Forecasts and WFM Tools | Doug Casterton

Dive into a riveting discussion with Doug Casterton, a celebrated leader in the workforce management (WFM) and customer care space, as we uncover the raw truths behind WFM tools and the often-misleading promises of vendors. With over 20 years of expertise, Doug has reshaped contact center operations globally, leveraging his passion for data-driven decision-making and AI.

In this episode, Doug and Irina dig deep into the flaws of modern forecasting tools, revealing why many vendors fall short of their lofty claims. Discover how manual experience often outperforms automated solutions, and why transparency is crucial for WFM professionals when choosing the right systems. Doug shares his journey from managing spreadsheets to becoming a recognized influencer and host of the WeWFM podcast, offering unique insights into vendor accountability and the future of forecasting technology.

Listeners will gain valuable lessons on what to demand from vendors, the importance of asking the right questions, and how smaller players are leading the charge in innovation. This episode is a must-watch for anyone navigating the complexities of WFM tools and vendor solutions.

Don't miss this eye-opening conversation! Subscribe now: https://www.youtube.com/@wfmunfiltered?sub_confirmation=1

Show Links:
 • RightWFM website: www.rightwfm.com
• Email: Irina@rightwfm.com
• Podcast email: WFMUnfiltered@gmail.com
• Podcast Directory listings: www.wfmunfiltered.buzzsprout.com/share
• YouTube Channel: www.youtube.com/@wfmunfiltered?sub_confirmation=1
• Doug's LinkedIn: www.linkedin.com/in/dougcasterton

Thanks for listening.

If you'd like to contact me about the show, you can email me HERE.

If you have questions about working with me on WFM projects and Consulting, you can find The RightWFM website HERE.

Please remember to subscribe and leave a review of you've enjoyed the show!

Transcript

Irina: 0:18

Hey, everyone. Hello,

Doug: 0:20

Welcome to 2025. Yeah.

Irina: 0:24

this is a great start of the podcast for 2025, but at least I have incredible guests. One that is very popular in our industry and that everyone loves. Everyone respects, especially me. So, I would like to hand over to you, Doug. Actually, are you in Berlin now? Shall I present you like going to Berlin or you're somewhere else right

Doug: 0:47

Yeah, no, I'm, still in Berlin. It's been my home for the last five years. Right now looking out the window, it's covered in snow, which is nice when you're indoors, but not so nice when you're out. But yes, yeah, Berlin's, my home at the moment. For the time being,

Irina: 1:02

Good. We're off to Berlin for the first time, I think in the show. And I'm so excited to chat with you. So, first of all, how are you doing and can you tell us a little bit more about yourself?

Doug: 1:14

sure, Yeah, I'm doing very well. And yeah, so I guess for my sins, I've been in and around workforce management for, goodness, I don't know, it's going to show my age. It's at least two decades. I think 25, 25 years. I, remember the days when everything was spreadsheets and there was no workforce management system. So that's, this is how old I am. I've done various roles across the UK, Singapore. Now in Germany and in the last last four years I've been able to really take workforce management as a bedrock and put that into customer experience. So my role has encompassed everything from customer care technology learning experience to the, vendor operation, and of course my passion Workforce Management.

Irina: 1:57

Thank you so much for that. And you're also quite active on LinkedIn. I remember we've been probably chatting for the last, I don't know how many years when you started with WFM together with the guys and you're also regularly writing articles and participating in different events. So can you share a little bit more about these initiatives?

Doug: 2:17

Sure, yeah. So, when I was living in Singapore one of the projects I had was to, create an outsourced Workforce Management, so Workforce Manager as a service. And the problem that I had was nobody knew what Workforce Management was. So it was pretty difficult to sell something when you, pitch up and I don't know what this is, what benefits is this to me. And that's really where I started writing. So I could, that was almost, that was two decades ago where, and at the time, the only people that were writing about the topic that I could find was, the system vendors. So they always had a, reason to obviously, because they're looking to sell software. But No, exactly. We're doing it for the love of the community.

Irina: 3:01

Absolutely.

