Key Metrics for Agencies: What to Track for Better Decisions

Sei-Wook Kim (00:01.91) On today's episode, we're talking about tracking performance at your agency, what to measure, and how the right metrics can inform your decision making. Peter Kang (00:05.979) All right, so this is an interesting topic and something that you and I have talked a lot about. When it comes to tracking performance at an agency, it's very easy to go overboard, just start tracking everything you can and build a fancy dashboard with a lot of inputs and numbers. But one of the things that you and I have learned over time is there's actually benefit to having less, and also just whittling down a dashboard or whatever report you're doing to an easily digestible number of key performance indicators and metrics that help us make better decisions. Today I thought it'd be great for us to share what kinds of things we track in terms of performance across our Barrel Holdings agencies, and also why these numbers particularly, and how we think about decision making based on these metrics. So maybe we can dive in and share what we do. Sei-Wook Kim (01:20.888) Yeah, it's an important distinction to think about what metrics can be lag measures — so they report on what happened in the past — versus lead measures, which you're trying to project the future as much as possible. So let's start with the lag. First off is revenue. When we think about revenue, we try to break it down into recurring revenue versus project revenue for the businesses, and also how much of the revenue came from existing clients versus new clients. Peter Kang (01:57.241) Yeah, why is it important to make the distinction between recurring and project revenue? Sei-Wook Kim (02:03.053) Some of it is really distinguishing how much is this almost a stable base of revenue that's recurring every month that you've sold in literal retainers, versus the project revenue, which could be a little bit more lumpy. So you can see where you may have sold in a big project and that lasts for X amount of months, and you can see the fluctuations in the business. In general, we try to move towards having more of a base of recurring revenue so the businesses are less reliant on winning new business. All right, the next is gross margin. When we talk about gross margin, it's revenue minus all the costs to deliver the work — so typically in our instance, people costs, salaries, freelancers — and then setting a goal for what we want to see on that. That really reflects how profitable are the projects that we're running. Peter Kang (03:02.138) Yeah, I think it speaks to operational efficiency — do you have a capable team that knows what they're doing, able to get things done on time in a satisfactory manner for clients? Gross margin is a great example of reflecting that. And then on the flip side, it reflects pricing too. Did you charge enough for that project so you have enough margin at the end of the day to cover the cost of your people for that particular engagement? That's a good one to track. Sei-Wook Kim (03:31.724) The next is EBIT. So looking at, beyond gross margin, what are all the SG&A costs in the business? Anything from rent to software to travel and anything else that you use to run the business. It's important to separate those two because this is more about expenses that typically fluctuate a bit, but some of it is pretty consistent month over month. This is where you may make decisions on reducing SG&A spend based on the performance of the business. Peter Kang (04:22.811) Yeah, so a big one is when our teams went remote, for example — you get rid of an office cost, which is a fixed monthly cost, and that has a huge impact on SG&A, so the EBITDA number goes up. But on the flip side, you can be staffing up on the marketing side, maybe on the sales side as well, partnerships, all those things which are not related to the billable work, the delivery side. That can increase the SG&A expense line and the EBITDA might go down. But the bet when you're making those kinds of investments is that revenue might grow because you're able to increase the number of leads and increase revenue in the long run. It's a balance. A lower EBITDA and higher SG&A doesn't necessarily mean you're running a business poorly. It could also be reflective of certain investments you're making. Sei-Wook Kim (05:17.291) The next metric is utilization — looking at the percentage of the billable work versus the total capacity of each of the team members. We like to break that down also by discipline, so each of the discipline leads can say, how is my team doing against our goal? For utilization, you can bubble it up to a company goal, but what we do is we look at utilization targets per role within the company. Just because a director at the company has a lower utilization doesn't mean it's bad. It's really based on what is the goal for that person and are we on track or off track? Peter Kang (06:10.908) Yeah, and how this comes up a lot — people probably ask, what is a good utilization rate? How would we answer a question like that? Sei-Wook Kim (06:22.283) Depends on the makeup of your team. Generally speaking, we target 75% plus as a blended number across the agency. But it really just depends on your specific business. Some businesses do trend higher, some maybe slightly lower, but it could work out — if your bill rate is extremely high, for example, you could support a lower utilization number. Peter Kang (06:51.16) Yeah, I also think it has to do with the roles. Like you said, some directors might have lower rates of utilization — that's okay. But then at certain levels, maybe a junior or mid-level, you're probably hoping that they're at 90, close to 100% utilization. Sei-Wook Kim (07:07.947) Yeah. So let's shift to lead measures, looking at the future. Starting with utilization — forecasted utilization is one aspect of mapping out what do we think the team is going to be working on the next week, two weeks, month, couple months, as far as you can go with some level of fidelity. We'll call that forecast utilization. Is there going to be a cliff where you have 20 designers and 10 of them are going to be busy for the next week and then they're going to drop off? This is trying to paint a picture for the future. Peter Kang (07:53.891) Yeah, this is a huge one. Let's talk a little bit about what are some things that agency operators need to be careful about when it comes to forecasted utilization. A cliff is scary — if you don't have enough new business lined up, it's very possible that you need to act quickly and make cuts. But at the same time, as we've experienced in the past, sometimes that cliff is not always accurate. Things get pushed, other things emerge all of a sudden. Why can agencies mitigate against that? Sei-Wook Kim (08:35.373) Yeah, any of these things, it's making sure the data is correct before making any decisions on it. When you look at forecasted utilization, you might assume that a project is ending at a certain date, but you go and talk to all the project team members, the project manager, the account manager, and they'll tell you, actually no, that's not correct — it's going to run for another two, three weeks. And we're talking to the client about a post-launch thing that isn't reflected yet. So it's really important to make sure you understand the full story of what your data is telling you before making any decisions there. Peter Kang (09:15.125) Yeah, and that gets at a really key point in all of this: the numbers can't be taken purely as gospel. They're more signals and directional. If you're running the agency, you've got to dig a little bit deeper, have the conversations, and surface what's really going on. Sometimes, for whatever reason, there are gaps or delays in people inputting the numbers, or sometimes people just forget to move things — that can really skew the numbers to look a certain way. If you're not verifying that and having those conversations, you can make some really terrible decisions that can have a big impact. Sei-Wook Kim (09:59.467) Yeah, and the other thing is making sure you look out far enough. If you just look at the next couple of weeks, there might be a cliff where a lot of people are available. But two weeks after that, there's a ton of new projects starting up and you'll need those people. So if you make the decision to scale back, you might need those people in two, three, four weeks, and you've got to make the call on whether to keep them or shrink the team. Sei-Wook Kim (10:28.605) The next is pipeline. This is from the biz dev team, typically — tracking leads that go to calls or qualified leads, and then looking at the number of proposals sent both from a how-many-proposals standpoint and what's the dollar value, and then the win rate. All of these give signals for whether we're getting enough opportunities that will lead to new work in the future. Peter Kang (10:56.041) Yeah, and this is something we're looking at every week because it speaks to the strength of the business. Because it is the lead indicator, we can see if you have multiple weeks of zeros or close to zeros when it comes to leads, proposals sent, and of course wins — you know there's going to be trouble on the horizon. If I had to pick just one of all these, it'd probably be pipeline as the one that gives you the immediate sense of what's the shape of this business. And one of the things is not just new biz in general, but also account activity — the work coming from existing clients as well. That's another way that agencies grow. Not all agencies get new clients every week or every month, but they rely heavily on existing clients to come back with new engagements or extended engagements. Having that tracked is super important for decision making. Sei-Wook Kim (12:11.022) Yeah, and this is definitely a plan that needs to go through the whole team — understanding, in that example before, if there's a post-launch engagement or post-launch retainer from a project, that needs to be created as an opportunity in whatever system you're using to track it so that you know it's going to be coming in the future. So that's from a deal perspective. What we do after that is convert that into a weighted pipeline