Hey, everyone. Kumar Dattatreyan here with Agile Meridian, and I'm joined today by my buddy and compatriot for the XSCALE series of podcasts that we're doing, Glenn Marshall. Glenn, thanks for being here. Today, we thought we would talk about throughput accounting, as it's one of the things that Peter Merel (Founder of XSCALE) talks about a lot in XSCALE as an accelerant to delivering value. We thought we would talk about what that is and contrast it to cost accounting, which is what most of the world uses. So, let's kick it off. Glenn, what say you? What is cost accounting? First of all, let's get some definitions out of the way.
Sure. Cost accounting is simply looking at your profit and your costs and doing what you can to lower your costs. In a sense, it's somewhat mechanical, and throughput accounting differentiates from that by being less mechanical.
Yeah, that's right. I'll just elaborate a little bit. Cost accounting assumes that there is a fixed cost to something. The cost of labor is fixed. The cost of marketing is fixed. The cost of goods are fixed. So, you say you're selling widgets and the widgets cost $3 a widget and you sell it for $6, then your gross profit margin is 50%, right? You sell for $6, you buy for $3, the margin is what you're left with. And that's fine, I think. I mean, it works in most cases, but what it doesn't account for is the variability in those costs. And that's what throughput accounting sort of differentiates. Right?
If I could just add one thing, cost accounting, if you wanted to improve your margin, you would look at ways to reduce your cost and just cut them to increase your profit. You might get rid of the building or something like that, just look for ways to cut costs, but the emphasis being on doing it mechanically.
Yeah, that's right. A good example of that, I'll say, one that hits close to home, is my own experience in my company, Agile Meridian. We had to let go of some people that were handling our marketing to save some money. Right? And so that was a cost. We were able to save whatever money we were spending on that. We thought we could assume what they were doing on our behalf, and thereby, we reduced our operating expenses and returned more of the profit that the company makes to the bottom line. And that is a good way to think about it. Right? Of course, there's a hidden cost to that, getting rid of the company that handled our services. And that hidden cost is now there's more of a burden on us to do that work. And we're not marketing experts. You know? We're not graphics experts. And so the cost of saving that fixed cost is that each of us partners are having to do things that we're not well equipped to do. Let's keep that example in mind as we run through throughput accounting, and we'll look at it from a throughput accounting perspective, as well as the cost accounting which you just articulated quite well.
Right. And so from a throughput accounting perspective, using the same example, our goal was to hire a company to help us with our social media presence and our email marketing presence and things like that by executing a strategy. The strategy was to build followership on social media platforms, mainly LinkedIn, Instagram, and YouTube. You know, this video here is on YouTube, and, of course, on podcast platforms. And so, improve our brand recognition that we, Agile Meridian, and the Meridian Point podcast, bring these types of hopefully valuable services to people that view it and listen to it. And so in that sense, our throughput metric was increased viewers and followers. Right? Over the past couple of years, our viewership and followership has increased by probably 1000%. So that's the throughput metric to say that, you know, we spend so much money, and we're going to expect a certain return from that. And that return is in our brand awareness and our brand viewership. Any questions on that, Glenn?
So that gets to the heart of throughput accounting. What was the relationship between these folks that you're no longer working with and your followership? Was there a correlation? The more time they put on it, the more followership you got. And now that they're gone, what's the impact?
That is still to be determined, but I can say that looking at the numbers, they have plateaued. We don't have as many new followers on Instagram or LinkedIn or platforms like that. And as you can see by my haggard expression, I'm working way more trying to do the marketer's job for our company. So I would say that the throughput has gone down. Yes, we are saving money, but our throughput metric, which is increased exposure to our potential clients and customers and the general public, has suffered as a result. So our throughput has declined even though we have saved costs.
Okay, so let's look at this from a strict throughput accounting perspective. You want to go and increase the number of followers. Was adding more capacity, rather with your third-party company, increasing your throughput? It sounds like not.
It is not. No, it is not increasing our throughput in terms of followership or subscribers, if you will, to the services that we offer. You know, one of the things that XSCALE talks about is the pirate metrics. And we'll cover that maybe in another episode of this XSCALE series of this podcast, but the pirate metrics, just in a brief snippet, the reason it's called the pirate metrics is the letters that make up the acronym. It's AARRR. And really all that means is, it stands for acquisition, activation... What's the next one, Glenn?
Activation, retention, referral, and revenue.
Yes, right. So in the case of the metrics related to this throughput measure, meaning increased followers and so on and so forth through this expense, what we were getting was a lot of acquisition. We had a lot of new people acquired into our network that followed us, you know, subscribers on our YouTube channel, people that were downloading our podcast, people that were following us on Instagram, following our LinkedIn page on LinkedIn. And in some cases, there was activation. So activation, they were interacting with the content. They were liking it or sharing it or referring our content to other people. That's the R, the referral part. The first R part is the retention. They stayed as subscribers. The people that subscribe to our YouTube channel generally stay on as subscribers to our YouTube channel, and they refer others to it. Now what it hasn't resulted in, and this is the key thing... So our metric, of course, was increased followership, increased subscribers to our various social channels, but that last R, right, the revenue which we weren't measuring, to be honest, to be truthful, but one of the things that myself and my partners started to talk about is, okay, we've made this investment in social media, you know, significant dollars over the course of the year. And it hasn't resulted in any, at least not that we know, any additional revenue. And so, yes, it did increase the key metrics that we were looking for, but at this point, we decided to suspend those activities to save the costs, the dollar cost of the expense, but at a cost of continuing to build a brand. Does that make sense?
But this goes to the key point. What was your goal? Your objective? Was it to increase followers or was it to increase revenue? It sounded like it was to increase followers and therefore it was quite successful.
Yes. But, as you know, the industry, the agile coaching, boutique coaching, is going through a bit of a downturn here. And so even though I would love to continue that and continue to build our followership, at the end of the day, the bottom line is the bottom line. And if it's not contributing to revenue growth in some measure, in some way, even though explicitly our goal was to increase followership, implicitly, I think anything that a company does should have some impact on throughput, right? On monetary throughput, increasing the top line revenue, goals, revenue for the company. So what do you say?
So continuing with this example, perhaps the original goal of followership wasn't the best goal. Perhaps it should have been a more revenue-based goal.
Yeah, I think, in hindsight, yes, I think it's a worthy goal to increase followers and increase likes and subscribers to your YouTube channel and such. In a way, the YouTube channel could be monetized if we had enough subscribers. We're not there yet, right? So we need a certain number of subscribers and we're close, but not quite. And so continuing on this path eventually would get the numbers we need to monetize the channel and start making some money from it. But since we're not there, I think a lesson here in throughput accounting is whatever you do, you know, you set a goal for your company, and you think, okay, we're going to... Give me an example, Glenn, of a goal. I think we can work with your example.
I think we can work with your example. If your followership was actually, you know, the funnel effect and you were actually getting some clients, some paying clients out of it, that would have been a good thing. You might have had so many clients that you would have to enter the market to hire some more coaches. And let's assume that you're able to do that, and you had lots of coaches. Well, given that you're getting stuff coming out the end, and you can still get coaches, perhaps you should go and spend more money with this third-party company to get them to do even more social media work.
Yeah, exactly. And so, you know, looking at Goldratt's definition of throughput accounting, the goal of throughput accounting is to understand the relationship between operating expenses and returns and to make decisions that maximize the throughput of the entire system. So while we got good results from our social media marketing experiment, it didn't benefit the entire system. It benefited just part of the system, correct?
Right. It benefited the part of the system that was to get more eyeballs on our website and more people subscribing to our YouTube channel and so on and so forth. But the entire system includes some monetary benefit. If it generated enough monetary benefit to offset the cost, then I believe we would have not gotten rid of that service. We would have continued to use them. But because it didn't, we were forced to make a decision, make a choice. Right?
So let's play with that a little bit. I think there's still some juice left in this. If you had lots of clients asking for lots of work, and you had coaches that had capacity to be used, what would have been the bottleneck?
I'm sorry. I don't understand your question in relation to the example.
Let's go to a magical world where you were getting revenue out of the followership.
Okay, yes.
And you had coaching capacity that you could call on to fulfill that revenue opportunity. What would be the bottleneck in that case? And where should you put your attention?
Okay. So in your scenario, you're saying, okay, the social media marketing campaign is doing more than increasing followership. It's also resulting in revenue, correct? People are finding us, they're using our services, and we are capacity constrained, right? We have not enough capacity to satisfy the demand that's coming through our virtual doors. Okay, perfect. So in that case, what would be the bottleneck?
Insufficient capacity.
Correct. And so we would address that capacity constraint by adding new coaches or coming up with different programs or offers that allow us to service more clients in a way that it doesn't consume all of our time and capacity. Right? So either of those ways might work, but the idea, and I see where you're going with this, is once you've solved for a constraint... So in this case, we spent money on marketing, and the constraint was that last R, correct? Revenue. We got the acquisition. We got the activation. We got the retention. We got the referral. We did not get the revenue, the last R of the AARRR, pirate metrics. Right?
And so the constraint was, in this case, we weren't converting, correct?
And let me just run with this for a bit. So if you were converting in this alternate universe, and you used all of your capacity, then you put the call out to your colleagues, got a number of crackerjack coaches working for you, and you still had additional capacity. So you addressed that bottleneck by bringing in more coaching capacity, what would the next bottleneck have been? And let's just play through this scenario.
Well, it's hard to say, right? So we addressed the capacity constraint. Because we've addressed it, we've gotten those crackerjack coaches and they're coaching. The next constraint might be something else. It might move somewhere else. Maybe it's in the referral space. People are unsatisfied with the coaching that they get, right? And so the constraint has moved from the last R to the next to last R, the referral part. What we're getting is people are using us once, and then they're moving on. They're not coming back to us. And that would definitely be something to address, but I know you're a top-notch coach, so I'm sure that your coaching clients would be very happy with you.
You might end up going all the way back to the beginning with acquisition, and you could then talk to this third-party company and say, "Look, you've done great work. We've got leads. We've turned them into revenue. We've hired more coaches. You've been doing such a great job. We actually think that you're the bottleneck now, so we would like to increase our spend on you so you can bring in more revenue to the pipeline."
That's a good point. So given that the coaches are all great and our clients love them, we may not have enough work just to sustain our cadre of coaches. And so you're right. The next bottleneck might just be that we have more demand than we have coaches. And so we need to generate more leads.
Bingo. Right? We generate more leads, and so we actually spend more money. And that's where the throughput accounting concept is beautiful. Right? It's geared to and tuned to how the business operates and where the constraints are. Right? I mean, in my scenario, it could be that we piss off our customers. And now we have a different issue. We have a glut of capacity, but not because of marketing. It's because we weren't able to satisfy our clients, but that's not going to happen because, you know, you know me and I would only bring coaches in that are like me. So anyway, but I get your point. And that's the beauty of throughput accounting and how it's different from cost accounting. If it was purely a cost accounting model, what would the response be from the accountant?
Look around. See what you can cut.
Exactly. And you look at companies that are doing this now, you know, companies that we all respect. Right? Tesla, Apple, Meta, Facebook, Amazon, they've all cut a bunch of people, laid off tons of people. I would like to think that they are more on the throughput accounting side. What do you say to that?
I couldn't agree more, but yes, I am actually quite disappointed at this fad of laying off people. I think that, rather than just mechanically cutting costs, look at where your bottlenecks are, and rather than lay off people, redeploy them to address those bottlenecks, and then get into a growth cycle.
You do this, and the idea is, you look at the complete stream of value throughout the organization, and you identify the current bottleneck, make sure that it has been addressed and value is flowing again, and then you find the next bottleneck. And by doing this again and again and again, you can get an exponential increase in throughput and profit. Right?
And I think in a way they are doing these things, they're just redeploying their resources. They save by laying off all these people, redeploying those resources, the money, and investing it into AI and AI-based technologies. Right? And so that is going to reap, I mean, look at what Meta's latest earnings report... They're projecting greater costs and lower revenue in the next quarter, in the next cycle because they're putting more money into AI and development of whatever their thing is.
