Call Centre Scheduling Software: Excel or Not?

Excel is enough for call centre scheduling software right up until your schedule has to follow a demand curve that changes every 15 minutes. After that you need a tool that shows, at a glance, where your staffing doesn’t match your required coverage. That’s the whole buying decision, and everything below is how to make it without overspending.

This is Part IV of the Erlang C series. Part I covered turning call volume into seats and then headcount, and Part III showed what a real daily schedule looks like once you plot required agents against actual ones. This one is about the tool that holds the schedule. The rule underneath it doesn’t change whatever you buy: your minimum coverage has to match the call volumes and service levels the Erlang C maths says you’ll need, hour by hour. A spreadsheet can hold that. It just can’t defend it.

Is Excel enough for scheduling a call centre?

For a single team on one shift pattern, yes. A grid of hours down the side, names across the top, a row for required agents and a row for scheduled ones. The gap row is the whole point, and you can build it in an afternoon.

The trouble starts when the numbers pile up. A day has 96 fifteen-minute slots and a week has 672 of them, and every one is a place where required and actual can drift apart. Add breaks, a sick call, a shift swap someone agreed to verbally and a second team on a different pattern, and the spreadsheet stops being a plan and turns into an argument. Nobody can say for certain which version is current, and the gap row only tells the truth if someone remembers to update it.

Show me a manager who’s still hand-editing a coverage grid on a Friday afternoon and I’ll show you a schedule that’s already wrong by Monday. That’s not a criticism of the manager. It’s a sign the tool has run out of room. Staff cost matters here too: agent salaries run 60% to 70% of contact centre costs, so a few hours a week of overstaffing you can’t see is real money, and so is the understaffing that shows up as abandoned calls.

My rule of thumb is simple. If you can’t see the required-versus-scheduled gap for the whole week in under a minute, and if changing one shift means checking three other places, it’s time to stop patching the spreadsheet.

What should scheduling software actually do?

Four things matter, and the rest is decoration. First, it has to show required coverage against scheduled coverage by time of day, so a hole is visible without doing arithmetic. Second, it has to let you schedule in small increments and account for breaks, because a 15-minute lunch gap in the middle of a busy hour is exactly where the Erlang numbers fall apart. Third, it has to handle restrictions, meaning the person who can’t work evenings on Monday, Wednesday and Friday, so you enter that once and stop policing it. Fourth, it has to get the schedule to your staff without you emailing a spreadsheet.

Everything else is a bonus you’ll pay for. Forecasting engines, adherence tracking and intraday re-planning are real features, and once you’re past a few dozen agents they earn their keep. Call Centre Helper’s buying guide lists the same short list I’d give you: rule-based scheduling, what-if planning, manual override for the planner, and self-service shift preferences for agents. Note the manual override. A scheduler who can’t overrule the tool will end up back in Excel within a month.

Skill-based routing changes what you need from the tool. If your agents have different skills, a single headcount number stops being useful, because eight agents who can’t all take French calls aren’t eight agents. That’s a scheduling requirement as much as a routing one. Decide up front whether you need it, because it separates the cheap tools from the expensive ones.

One warning on the gap row. In my own worked schedule, 100 required agent-hours against 68 actually staffed left the team 32 hours short. A tool that shows you that number before the week starts is worth more than any feature list.

Whatever you pick, prove it on a week you already know the answer to. Load last month’s real shifts, breaks and time-off requests, then check whether the tool’s gap row matches what actually happened on the phones. If it says you were covered on the Tuesday afternoon everybody remembers as a disaster, the required-coverage numbers behind it are wrong, and no software fixes that. Fix the Erlang inputs first, then blame the tool.

Scheduling Employees 2000 or When to Work?

Two packages, two ends of the market. Scheduling Employees 2000, published by Guia International, is the step up from Excel. It’s fairly inexpensive and easy to use. You enter a single department or team and it checks that you have the right coverage by time of day, with drag and drop scheduling in 15-minute increments that accounts for breaks. You can see at a glance when staffing levels don’t match required coverage, and it tracks time and labour cost by the hour and the week. It isn’t fully web-enabled, but you can publish the schedule online so staff have a current copy. I recommended it when I first wrote this and I’d still point a single-team manager at it. The site now sells it as Scheduling Employees for Windows, so search under that name.

When to Work (W2W) sits at the high end. It’s fully web-enabled, so staff can reach the schedule from anywhere and your reporting and control get simpler. It’s very customizable and has automated staffing calculators that work with skill-based routing. Restrictions are its best trick: enter that someone can’t work evenings on Monday, Wednesday and Friday and the system will not let you schedule them then. Staff get a notification whenever their shift changes, there’s a company bulletin board for announcements that touch several departments, and different teams and skills can each run their own schedule inside one application.

Shift trading is the feature people underrate. Letting similarly skilled employees swap shifts themselves, with you approving the result, removes a whole category of small fires from your week. It costs significantly more than the simpler package. When you’ve reached a certain size, though, that’s a bill you’ll be glad to pay, because the alternative is your own hours.

Which one? Under one team and a handful of shifts, the cheaper tool. Multiple teams, multiple skills or staff who need to reach the schedule from home, the web-based one. Both have trial versions, so test with your real shift patterns before you commit.

Who answers the phone at 2am?

Nobody, and that’s usually the right answer. Scheduling software will happily let you fill a night shift, and Erlang C will tell you what that costs. It won’t tell you that staffing two people at 2am for one call, with a two-hour SLA, is a waste. I covered how the daily schedule and its overnight gap play out in Part III, so this is the short version.

An outsourced order-taker is often the better tool for that window. Someone answers the call, opens an incident and gives the customer a reference number, and your own staff pick it up when regular operations resume. The customer got a human, the incident clock started and you didn’t pay a full shift to wait. It only works if the handoff rules are clear, so tie it to your tiered SLA matrix and your escalation matrix. A Critical incident that lands at 2am needs a named person to escalate to, not just a ticket number.

If you have questions about any of this, use the contact page and I’ll answer or at least point you in the right direction.

Hutch Morzaria

Hutch Morzaria is a CX and Support Leadership professional with 19 years of experience building and leading support organizations across SaaS, Fintech, and enterprise technology. He has held Director-level roles at Q4 Inc, AudienceView, Johnson Controls, and others, and holds ITIL Expert certification across V3 and V4.