Doug: 3:03

And, I really didn't, I continued it. So even after that particular project had concluded I found one that, that I really enjoyed. Brilliant content out there on this topic. Two, that there wasn't much still going out. And the third, piece for me was it, really helped me to expand my thoughts. So as I'm writing and researching a topic, I'm actually also learning. So that was a, huge benefit that I wanted to continue. And then in the last few years I, came together with a few other passionate workforce management people. And we decided to create a podcast called WeWFM. And Yeah, so that, that's also been extremely fun. I've just got to meet really special people in the community, people like yourself, but all different spectrums of different types of workforce management, and really shine a light upon their efforts and the types of things that they're doing and build, profiles, and build. Really our view and vision was to, to, really uplift the workforce management community in the lives of others. Because often people don't really know what work management is or what, or the value of it goes, unsaid.

Irina: 4:14

And I think you've done an absolutely incredible job and this is a shout out to the WeWFM podcast and to you, Doug, and to the guys as well that are also involved with the podcast because you were the first one to create something out there to create type of a community that's not a vendor specific that is bringing everyone together. And I absolutely enjoy what you have done and keep on doing. So keep up the good work. Definitely.

Doug: 4:43

Thank you. And vice versa, what you put out is great. I think I've always taken a bit more of a filtered approach. I love your unfiltered approach. I think you're very brave. I'm not brave enough to do it. But yeah, what you do is

Irina: 5:00

Maybe I'm just not smart enough. People use the term politically correct as like a kind of covering the fact that they're just smarter. I'm just like, let's say it is I make the effort to cover something. And on that note, we have a special topic. And the thing about this podcast is that usually I agree with the guests on a high level topic, but they have no idea about my questions. So he doesn't know what I'm about to ask him, but it's about forecasts. specifically in Workforce Management Solutions. So first of all, let me ask you Doug, this very important for myself question, because I'm about to rant a lot about it. Can you call a solution, a Workforce Management Solution without being able to forecast in this solution?

Doug: 5:53

I, no, I, oh goodness, that's a really great question. Not a holistic workforce management system. There are components to workforce management, but at the end of the day, It's like everything when it comes to any type of modeling, you're putting rubbish, your input really determines your output. So rubbish in, rubbish out. So therefore, yeah, if you're not forecasting with some level of accuracy, Then what the hell are you doing when it comes to the scheduling or the capacity planning or even real time, you, actually probably just solely relying on your real time team to, to accommodate for the gaps that you've got in your, staffing plan. So that's a half answer, but yeah, I don't think you can without, without a good workforce forecasting system.

Irina: 6:41

this is interesting for me because lately I've been talking to a lot of vendors that are claiming to have a solution. Amazing WFM solution, and when I'm starting to review it, it seems that it completely lacks the forecasting capability. So my first question is, okay, if you don't have the foundation, then at best you're a scheduling tool. But if you don't have a forecast

Doug: 7:04

Yeah.

Irina: 7:04

which is basically the groundwork to create your schedules, like, how the hell are you a workforce management solution? But this was okay, I'm gonna accept your answer on this one, cause I entirely agree with it.

Doug: 7:20

The only reason I was sat maybe a little bit on the fence on it is just that I know that you can produce forecasts outside of workforce management systems. So that's, the only kind of caveat. Do I think that the, in the modern day with, some of the challenges that we're now facing versus where we were two decades ago, things like, The, emergence of like customer behaviors that's much more volatile than it was at least what I experienced two decades ago. I think that if you're relying upon kind of very manual efforts now, either with a simplistic workforce management forecasting system inside the system or outside of it, I think in many cases, you're going to struggle to get a good schedule or, staffing plan. It depends like on it. You may be in a simplistic, very stable volume that you can just rely on a very simplistic. Manual forecasting spreadsheet. You probably could get away with that. But these don't exist for very, they're not, they're quite rare in today's modern context. In my opinion.