Here's the problem with layoffs that I see. Google has done some very good work on this, around psychological safety. What is the main reason that some teams are better than others? And the key finding they found with psychological safety is that people feel comfortable speaking up. They're not afraid of being ridiculed or losing their job. And when you have a layoff, people get scared. They go into a shell. They don't want to go and stick their neck out because they're afraid they're going to be the next one to be laid off.
I've been at companies where that mindset is pervasive. Yes, there's a bottom line savings in terms of expense, but there is another cost which people don't take into consideration, and that is basically the psychological trauma that people go through when there's a layoff, particularly if it's an across-the-board layoff. Far better to go and take that capacity that you identified as "we could be without" and redirect it to eliminating bottlenecks.
Yeah, no, I'm with you. I just think that there are reasons beyond, it's not as simple as that when it comes to these big companies, but as I transition my practice to helping more small companies, I see that throughput accounting is such a powerful means for them to increase the efficiency of their operation, improve the profits that they generate, and also satisfy their customers like never before, all while keeping their employees and growing their employee base and, you know, supporting their employees' growth and so on and so forth, because it's not just focused on costs and waste. It's focused on improving the top line. They're all interrelated and throughput accounting draws the relationships between these two things in a way that cost accounting alone cannot do.
There is a bit of a subtle point around throughput accounting, and that is when you've gone and optimized the flow and eliminated the bottlenecks, you look at what's left over. And if there are things that are not helping with the flow, you redeploy the capacity from those areas. So there is kind of a cost accounting angle, but the key point is to focus on throughput and limiting bottlenecks.
And I would say that element that you talk about is really a systems thinking approach to how they run their organization, whether it's a small business or a large corporate structure. Right? It's thinking about the system as a whole and, you know, thinking about where the flow has been blocked or is constrained and redeploying resources to lift the constraint, knowing that that constraint may reappear somewhere else and being ready for that.
An analogy that I often use when I'm talking about this is logs in the river. You know, you're throwing logs in a river. It's a big river. There's a lot of logs. At some point, they're going to jam up. That's your bottleneck. Is there any point in throwing more logs in the river? No. Is there any point in trying to make the logs flow better ahead of the bottleneck? No, that'll just make the bottleneck worse. Is there any point in going downstream from the bottleneck and making that more efficient? No. They're already sitting around doing nothing because of the bottleneck. You go to the bottleneck, you optimize it, you change the curvature of the river so that the logs can flow more easily, and once you address that, you look at the river again. There will be a bottleneck somewhere else. It never flows optimally. This is how value flows through an organization.
It's a good point. Although, you know, changing the shape of the river has consequences.
It's a metaphor, right?
It reminds me of our civic planners and how they build roads and keep making them wider and wider and keep building more and more houses. And the flow doesn't improve ever. It just gets worse. And that's because everybody has to go downtown, because everything is centralized. Whereas if you distributed a bit more, continuing with the river metaphor, if you have some canals or some tributaries to distribute the work or the flow rather, you know, that can help. And definitely in terms of cities, what they're doing with the latest work is what they call "live, work, and play." So you build little regional centers around the city, so everybody doesn't have to go downtown.
Yeah, exactly. Exactly. For organizations, small companies or large, what would you say some key things are to put throughput accounting into practice?
I would say look at what your goal is and what your measures are, and obviously measure that. And then find how you're delivering that and just look at each of the different steps in that delivery of value towards that goal and just keep relentlessly focusing on that and just keep optimizing it.
Yeah, a good tool for doing this is a Kanban board. But that's really an absolutely critical measure of how the business is doing. And if you're not looking at that, looking at things like how long it takes to flow through this pipeline, how much is coming out the end, how much is in progress at a given point of time, how many defects or rework are coming out the other end, are the different steps along that process, getting back to bottlenecks, are they overloaded? Those kinds of measures, I think, are extremely important, and organizations really need to pay attention to them so they don't get into a difficult situation and have to do difficult things like lay off people.