by month. For each deal in the pipeline, we'll say this is a five-month engagement and we think we'll deliver the work evenly across five months. Then, depending on the stage of that deal, we'll put a weighting on it. So we can map out what's our committed revenue and what's the future revenue, and with that, what do we think our next two, three months is going to look like? Are we on pace? This goes in parallel with the forecasted utilization, but from a numbers and financial perspective. Peter Kang (13:19.484) Yeah, and we could probably do a whole episode on this part because we're talking a little bit about booked versus recognized revenue — some accounting stuff here. But it's very important that we just underscore this. Let's take, for example, let's say we sign — just to make it easy — a $500K engagement. From a bookings perspective, you go, cool, this quarter we booked a half-million-dollar project, awesome. But from a recognition standpoint, like how it might show up in the weighted pipeline view, let's say this is an eight-month project — that's $62,500 a month over the next eight months. How should we think about that in terms of how it shows up on the weighted pipeline? Sei-Wook Kim (14:20.515) So we would break it down. And the other thing that's important is when that's starting. You could sign it now, but maybe it's kicking off in two months for whatever reason. In your future pipeline, you'll see a two-month gap, and then $62,500 for eight months beyond that, month over month. And yeah, you might see a gap in your projections, but as much as possible, having some kind of sense of when the work will be delivered is important because, to your point, a $500,000 project sounds great, but you don't know — is that a short engagement, a long engagement? Is that a 24-month engagement that you just signed? The timeline is very important. Peter Kang (15:01.131) Yeah, and this is super important because this all flows down to some of the things we've talked about in the past, like cash flow management. You can have a drought — let's say you just have low revenue months — but then all of a sudden you sign a ton of business and you think you're golden. But if those projects are not slated to start until months from now, you might still be in a drought and not be out of the red until those projects start. This is key for a lot of folks to think about: just because you sign stuff doesn't necessarily mean you're going to be in the black. Sei-Wook Kim (15:45.884) Yeah. And the other layer there — if you sign a lot after a drought, you may need to step up even more than what you have now. You may have to increase your cost basis to deliver that work in the future. Being able to visualize that is important. Peter Kang (16:01.171) Yeah, and that goes all the way back — the forecasted utilization might pop, and then all of a sudden your gross margins might go down. So this is all stuff that, as you get more comfortable with it — and these are just a handful of metrics, but this is the core stuff and there's a lot of nuance to it — as you get more used to this, you see how they play with each other, and then you can start to make decisions using these numbers. All right, cool. That's the core. We kept it pretty simple. It's not that much stuff, but obviously there are a lot of other things that we can measure and do. We keep it simple and we think about these others as secondary numbers. Maybe we could just quickly reel off some of the other things that we do consider and layer on, but that aren't looked at as closely as the others. Sei-Wook Kim (17:02.308) Yeah, and a lot of these also depend on the stage of the business and how large you are, the maturity of the business. If you're a small shop with a few people, some of this stuff may matter less versus a more mature organization. So I'll rattle off a few of the other ones that we look at. Accounts receivable aging — that's important to look at to make sure that you're actually getting paid by your clients. This is more for cash flow, to make sure that on paper, from an accrual standpoint, you may be very profitable, but if no one is paying you and they haven't been paying you for months, it's probably a flag that you should take some action on. That's an important one to stay on top of. Another one is client concentration — being sure to understand how much of your revenue is coming from which client. Do you have one or two clients that make up 50% of your revenue? That could be a risk if that one or two clients ends up canceling a project or churning, and then how do you replace that revenue in the future? Another is net revenue retention. That number is looking at — of the revenue that you did in one period of time, how much of that carries forward? Last year a client did X amount; this year, how much of that revenue can you expect, or does it grow or shrink? It's really showing whether clients are growing with us or churning year over year. Peter Kang (18:43.881) Yeah, and it's important to note that for one-time project-based businesses, this number is going to be lower to some degree. This is also why, going