Irina: 8:22

I also agree with you that if we have the possibility to import forecasts from the outside, which in a BPO world, it's a given that you have to do that in your solution. I can agree with that. But if you don't even have the possibility to bring forecasts, you're just relying on, I don't know, your input, like five people here, 10 people there. It's. It's a scheduling tool, it's not

Doug: 8:45

Yeah. Yeah. Yeah. And what are you scheduling against?

Irina: 8:48

Well, your opinion, your gut feeling.

Doug: 8:51

or just static shifts without any flexibility.

Irina: 8:56

Yeah, but how good is your coverage then if you have nothing to compare it to? It's a, yeah, completely other question. But this is a great segue to, our main topic about forecasting WFM tools. So, I'm curious about your So, opinion and experience because when I have been brought to implementations working on both sides, on the vendor side and as a client, usually you start with covering the solution from start to end. And when it comes to forecast, especially if you have a client or a prospect who is a vendor, Very mature and very experienced in forecasts, they tend to drill into the details, like how do you do the forecast? What, how does the engine work? And I would say probably 99 percent of the vendors out there, it's like, it's just algorithms. Algorithm, and nobody can tell you exactly what kind of algorithm it

Doug: 9:52

And by the way, The use of algorithms in itself, in my opinion, is also part of the problem. But I can dive into that if you'd like. So I think that, I think that there are lots of different things that you need to look out for in a workflow connection tool to get a really best in class forecasting tool. So one of the things I just mentioned there is, so the use of algorithms, as in Erlang C, for example Erlang C was never designed anyway for, how we use it, even in a single skill context, let alone a multi skill It doesn't, we all know, I think it's pretty common that LNC doesn't include abandoned rates. So straight away you've got a, you've got a potential for overstaffing and it does also overstaff as well. But the, fact that we've now had Nearly 30 years of simulation models and yet we're still using an algorithmic or formulatic approach to deriving FDE. For me, it's crazy. Now I know that some workforce management systems do have simulation in it, which is great. But with compute power as it is today, I think the only reason not to use simulation models was just the amount of compute power that you need. That's not the case. Compute power is so much cheaper than it was, especially now with the emergence of AI. And, the excuses for not using simulation models, for me, I think it's just a case of somebody hasn't got round to developing it or hasn't seen the value of it, but it's a massive flaw. And then in terms of the actual sort of forecast input, so say the contact volume or something like this the, they're very different approaches like, so from a historical Time series perspective, there's various different models that you can be using, not just simplistic time series. So having a system that has a best choice algorithm or, method in order to forecast. So you take some of the brain power out of somebody needing to use multiple methods. The ability to, automate repeat contact detection I think is also important. And and event handling capabilities. So you've got the time series historical side of things, but just as important just as it is with stop broking, history isn't necessarily a good guidance for the future, especially in a business that's growing or changing quite a lot. So you need to really also have great event handling, not just for The things that might be happening like a marketing or product, but also like to, incorporate your normal kind of seasonal patterns that you might get, like for example public holidays or these sorts of things. And they're equally as important as the, AI algorithms and all the other types of things that I mentioned earlier.

Irina: 12:32

I'm going to throw a curve ball at you.

Doug: 12:35

Oh, here we go.

Irina: 12:38

In all honesty, how many vendors do you know that they are doing a great forecast?

Doug: 12:45

Yeah, no, Two, maybe three maybe one. Yeah, maybe

Irina: 12:50

And I think we agree on which one it is.

Doug: 12:53

might add a two, two or three, two or three might be there. I'm only saying that tentatively because I haven't used their systems recently as much as one. So, yeah, not many. Not

Irina: 13:06

Right. So my opinion on this one is that I know of one that is really taking the How do I say the mathematical approach to forecast and they can really in detail explain to you how their system works and what is the output and what is the input and just explain in details the forecast. There are some that can get away with forecast that can look oKish, especially if you don't have a lot of experience as a forecast. Majority, and they're never gonna say this one out loud, that's why I'm here to put things in front. They're using only moving average. So they're all algorithm and machine learning and bullshit here and bullshit there. It's actually a moving average. That's it. Enough. That's it. That's the approach.

Doug: 13:56

Yep. I would say the vast majority are that in my experience. Like I say, I can only think of three that have gone beyond moving averages. It's true. With a bit of event management in there as well to be fair to us. But yeah, that's it. Agree.

Irina: 14:13

that probably the one that we're thinking that they can get away with looking a little bit more advanced. It's still, in my opinion, limited in how much. How much input can you give? If you're saying, we have X amount of, I don't know, orders and this is gonna be this impact on the contacts and this, is something that you just can't input in the system. You can potentially do some kind of an impact on events or occasions, but everything is not possible. Very manual or it's not working and it's a dust in the eyes. So the thing that worries me is that we know very well that when we're going to conversations with vendors, forecasting and automated forecasting is a unique selling point for every single one of them. So what's your opinion? How automated the forecasting actually is in WFM tools? A good forecast, not just the rubbish that you will have to manually change anyway.

Doug: 15:14

I would say 90 percent of, the forecasts that are produced in systems that are either done outside or are, or at least start outside. And maybe they're just using the system, the moving average to break the volume down to, like 30 minute, 50 minute intervals. But in terms of the daily or the weeklies or monthly volumes, Maybe they've done something in the other system, but they've probably done outside of it. And they, cause again this goes back to the fact that another thing that's been under invested when it comes to system development has been capacity planning. So that, that usually is done in a spreadsheet. It usually has its own forecast and usually feeds into the WFM system. So there's a kind of another thing that's missing, but it's also probably one of the reasons why people don't use the systems aside from the fact that in your words they're pretty shit anyway. So you don't get a lot of value other than being able to break it down to 30 and that's it. And that, and using a moving average for that yeah, you're not going to get great results. Even breaking it down. cause yeah, the, differences in, volatility from a day to day or from seasonality. can be drastically different from week to week or month to month. So it's not just about the monthly volume, it's also about what happens in that particular day or week or month. The example being, say, just after a public holiday. You often see a reduced volume on the public holiday, but you'll see a big increased volume. the next day. Now that doesn't all land nicely evenly across the day. That'll be a different profile to, to what a normal Tuesday might be. So that's where you need to even with, even on a moving average, you need a quite dynamic moving average system, which they don't have on the whole.

Irina: 17:07

That suddenly turned into a very sad conversation. I'm just now thinking about

Doug: 17:11

it is. Yeah. I'm getting depressed as I

Irina: 17:13

it. it. is a bummer. Oh gosh. And on top of it, it gets worse from here. This was our pick, by the way, in the conversation. The other thing that bothers me is I have worked with so many incredible forecasters, forecasters that are like, they're blowing your mind with the things that they can produce in Excel and the way that they know their business. And usually in conversations with vendors, the first thing that they're asking is, okay. How can I affect, rather than manually, what kind of a simulation your tool is using? And let me think, I know exactly of one that you can actually choose the simulation that you would like to use. Everything else is Vendor Secret We use over 150 algorithms, but we cannot name one. But it's very sophisticated. It's complicated. And I'm wondering, do you think it's a good idea that these things become a little bit more open to the user so they can affect the simulation models somehow? Or they can choose at least from the simulations which one to use?

Doug: 18:25

I do. Yeah. If you've got a skilled forecaster that is able to. To determine which model is best to use. Absolutely. Having said that though, I know one vendor that's built inside their forecasting tool. Something that I mentioned earlier around this, best choice selection. So, it only really works on, historical time series type forecasting algorithms, but the, if you've got an AI system that's able to look back in history at all of the different volume and say, and then do a simulation to model which model is, which forecasting model is, best to use. This can really save you a lot of time. But, I think even in those circumstances you should have the ability to choose because you might not agree with it, or you might know something that's going to happen in the future that the, best choice system can't possibly know about. And therefore, the, I don't know, 2025 is going to be very similar to, And this, these types of things happen sorry, 25 have, I've gone back to the future there. It was very, similar to 23 and therefore these types of things are going to happen. Therefore this model forecasting particular algorithm is the best to use versus this one, which I used last year. Yeah. In those circumstances, for sure. I can see a lot of value. And I, and by the way, on my theory on, on why, Often the, vendors will say Oh, it's magical. It's the people actually selling the tools or doing the pre sales, they don't know. That, that's my experience. So whenever, I love to really dive deep into how does your system work. And as I start to pick the questions out, I usually find, oh, sorry, I don't know. I'm going to need to go and ask the product team for that. No one's ever asked me that. And maybe that's something that the workforce management community needs to step up on. And it's we need to ask these questions more and insist that we get answers on them. Because they're so, so important to, to, to the development of our, Discipline, because we need the systems to step up. But also you can't hold account, you can't hold them account unless you're asking those sort of questions and insisting that they answer it. And it is hell, by the way. Cause I'd ask that question and it takes weeks, and can we move on to you still need to answer me this.

Irina: 20:37

Yeah, I, find it ridiculous where even within the organization itself, you can't get that information as a consultant or as a account executive or what have you. And the thing that really bothers me, and I'm very passionate about it because someone like you that has been in the industry for 20 plus years. You can ask those questions, but imagine that you have been on that position for three months and somebody tells you we're going to get a WFM tool then you're getting someone with I don't know, a suit who is explaining you how great it is, how expensive their solution is, and how it's going to make your world magic. And then you have nothing to compare it. You don't know yourself what to expect. Often the decision is made by somebody above you, like a contact center manager or operations manager, or have you because of the name of the the vendor. And then you have no impact as well. So why would they change their recipe in their software? If they're selling it anyway, why would they invest? They're not stupid. They wouldn't waste. Resources to change something that's working.

Doug: 21:49

No, and So you've, hit the nail on the head when it comes to the decision makers are often not workforce management people. And the other thing that I see, okay, ever more than when I first started out in workforce management is the WFM tools are often packaged as, just an add on to other technology usually telephony systems and such like that. So it's just the freebie that comes along with it. And that for me gives no well, no economic incentive for, the vendors to do anything with it. If they know that one, you've got legacy system inertia where once you put a WFM system in, and this is the nature of Workforce Management, everything's tied to it, everything's linked. It really embeds itself in your business. So you've got this, you've got this sort of system inertia where it's actually quite challenging to switch it out. And then you've got on top of that the lack of economic incentive from the vendors part, because why would you spend money on product development when you're not selling it, you're not making any revenue, it's just something that You add on in order to sell a larger piece of software.

Irina: 22:58

But it's, so sad because we keep on saying that we're investing in AI customer experience and things are shifting and changing. And if you think about it, and that's why I hate so much the conversation about AI, because even in modern technology, there are so outdated capabilities that are ridiculous. But if you're listening to the sales conversation, it's the number one solution that is doing such amazing results. No, it's shit. It's automating some stuff. Okay. It's making your life better in some aspects, but it's not how you're selling it. So now when we're introducing KI, you're basically saying. You've been talking rubbish for the past 25 years. This is how you've been selling your solution. And only now something will change because we mentioned the terminology of AI. So I, I don't know how, Doug, you're somebody who is also very influential. What can we do for the community? Because my concern is that overall on the topic of forecasting, there is not a lot of knowledge out there, how to do a proper forecast.

Doug: 24:08

So I think it's like with every technology purchase, especially if you're contending with free, you've got to prove the business case. So, I think the, first thing that I would encourage workforce management teams is to get involved in those business cases. So assess, for example, where are you at with your forecasting accuracy today? And run some sort of scenario planning where you're looking at what does that mean in terms of either overstaffing or even understaffing, so therefore you're not meeting service level. And then calculate what the return on investment will be if you improve your forecasting accuracy. And then the other tip that I would add on to that is that in every RFR, RFP that you're running for a Workforce Management System, Ensure that you've got a vendor that only provides a dedicated workforce management system. So it doesn't have any of these other packages use that as a comparison to some of the other ones. Some of the other ones that maybe are a bit bigger and have other software packages. Because you'll tend to find that the innovation is being led by the smaller players that are dedicated to workforce management. Whereas, I don't, it's not to say there are some vendors, there are the bigger ones that are innovating, to be clear, and I want any of those to think that I'm not saying they aren't.