Yeah. I mean, ultimately, it's a mindset shift. Everyone has to be rowing the boat in the same direction and know where they're going, and that takes a little bit of work, especially with larger companies.
Yep. And so the reward structure may need to change, you know, tying bonuses or promotions to overall system throughput rather than department-level metrics, right, is an important thing. I mean, if you're operating a small company, a small business, of course, maybe you don't have those types of constraints to think about. But as a small business owner, you have to think about how to tie your goals, the small business goals, to every employee that works for you so that they all understand what the goal is and the kind of work they need to do to achieve those goals. Right? Everyone needs to be rowing the boat in the same direction. All vectors pointing in the same direction.
I think you just brought up an excellent point. If you have departmental rewards, that end... I'll just pick a silly example. Let's say the software development shop is rewarded by the number of things they pass off to QA. It could do a wonderful job of passing all kinds of things off to QA, but QA is backlogged. So they're working faster, giving more stuff, but they're just making the jam-up in QA worse, and the product isn't getting up to the customers.
Yeah. So rewarding developers for the amount of what they do is not a good reward model. You want to reward developers for working together as a team with QA and with deployment so that you can actually put product out. It's a whole "logs in a river" model, right, the metaphor rather.
Yep. Yep. Doesn't matter if they have partial logs. If they don't pass QA, then there's a jam-up.
Exactly. Yeah. And then, you know, I think another reward mechanism or mindset shift would be rewarding, to your point, cross-functional problem solving, you know, QA and developers or sales and engineering. We have a client where we got leaders from each of the departments to work together as a cross-functional team. And the rewards, the gains they got were astounding. And one of the testimonials they gave was that it was the first time that sales and engineering worked together on something. And of course, now it's a matter of habit. You know, they do this all the time, but sales suggested something to engineering that would improve the usability of whatever it was that they were building. And this is a manufacturing firm. Right? So rewarding collaboration, cross-functional work. And again, if it's a big company or small company, it doesn't really matter. I used to run restaurants, and I used to have meetings twice a day between the front of the house staff and the back of the house staff just so that people would know what the specials were, how much inventory was on hand for the steak special or whatever it was. It was about cross-functionality and people being able to communicate and collaborate with one another.
I've worked in a number of companies that have problems with bugs. So the engineering was working well, QA wasn't catching things, and the customers weren't happy.
That doesn't end well, like in a restaurant. You have to have happy customers, or you're in deep trouble.
Yeah. Manufacturing is the same way. "Keep the line moving. We'll fix it later." That doesn't work out so well. Right? You got to have a huge parking lot where you're fixing, if you're talking about cars, where you're fixing the cars in the parking lot, your car parking lot's going to get jammed up. Or you could go and get the dealers to fix it. If they miss and it goes there to the customer, that's not good news.
So, you know, in terms of these logs through a river, while maintaining good quality at all times...
Yep. Absolutely. All right. So, throughput accounting definitely the way to go. Hopefully, this video provided some useful information on how you might use it, you know, thinking about it from a throughput perspective, what is it that you're working on and what are all the... You can't just think of, you know, like in my example, increasing the number of followers and subscribers. Not enough. Right? How does it help the system as a whole in terms of all those things, but also revenue? And because that wasn't there, we had to give that up. And it's not that we're giving it up. We're going to go back to it. It's just we need to rethink how we spend that money, that investment, and that investment just for followers and subscribers isn't enough. Even though that's great, I love all of you people watching these videos. But from a business standpoint, it isn't enough. Anything to add, Glenn, before we sign off?
Nope. I think you've covered it well. All right.
You too, I guess. I would add one thing.
Okay.
Throughput accounting is somewhat of a relatively new notion, and most organizations frankly are not using it. They're using cost accounting. So this will be a bit of a radical shift if you bring it to your cost accounting folks, but you can have both existing at the same time. You have cost accounting and throughput accounting. In terms of where to spend your money, look at throughput accounting and give attention to the bottom lines.
Absolutely. I like that. All right. Thanks a lot for watching, and we'll catch you in a month with our next episode on things Agile, XSCALE, business agility related. Thank you for watching. Bye-bye.