back to our initial revenue metric, if you can get clients on more of a recurring basis, then year to year or quarter to quarter you can have a higher net revenue retention number. If they spent whatever $25K in a quarter and the next quarter they're spending another $25K, you've basically retained them at 100%. And even better, if you've added some work on top of that, you could be over 100%. This is an important number that, especially as the business gets more mature and has some history to it, is going to be important to track. Sei-Wook Kim (19:33.519) Yeah, and related to that is looking at client retention from the specific number of clients as well, because net revenue retention can sometimes mask things — there could be one client from the past that increases a lot and masks other clients decreasing. The number and who the clients are are both important to look at. Sei-Wook Kim (19:58.148) On the employee side, looking at employee retention as well, tenure, and if you can track employee satisfaction through a variety of ways, that could be important to show the overall health and stability of the business. Peter Kang (20:17.316) Yeah, one of the things I think about is employee satisfaction — this could be different surveys you might send out or other measures. There are employee NPS-type surveys you could send out too. That's almost like a leading indicator. And then you could argue that the turnover numbers — what percentage of employees do you retain in a year — could be a lagging one. And then the ultimate lagging one is your Glassdoor reviews. You can look at some numbers there and obviously if that number is suffering, you know there are some issues you might be having on the employee side. Sei-Wook Kim (20:57.242) Yeah, definitely. The last one here is project and client profitability. We've talked very high level at the metrics, but breaking it down on a client level and project level — you may have scoped it for a certain level of profitability, and going through, doing a debrief after each project and saying, all right, this is where we landed from an actual cost and delivery standpoint. As you do more and more projects or engagements with clients, you can say, all right, who are our most profitable clients over a period of time? And make decisions as needed from there. Peter Kang (21:38.832) Yeah, and this is a good practice on a quarterly and annual basis. Quarterly helps you tighten up some of your ops and highlight where — hey, what can we learn from the project that was not profitable or where we struggled? And then on an annual basis, maybe looking at the clients and saying, hey, going forward, maybe XYZ clients are not the best fit for whatever reasons, and the numbers back that up. These are worth looking at, especially as a leadership team, maybe with certain extended team members as well, to really hone in the strategy of what you're going to do going forward. Sei-Wook Kim (22:18.721) Yeah, definitely. All of these numbers — there can be a lot of numbers to track. It really comes down to what you can actually look at and take some action on, or else you'll get to a point where you're just tracking too many numbers that aren't meaningful enough for you to affect any change. That's a really important thing to look at: review what you've been tracking, and do you still need to be tracking those numbers or should you be tracking something different? Peter Kang (22:51.087) Yeah, and a good litmus test for this is — on a weekly basis, or however often you look at it, whenever certain metrics just become a bit wallpapery, where you're just glossing over it and not really paying attention, just going through the motions, and there's no real conversation or critical thinking around that number — it's probably not that important and probably not worth tracking. It's almost like, all right, we should probably take that out and go back to the things where you almost have to feel like, hey, I need to look at those numbers, I need to know what's going on. There needs to be that pull, that sense of urgency around looking at those numbers. And it could very well be that some of these numbers you just need to look at on a much broader interval — once a month is fine, once a quarter is fine, once a year is fine. Distinguishing the cadence at which you look at these numbers can also give you a better sense and not overload you on a week-to-week basis. Sei-Wook Kim (23:57.682) Yeah, for sure. And I think in all of this — you mentioned this before, Peter — metrics shouldn't replace judgment. You have to look at this and decide, is it telling you the right thing that you're trying to make change on? And from there, it's your decision to take any action. Peter Kang (24:21.73) Yeah, awesome. We've shared a lot of stuff, but as with anything, we're always evolving as well. There are other metrics and other pieces of data that I'm sure will evolve to collect and look at more closely in the future. Hopefully this has been helpful to share how we look at our agencies across the portfolio. Thanks for joining us. Take care. Sei-Wook Kim (24:46.593) Thanks.