Resource planning looks complex on the surface, but many of the biggest planning problems start with simple mistakes. Forecasting, capacity planning, scheduling, and real-time management all sit together, and each one depends on the other. For example, a misunderstood demand signal can create the wrong forecast. It can then affect the capacity models and schedules based on it, and put pressure on intraday teams when the demand is not met. A simple mistake upstream can easily travel downstream if left unchecked and create a big impact within the functions you’re planning for.
The good news is that these mistakes also have a simple fix. In our recent webinar, Ben Heslop, Director of Resource Planning at ResQ, and founder and author of Smooth Planning, covered 6 planning mistakes in detail and what to do instead. This blog presents the key takeaways from the webinar.
One of the most common planning mistakes is jumping straight into the forecast as soon as a number changes.
How can this mistake look? Consider you have a one-off, really busy week that you naturally try to smooth out of your forecast as a random event. Sharing an example from his early days, Ben mentioned a time when he saw 3 really busy Mondays and thought he’d spotted a pattern, and adjusted his forecast. However, he later discovered that it was due to high absence on Mondays, and when absence went back to normal, the spike was gone.
This happens more often than you’d think. Volumes rise, and the instinct is to push that change into the forecast. Volumes drop, and the instinct is to do the same. But a number on its own only shows movement and does not explain why the movement exists. Ben advises pausing to understand what drove it and what's likely to happen going forward.
Before any change enters the forecast, answer the three questions:
Planning cannot sit apart from the rest of the business.
Customer behaviour changes for many reasons, and often the business itself drives those changes. New marketing campaigns, system changes, bill runs, product launches, and technology changes can all affect demand.
When planning teams don’t know about those changes beforehand, the forecast moves off track, and real-time teams face the pressure. Average handle time (AHT) gets longer, queues build up, abandon rates rise, and everyone starts asking why the forecast missed the mark.
Involve planning teams early on and make planning cross-functional.
Create a demand calendar that holds anything likely to affect volume, workload, or customer behaviour. Make it cross-functional and keep it updated with input from departments such as:
Then review it continuously, check whether dates change, and look back after the event to compare what was signed off against what actually happened.
If an event can move customer behaviour, planning needs to know before the customer does. For contact centers to deliver a consistent customer experience and maintain a competitive advantage, it's critical to integrate WFM data. Modern WFM software is designed to support this by providing one source of truth for all your WFM data. When your forecasts, capacity, and scheduling all live in the same platform, it's easier to align with everyone.
Historical data helps, but only when it reflects the business as it exists today.
More data does not always mean better data. A long history can smooth out patterns, hide changes, or include information that no longer fits the current operation. Customer journeys change, technology changes, contact channels change, and business structures change too.
If the business introduces chat, splits a department, moves from paper bills to digital bills, or changes a process, older data may stop being relevant. For example, your business introduced bots to handle simple contact types. You want to forecast demand based on year-on-year variants; you use the 5 years' worth of data you have. But 2 years ago, customer service and sales were one department, and it was later split into 2. The 5 years of data are not an accurate representation of where your business stands today. The only relevant data to be used in this scenario, Ben advises, is the last 2 years.
Check:
A forecast only works when the data behind it matches the business being planned for now.
Complexity has a place in planning, but it has to earn that place.
It can be tempting to build a model that looks sophisticated. More inputs, longer formulas, more weightings, and more layers can make a forecast feel smarter. But complexity can also hide mistakes. Sharing a personal example again, Ben mentioned a previous role where he joined the largest planning function of his career. Every planner worked from their own copy of the same capacity model template, and years of additions had turned it into what he called "an absolute monstrosity": hidden sheets and formulas too long to trace.
The model consistently predicted better performance than the operation delivered. The business averaged that gap and added it back as a correction. If a department's gap was 7% on abandon rate, planners added 7%. For operations leadership and finance, that meant headcount decisions rest on a number nobody can explain.
The actuals and assumptions matched, so Ben worked through the formulas and discovered that the cause was an inaccurate shrinkage calculation. The model added a flat 10% to the staffing requirement, which left required FTE short. It was a simple mistake, hidden in an overcomplex model.
Start with a transparent baseline model. Having that can make this kind of error visible at its source. Modern Capacity Planning software uses the same validated calculation logic for every plan, so planners don't maintain their own copies of a template. This ensures everyone is working from the same numbers and removes unnecessary complexity.
Always validate the model you have by asking:
If extra complexity does not improve the forecast or the operational decision, it does not belong.
Outliers can create trouble in both directions.
It is easy to smooth them out or delete them from the data. But if the outlier signals a new pattern, removing it can hide the next arrival profile. On the other hand, some outliers should not stay in the forecast at all because they do not reflect future demand.
That means each outlier needs individual judgement. For example, when working with Excel, you can see a huge spike in contact and smooth it or just delete it from the data you’re going to use for forecasting. What if it was a pattern you will now miss?
This is where modern WFM software can help remove this mistake by spotting patterns quicker and deciding what to use and what isn't yet a pattern enough and seen regularly enough to include as a pattern going forward.
First, understand why the outlier exists. Then decide whether it belongs in future planning:
Examples from planning practice include outages, marketing-driven demand, repeated contact from poor service conditions, and process changes that reduce contact through digital containment or self-service.
The key is simple: understand the outlier before deciding what to do with it.
Plans go wrong sometimes, and that's normal. The real mistake is defending the plan instead of learning from it.
When a plan misses, teams can fall into one of two traps:
Neither approach helps the next planning cycle.
Turn every result into learning.
When things go wrong, review:
When things go right, review those too. Great performance also contains lessons.
The most effective workforce management improvements often come from simple habits: