How To Turn Your Agency Into A Profitability Machine

About this episode:

In this episode of the Agency Spark Podcast, Sara interviews Marcel Petitpas on how to turn your agency into a profitability machine. They discuss the agency profitability flywheel, creating accurate estimates, value vs. hourly pricing, the agency pricing quadrant, tips for tracking time, client reporting and more!

Marcel Petitpas is the CEO & Co-Founder of Parakeeto, a company dedicated to helping agencies improve their profitability by streamlining their operations and reporting systems.

He’s also the fractional COO at Gold Front, an award-winning creative agency in San Francisco working with brands like Uber, Slack, Keap and more. As well as the head strategic coach at SaaS Academy by Dan Martell, the #1 coaching program for B2B SaaS businesses in the world.

In his work as a speaker, podcast host and consultant, specializing in Agency Profitability Optimization, he’s helped hundreds of agencies around the world improve profitability and cash flow in their business.

When he’s not helping agencies make more money, he’s probably watching “The Office” or “Parks and Rec” on a never-ending loop and eating breakfast foods for every meal of the day.  

Core ideas:

  • Unpacking the Agency Profitability Flywheel & the Agency Pricing Quadrant
  • Creating accurate estimates by understanding where your risk is
  • Understanding when to use value-based pricing versus hourly-based pricing

Show notes:


Sara: Welcome to the agency spark podcast. This is your host, Sarah Nye. And today on the show, I have Marcel Pedopaw CEO and founder of [inaudible] a company that helps digital agencies improve their profitability. Marcel is also a speaker podcast, host and consultant specializing in agency profitability optimization. Welcome to the show.

Marcel: Marcel. Thank you for having me, Sarah. It's great to be here.

Sara: Of course. I want to start the show diving into your background story a bit. So what led you essentially to where you are today?

Marcel: First ever business was an agency like a lot of people, and we were actually selling real estate services. We're doing VR for real estate agents. That was back before Matterports and like apps on your phone could take 3d images of a house. And that was the first time that I got in touch with how hard it is to run things profitably, especially when you're selling to people who are frugal and don't have a lot of price sensitivity. And I actually ended up walking away from that business because I couldn't make the margins work to scale. So that's where I became intimately in touch with this problem set fast forward, a few years, many other startups and failures later, and a friend of mine, Jared called me up and said, Hey, I'm sick and tired of building spreadsheets in my business to answer these simple questions, there's gotta be a better way to do it. And that's when we started bear Quito and decided we were going to solve agency operations for good.

Sara: I love it. And one of the elements that you created as part of this is what you call the agency profitable profitability flywheel, which is a four point feedback loop that builds data driven processes in profit improvements within an agency. So I will be honest. I don't know much about your flywheel yet, but I would love to dive into what our to start. What is the flywheel? What are some, what are the four key pillars?

Marcel: Yeah, that's a great question. I know that there's some linguistic gymnastics. You had to jump through just to tee that up. I appreciate you giving it a good try. Um, but it's complicated stuff, right? And at the core of it, our POV is that it's hard for agencies to answer simple questions. Are we making money on projects? Do we need to hire people? And when are we scoping things accurately? When we think about operations as an agency, really at the end of the day, you need to be able to make good predictions about client work. How much is it going to cost for you to deliver? What kind of time is it going to take over what time horizon, what kind of skills are required within that? You need to be able to plan that out because if you want to have good cashflow, good profit, if you want to hire people at the right time, do good resource planning.

Marcel: All of those systems in the operation space rely on those assumptions. And so when we think about the flywheel, it's about creating a system so that every time you do a project, that system gets better and designing it in such a way where it doesn't rely on any one person in the organization to happen. It's a process and that process can be owned by the team and it can be baked into your system for how you do projects. And so there's four basic elements to it. It starts of course, with estimation or scoping. So making sure that you have a system for how you come up with assumptions for client work and the two elements that are most important about that are the structure of those assumptions. And I don't mean the font that you use in the document, but more so what is the data schema?

Marcel: Is that a client with deliverables and within each deliverables or tasks, or is it a contract that has perhaps different phases? And then within each phase you have different roles. So what is the structure of the information, the structure of the assumptions, because that's going to impact how you need to set up all the tools for tracking time, managing projects, doing finances so that you can actually tie those pieces of data back to each other and answer the question. Did we even scope this right in the first place? And then the second thing is the methodology. If the way you come up with those assumptions is different. Every time, then you're going to start to run into issues where you can't compare one estimate to another. Therefore you don't have a foundation that you can iterate on and improve from. So it starts with estimation.

Marcel: Then it goes to tracking actuals. So can we actually keep track of what actually happened? And then compare that to what we thought was going to happen and see where the gaps are. Then we get into reporting and meetings. So do we have a cadence on which we're reviewing what happened versus what we thought, identifying outliers, and then having a conversation with the team about why was this project a slam dunk and why was this one, a dumpster fire? And what can we learn from those two things? And then the last piece of course is process improvement. What did we change about the way that we scope the way we deliver, the way we structure teams, the way we interact with clients so that we can close those gaps over time. Because the piece that I think a lot of people overlook is maybe your estimation is bad because you're bad at estimating and you don't have good data, but most of the time it's because the thing you're trying to estimate just isn't that well-defined. So if we can improve the definition of that thing over time, then inherently, it becomes a little bit easier to scope because it is an in and of itself is more clearly structured.

Sara: Yeah. That makes a lot of sense. Let's start with someone that's in the very beginning. And so maybe they're stuck in spreadsheets or feel like they're making stuff up at this point. They haven't done a lot of S accurate estimates. Do you have any advice? Because obviously it sounds like it gets, obviously if you set the, put the system in place, it gets easier over time to get better into improve. But if you don't have the experience yet, how do people start with making estimates to be?

Marcel: Yeah. So I think the advice I would have to, even the freelancer, that's doing this, that has aspirations of hiring other people to do the work. It's just start with something, get a standard format in place and think about it in terms of what hats am I wearing. Usually the best data schema is think about it in terms of roles. So if I was going to hire people to do this work, what are those people? And then you structure your estimates so that you're thinking about, okay, let's simple example. I have to build a website. How much time am I spending being the project manager, emailing the client, scheduling a meeting, keeping things up to date in a sauna. How much time am I spending being the web developer? How much time am I spending being the designer and how much time am I spending being the copywriter?

Marcel: How much time am I spending being the strategist? And just start to track your own time in those buckets and do that a few times so that when you get to a place where you think, okay, I think it's time for me to create some leverage here. You have some clarity on, okay, over the last five websites that I built on average, it took me 30 hours of design time. So now you're, de-risking that investment significantly. You can set a more clear set of expectations for that designer. You can predict your own costs better. You can actually figure out ahead of time, if you can afford that investment or not. It takes a lot of the ambiguity out. So it's just about coming up with a simple structure and then tracking your time and comparing that to what you thought was going to happen. It doesn't have to be any more complicated than that. That would be the starting point, I think for everyone.

Sara: Yeah. I love that a lot. When I'm thinking about time tracking in the past is I'm just tracking my time. I'm not necessarily saying I'm tracking it as I'm doing client work for this client. I'm not necessarily saying I'm doing this specific design work or development work or whatever it might be. And so I love the idea of even getting it narrowed down more, as you said, then when you're hiring in the future, you know how much time you need from a designer or someone like that. Another question I get often when people are just getting started in the consultancy role is how do I charge for my time spent? How do I figure out what my hourly rate essentially?

Marcel: So great question. There's two things to consider here. The first is just the foundation of what needs to be true for your agency to be scalable or for what you do as a freelancer to actually translate into being something that you can hire people to scale. And when we think about this, the most important benchmark, the first thing that you have to figure out as a consultant or as a freelancer is is there enough gross margin and what I'm doing for a business to get built around it. And this gets overlooked. So often we get people reaching out to us all the time that they think that because we're the profitability people. So they come to us and say, Hey, we're not profitable. And then they think it's because they're spending a little too much on their legal firm or their accountant, or they're spending too much on software.

Marcel: And it's no, the reason is because your services aren't fundamentally profitable. You don't make enough money. Every time you sell a new client, to be able to afford, to spend a reasonable amount of money on overhead and still have enough money left over at the end of the year to compensate you for the risk of running the business. So that number should be on a project basis, targeting 50 to 70% gross margin. Now let's break down what gross margin actually means. There's two numbers you have to know in there. The first is your revenue, but it's not just your gross revenue. It's your agency gross income. So what is the difference between revenue and agency gross income? We want to strip out all the revenue that doesn't belong to us, or this is not our responsibility to earn. So examples of this would be templates or stock images, external vendors that we have to hire ad spend print budgets, right?

Marcel: That money is flowing through us into another entity. What we want to do is be able to measure when we have work or revenue that we have to spend time to earn, how efficiently can we do that? How little time can we invest or in the same amount of revenue, that's where we create margin. That's where create scalability for the business. So start with calculating your AGI, taking your revenue, stripping out all the pass through, and then you want to look at your cost basis. And the cost basis is mostly much time. Did it cost me to do this thing? And then what you can start to do is model early on, okay. Roughly how much would it cost me to hire somebody else to do this? You don't need to be in that place to do that math. You can just get on Upwork.

Marcel: You can get on free up. You can get on working, not working. There's tons of places. You probably have a network maybe through duct tape marketing, where you can connect with some other people who do this kind of stuff and just say, Hey, what's your hourly rate. And then you can start plugging in, okay. If I pay a designer $50 per hour and it takes them 10 hours, then there's a $500 cost. There, you divide that by the margin you want to have in this case, let's shoot high. Let's say 30% that tells you what you need to be charging for that person's time or what the hourly rate needs to be. So set up your pricing, set up your projects with that modeling in mind so that when you get to a place where it's time to scale, then you don't run into that bottleneck of realizing, oh, like I'm not charging enough to actually hire people. And then of course, if you already are hiring people, then make sure you've got enough margin there. So that have enough leftover to run the business on the backend. So it start with margin. There's a whole other discussion here about thinking about risk and how risk changes as you get better at estimating projects and how that should impact your pricing strategy. But I'm going to plant that seed and come back to you to see if you want to go down that rabbit hole together.

Speaker 2: Yeah, I think let's go down the rabbit hole.

Marcel: You opened it up. Yeah. So I think this is a concept that not enough people are thinking about, but we have, and there's a video that I can share on this. If you just look up agency pricing quadrant on our blog, you'll see the post on it. But I have a way of thinking about pricing, where you have to consider two elements essentially to figure out the right pricing strategy. And I think in the industry, we get bombarded a lot with billing for your time is crazy. You should never bill by the hour, you should only do value-based pricing. And the reality is that that's not necessarily appropriate for everyone. And the reason for that is not everybody has the same amount of risk and the type of work that they do. So there's two axes to figuring out what the right pricing model is for you.

Marcel: The first is value, how much value is there and what you do value is generally determined by two primary things. The first is how commoditized is your service, right? So are you doing general website design development, general graphic design, but you're competing against the design pickles $400 a month for unlimited graphic design, or are you a graphic designer that helps B2B SAS companies and enterprise visually communicate their complex systems for how they transmit data from one system to another. It's a very niche, very specific, much more high value, same service, but positioning creates value. The second is relative value to the client. So what's a logo worth well to a hundred thousand dollar business. That's selling pizza, maybe locally, not that much, a couple of hundred bucks to an airline that has to spend $50 million to put it on a fleet of airplanes. That logo is probably worth a couple million dollars.

Marcel: So there's a relative value difference, depending on who the client is, you have to assess those two things. Who do you sell to and what do you do for them? And how many alternatives do they have? That'll help you figure out are we at the bottom or at the top of the value scale? The second axis you can think about this going horizontally is how much risk is there in the engagement. And very simply put risk comes down to how accurately can we scope this thing? If you build WordPress websites on Elementor for lawyers and you've done a thousand of those websites, you could probably get to within 10% of the hours for every single task on autopilot, that's super low risk. Whereas if you build custom software on a brand new development language, that's not very well-documented oh, and you're building something that's never been built before.

Marcel: And the specs can't even really be clearly defined, but that's extremely high risk. It's basically impossible to figure out what that's going to take. So understand how much risk there is. And this is going to change of course, based on where you are in the life cycle of your business. You might have a service that can become low risk, but it's not there yet because you've never done it before. Or you've only done it once or twice. Maybe you weren't tracking your time or maybe your time tracking data is not useful because it wasn't structured properly. Over time. You can start to move down and lower the risk, but you might be starting at a high risk place. So that's your quadrant risk along a horizontal axis. And then vertically, you're thinking about value. If you're in the bottom left-hand corner, low value, high risk, probably just bill by the hour, what you're doing in that situation, you're asking the client, Hey, let's share some risk.

Marcel: You want me to tell you how much it's going to cost, but I can't do that. I don't have the information. You're asking me to do something that's like really risky. That has a lot of variables. So let's do it by the hour. And then you get, you're getting the value. And in that case, you just make sure you have enough margin in what you're paying somebody versus what you charge the client. And you try to make sure you're not eating that many hours. You're playing the utilization game when you're billing for time and materials is just how much time can I spend doing stuff that I can build to the client and then making sure that I set good expectations. So they actually pay me for those hours and don't push back. If we move over to low value, low risk, what we're doing there is doing flat fees.

Marcel: The reason for that is we can arbitrage the fact that we have low risk to take the risk off the table for the clients, and ideally become really efficient so that we can create a higher hourly rate without scaring the client off. Because Sarah, if you came to me and said, Hey, I need a website. And I told you it was going to be $500 an hour. You'd probably tell me to go take a hike. But if I told you, yeah, no problems there, I can get you a blazing, fast SEO, optimized, high converting website, specific for your niche, and I can get it to you in two weeks. And it's just going to cost you $5,000. You'd be like, where do I sign those two things to me, the seller might be the same. I might make 500 bucks an hour because I have a great process.

Marcel: And I know exactly how to do that in a systemized way. So I arbitrage the little risk. It gets you certainty. I get myself efficiency. I win by doing flat rates, moving up to the upper left-hand side of the quadrant. This is high risk, but also high value work. This is where we do something called abstracted time and materials. This is what companies like MediaMonks does for example, and they scale to a thousand plus employees building complex websites for enterprise doing this model. You're not saying we're billing by the hour operationally. You're modeling it that way. But to the client, you say, Hey, you need a very skilled cross-functional team to get this thing done for you. We're going to basically give you that team for $20,000 a week. And we estimate this is going to take between 12 and 18 weeks to get this deliverable done for you.

Marcel: So you're baking your margin in, you're taking you're operationalizing the risk. You're sharing it with the client, but you're using the fact that it's high value to abstract the conversation away from the hour. So you're not talking to them about it's 1 75 an hour. You're saying it's 20 grand a week. And here's all the skill sets that you're going to get in that team. And here's what we think it's going to take to get to that end goal. And by the way, client, the difference between 12 and 18 weeks is how much you've got your stuff together. If you're late getting us feedback, if you're not engaged in the meetings, if you're not turning things around, then it's going to take longer and you're going to eat that cost. So that's high value, high risk. Finally, the quadrant that everybody wants to talk about is high value, low risk value based pricing, right?

Marcel: Taught. But as you can tell, it's not appropriate for everyone. And it's maybe something you have to work towards as you lower your risk. But in that case, you're saying, okay, client, what is this worth to you? This conversion rate optimization, this website redesign, oh, you do a million dollars a month in e-commerce sales. We think we can improve your conversion rate by 2%. That's $2.4 million in extra revenue for you this year. Would it be fair to pay us 10% of that upside $240,000 for this project? Yes. Okay, great. Now the price is completely abstracted from ours. We're basing it on value. We still want to make sure that we can scope it and make sure we have that 70% margin, but this is where we can get to 80, 90% margins because the basis for our pricing is what is it worth to them? It has nothing to do with what it's costing us. And we can really start to arbitrage, both certainty and low risk and arbitrage high value to really stretch that gap. But you can see that's a very advanced level of pricing and you have to have positioned yourself for high value and you have to have created the processes to lower your risk in order for that to really be a viable pricing model. So that was a lot, but hopefully it's insightful for everyone that was listening.

Sara: And now a word from our sponsor term. Again, states are proposing laws that will allow consumers to Sue businesses anywhere in the U S for not having a compliant privacy policy. Any website collecting as little as an email address on a contact form should not only have a privacy policy, but also a strategy to keep it up to date. When laws change, learn more about how term mageddon can protect your business at [inaudible] dot com. Now back to the show. No, it definitely was insightful for me. So I appreciate you going down that rabbit hole. Um, we'll definitely include a link to your video that, that goes into that in more detail in the post. So to me, obviously what I've been hearing in preaching a lot is the whole value based pricing. That's what we try to do in our business. And so I really love though how you laid it out to really simplify.

Sara: It really matters on where you are at as a business, but also where your client is that as well. And so I think that's such a smart way to visually lay that out. So thank you for sharing. My pleasure, really. I have a few more minutes left, but I do want to dive some of the other elements that you talked about. So let's say someone's estimated a project they're starting to work on it. They haven't really nailed tracking yet. So how do you have any tips on how someone can be successful in terms of tracking time?

