communication, and allows teams to consistently and accurately estimate work.