Irina: 25:28

Go after you.

Doug: 25:30

But, on the whole, it's, the, smaller players that are innovating. And I'm not saying you. can't get them, I'm just saying, put that in your RFI and RFP to compare it and specifically look at the, what they provide in terms of those, forecasting elements and compare it with the other sort of sales pitches that you're hearing. And then you will quickly see that there's a big difference.

Irina: 25:56

I love that you're saying that and the only other thing maybe that I would add in terms of advice specifically towards vendors is to be more transparent about their solution, even within their organizations. And personally, what I have found worrying is that you have a lot of solution consultants that are able to tell you, How their solution might work, but not necessarily to talk to you about the processes behind it. And I find that the very button clicking orientated and this often cannot apply to organizations because every reality, every business is different. And just the fact that you're saying, okay, Click on Open, and this is gonna expand something, and then you're gonna see the the screen. That's not enough for me from a solution consultant, so educate them more on WFM as well.

Doug: 26:53

It's a great tip, and I think when you were saying about, yeah, Doug, you've got experience, you'd ask these sort of questions maybe somebody with less experience wouldn't I think that's, it's a different topic, but I think that, One of the advantages that I have that I believe I have over somebody that hasn't ever used spreadsheets to do something very manually is that I understand the nuances and the detail behind the button clicking. So when, the WFM system outputs something, I can pretty much use my sort of gut feeling around, is it doing something that I expect, or is it doing something unusual? And I think that's where some of the ones that haven't had that sort of manual experience or haven't spent the time to really go and Deeper in terms of the theory around the back end, how do their algorithms work? What's behind the button pushing? they, struggled to assess their output. They literally, I've seen outputs there that they looked really great, but we could quickly see that there's something quite not quite right there. and I think that so if you haven't had the, sort of The experience of doing something manually, that's not to say that you can't still go and understand the theory. And I think that's an area where we're losing a little bit in the workforce management community, that sort of more in depth knowledge. We're becoming a little bit like, we know which buttons to push and that's it.

Irina: 28:20

Thank you so much, Doug. Any final words of advice towards the community, towards

Doug: 28:26

No. No no, I think I've probably just upset everybody. So,

Irina: 28:30

Don't worry. Probably they're thinking, Oh my God, Doug is absolutely phenomenal. He's so

Doug: 28:35

oh, whatever.

Irina: 28:35

correct and she is shit, so cancel her.

Doug: 28:39

Was I? Okay. Well, that's my British in me, I guess being politically correct or trying to be, but yeah.

Irina: 28:44

No, you've been absolutely brilliant. And as mentioned, I appreciate so much what you have done for workforce management in the community. And on this note, I have two things. First of all, let's. Let's actually challenge vendors if they truly believe that their forecasting works really well. Okay, let's make a webinar. Show us how your forecasting works. It might bring you some leads. If you're right spot on, we're going to challenge you on it. But the second one is that, please tell me, how can people find you?

Doug: 29:17

So I'm, always active on LinkedIn. So, and I'm, Anybody with a Workforce Management or Contact Center background, reach out to me, I'm going to instantly accept. I've got a, like I mentioned earlier, we've got a podcast and blog on WeWFM, so you can figure that out. But yeah, generally I'm very available for the Contact Center and Workforce Management community especially. And by the way, on your challenge, for those vendors that might be listening out there I'll be very happy for you to come onto the podcast. We'll interview you. I'll even write about the software. If you, so that's the challenge here. If you can show me that you have best in class forecasting software, that say, marries and competes with say, supply management, which really has great forecasting. I'm happy to give you some free promotion on it. So there's the offer and the challenge out there to the community, but also to the vendors.

Irina: 30:11

Perfect. Let's support the community. Let's see if vendors are going to jump on that challenge. And thank you so much, Doug, for agreeing to have this conversation with me. It has been absolute pleasure and you're welcome anytime.

Doug: 30:26

very much. I've really enjoyed it.