Marcel: The biggest thing is making sure that the schema, the structure of the data looks the same as the estimate. That's probably the easiest thing to overlook anyone that's listening. Chances are, if you're anything like all the clients that we've worked with, when you hold up an estimate and then you go into your time tracking tool, everything looks different. It's called a different thing. It's designed over here and it's wireframe V2 task in the time-tracking tool. So if you just want to answer the question, did this take the amount of time that we thought you got to spend a lot of time cleaning up that data to get that answer, and you've got better stuff to do with your time. So you're just not going to do it. So you want to make that as frictionless as possible. And that just comes down to thoughtful data schema design.

Marcel: And this is the kind of stuff that we do as a consulting firm is like, what should that data scheme would be? What is appropriate for your business model? You don't have to overthink it, just try to make sure they match up. And then it's easier to get the answers. That's really the core of it. It doesn't matter what tool you use. It doesn't matter if you're doing time sheets or if you're using a resource plan as the basis for time tracking. And it doesn't have to be a hundred percent accurate. Don't over-optimize on accuracy, optimize for structure and basically compliance. Try to get as much time into the tool as possible. Don't worry that it's not exactly right. That's not important. It'll be directionally helpful. And really at the end of the day, that's what you're looking for. Yeah,

Sara: Absolutely. So then when someone has a tracking figured out they're ready for reporting and review any advice around their,

Marcel: Yeah, set a cadence that's appropriate for the rate of change in your business. So if you do a lot of short projects, you probably want to have these conversations every week. And if you do a lot of short projects, you probably want to just, instead of trying to do a retroactive at the end of every project, just have a meeting at the end of the week. And you can look at, okay, here's 10 projects that we finished this week, which ones were awesome, which ones were crappy. Let's talk about the outliers. But if you have big projects that lasts a really long time, maybe you just do a retroactive at the end of every project. So don't get too caught up in what is the right cadence. That's okay to customize that for your business and the rate of change, but make sure that at the end of an engagement or at the end of a time period, where you have had enough change, you can look at.

Marcel: And the most important thing to look at is essentially what was the furthest off from the estimate, either in a good way or in a bad way, where did you go way over? Where did you go way under? And let's focus in on those things and try to extract insight from the team on why those things happened so that you can make changes to the process and tighten up the system over time. That's really as simple as it is, and you don't have to overthink it, just start having the conversations and the team will work it out.

Sara: Yeah. And then that takes you directly into the last pillar, which is process and improvement. So you take that and I'm guessing yes,

Marcel: Go for it and make the changes. Now here's the key, here's the key. And this is really important in that second part, the meetings try to stay away from pushing answers onto your team. It's really important that the team is the one that is surfacing the ideas cause they're on the grassroots, like they're in there doing the work. So they'll probably have better ideas than you will. Um, but even if you know what the answer is, try to get your team to come forward with that because when they came up with it and you go, wow, Sarah, that's a really amazing idea. We should change the way that we ask our clients for the copy for the website. Because if we do it the way that you did it on that client, then that'll save us a ton of time. Would you like to take care of, take the lead on implementing that change? Then you're way more likely to not only go out and actually do that, but follow it, maintain it, hold your peers accountable to it. So we want to create buy-in. We do that by inviting the team to get involved in that conversation and really pulling them into surfacing the ideas that we ended up going forward with that.

Sara: Yeah, I love it. Awesome. Thank you so much for sharing everything today. If people listening would like to connect with you online, where can they find you

Marcel: For those that are listening that want a set of free tools to implement everything that I talked about without spending any money on us or anything else you can head to pare forward slash toolkit, where I've put together a series of training, videos, templates, checklists, cheat sheets, to help every agency in the world get better at managing their profitability. So I encourage you to go and check that out. I will spam you with emails after you can just unsubscribe, if you don't like them and for anything else, find me on LinkedIn. I'm always happy to nerd out and chat with you about profitability.

Sara: Awesome. Thank you so much. And thank you all for listening to the agency spark podcast. We will see you next time.


Like this show? Please leave us a review on Apple Podcasts here!


This episode of the Agency Spark Podcast is brought to you by Termageddon, a Privacy Policy Generator. Any website collecting as little as an email address on a contact form should not only have a Privacy Policy but also have a strategy to keep it up to date when the laws change. Click here to learn more about how Termageddon can help protect your business and get 30% off your first year payment by using code DUCTTAPE at checkout.


You may also like