What Do Great Planners Do?
A Composite of Strategic Planning Tips
By Ric Kosiba, Real Numbers
At the end of last year, SWPP sent out a series of tips that I authored that were titled, “What do great planners do?” I probably had more feedback around these tips than any other that I have ever written. Fun!
We thought it would be good to compile the five tips together and make it a single article to put them in one place, so here it is!
What Do Great Planners Do? Part 1.
I recently had a conversation with a planning manager who asked about the regular tasks that a solid strategic/capacity planner should be performing every week. So, I started making a list and thought I’d share it with all of you.
Let’s start at the beginning — data. Great planners:
- Read in new data at the appropriate time intervals (usually weekly) and from all appropriate systems. There should be a cadence to this, say, every Monday morning we grab files from our contact platform, from our HR system, and from our WFM system. We organize it for analysis.
- Adjust data for definitional differences across systems, if needed. Different systems may have slightly different definitions of things like productive time, service level, abandons, etc.
- Annotate data. If something out of the ordinary happened last week, it is always a good idea to understand and note what happened. Was there an outage somewhere? Was attrition suddenly higher? Did some external event drive contacts? Was the mix of calls suddenly different, leading to higher or lower AHTs? Did the local high school basketball team win their playoff games leading to higher “sick” time? Whatever changed, we need to keep a diary in our data so we can understand why, when looking back.
- Correct anomalies. If something explainable or unexplainable happens, that is out of the ordinary, it is important to make sure that the anomalies do not throw off our forecasts. Big, one-off events also fall into this category. In this case, we both annotate the data and change it to something normal, for forecasting purposes. It is important to keep regular, unaltered data, and a separate data set for forecasting purposes. The best forecasting systems will automatically do this for us.
- Calculate lost time. As discussed in an earlier tip, lost time is unaccounted time, usually from a disconnect between our various systems, and usually representing the difference between our top-line agents, minus all the shrink, and our bottom-line agents. It is good to keep tabs on this, and to maybe even forecast it.
- Determine the difference between previously forecasted values (volumes, AHTs, controllable and uncontrollable shrink) and actual values, from a few weeks to several months out. This is one of the core responsibilities of every forecaster, to demonstrate how well we have performed. The proof is in the pudding with every new piece of data.
What Do Great Planners Do? Part 2.
There is a cadence to being a great strategic/capacity planner. Here we discuss the steps an experienced planner might take to turn data into forecasts, and data and forecasts into some decisions. Great planners:
- Monitor main plan drivers relative to our expectations. Is attrition coming in high or low? How about shrink?
- Show the operation’s misses relative to the plan. Did we miss our hiring targets? Our attrition targets? Show our misses and the cost of each miss every month.
- Act as a canary in a coal mine. Looking at the misses, is there a reason for it? Will performance come back in line? We will need to decide when to tell management and what our responses to operational changes should be.
- Calibrate and validate forecasts. With new data, we should check to see if our forecasting models require tweaking. This means we will rebuild our models to see if a new set of forecast parameters or a new math model type would predict better. Do note, however, that a forecast model should not change significantly with just a smidge more data! We do not want forecasts to vary significantly every week!
- Forecast everything. When we are building plans, there is a tendency to always discuss volume forecasting. But certain shrinkage categories may be seasonal and important to analyze. Attrition is often seasonal and predictable. Handle times require their own forecast. Sick time can be seasonal.
- Forecast different locations differently. Shrinkage may be locally seasonal and require a separate forecast. Attrition, too.
- Track which methodologies perform the best. The various data streams likely require different forecasting methodologies/models. Try them all and keep track of which methodologies work with which of the metrics.
- Consider combining forecasts. Can we combine forecasting methods to produce a better ensemble forecast? Many statistical tools can do this for you.
What Do Great Planners Do? Part 3.
There are standard strategic/capacity planning tasks to be done! Let’s discuss the steps an experienced planner might take to turn forecasts into staffing required. Great planners:
- Protect the spreadsheet or scenario! Make a copy of the current official plan. Name it something everyone will understand. Then make another copy and send it someplace safe. Put it on a thumb drive, too. Make another last copy for fun.
- Move the plan forward. Once the new forecasts (volumes, shrink, attrition, AHTs) are done, it is time to enter the new expected forecasts into our spreadsheet or system. And so,
we need to add time to the end of the plan, roll forward the date. Although it is a standard task, we all know we can mess up a formula that might be hard to find later, so we need to be careful!
- Do a quick staffing calculation validation. If I were to plug my actual performance last week — my actual volumes, attrition, top line staff, handle times, etc., use service achieved as my new goal, and plug it into my spreadsheet or system, will it tell me I would need the number of staff I actually had available? If so, great. If not, uh oh.
- Do another quick staffing validation. If I plug in my last week’s staffing into my model, will it accurately predict service level? Abandons? Occupancy? ASA?
- Have a meeting. Explain what is different in the contact center data from what was expected. Was attrition higher? Were volumes out of whack? Are we noticing a trend different from our previous plans? If there are differences that look like a kink in performance, it is time to get management on board that changes might be necessary (to the budget).
- Put together a first pass of the weekly staff requirements. 18 months out is a good rule-of-thumb.
What Do Great Planners Do? Part 4
Let’s continue with the next set of tasks, turning requirements into cap plans. Great planners:
- Compare staff requirements to our last staffing go-around. Are the new plans very different from the last plans? If so, is it because of us — our staffing models — or is it because our environment is changing? Models can be weirdly sensitive to changes (especially AI models!) and if the world hasn’t changed much, but staffing has, then we need to delve back into our models and processes. We can’t recommend changes if they are just because of our math.
- Check to see what’s in the works. You’ve previously published hiring plans, are they being met? Just because we say hiring is necessary, it doesn’t mean that HR has been able to meet our plans. Check with them to see all is still good.
- Develop new variable labor plans. Resource plans are a combination of hiring, overtime, undertime, training, and others. Does our planning process balance them all out? Have they changed from the last plan? If, for instance, HR cannot hire to our plan, we will need to make changes.
- Communicate with all stakeholders! Make sure we have a meeting where we decide, with HR, center management, and our VP of contact centers, any changes with our variable labor plan. Show everyone how accurate we are with our forecasts and our staffing models. If we can make changes interactively as questions arise, that would be awesome.
- Lock down the final-final version of this plan. After all interested parties have blessed this resource plan, make it final!
What Do Great Planners Do? Part 5
Let’s discuss the fun stuff — business analysis. Great planners:
- Answer our exec team’s what-ifs. I have never sat through a cap planning meeting where I left without being assigned some more what-ifs. If you’ve got a great planning process, you can (powerfully) answer their what-ifs during the meeting. But taking them away and following up with the answers to their what-ifs is of value, too.
- Discuss trends. As part of your assessment of the plan, each important line item (volumes, shrink, attrition, handle times…) should be analyzed for variance to plan. Change introduces risk!
- Display marginal analyses. This is a simple question that asks, “What happens to my operation if I add one more person? Or get 100 more calls? Or have our handle times increase by 20 seconds?” How many more calls will you handle? How much does your service level change? How does occupancy change? Costs?
- Produce sensitivity analyses/graphs. The more robust type of a marginal analysis is sensitivity analysis. In this, we use our staffing models to vary performance drivers and see how performance changes. In other words, we answer questions like, if attrition increases, how bad will service get? When should we start to worry?
- Do some risk analyses. If your business is in an era of growing uncertainty, what is the risk of sticking with our current plan? What would throw it off? Can we take steps now to ensure we avoid future service issues?
Ric Kosiba is a charter member of SWPP. Ric is a founder of Real Numbers, a contact center capacity planning and modeling company. He can be reached at ri*@*********rs.com or (410) 562-1217. Please know that I am very interested in learning about your business problems and challenges (and what you think of these articles). Want to improve that capacity plan? You can find Ric’s calendar and can schedule time with him at realnumbers.com. Follow Ric on LinkedIn!
(www.linkedin.com/in/ric-kosiba/)