The Data Behind the Deal: B2B Distribution Through an FP&A Lens with Ryan Cosby

Resources Mentioned:

• Salon Service Group — Ryan's company
• Qlik — BI and analytics tool used by SSG
• Vertex — Sales tax automation platform
• Imbibe Magazine — Ryan's go-to for drinks culture, recommending it as a great example of a media brand quietly and successfully reinventing itself.

Full Transcript 

Aaron: Welcome back to the B2B Uncut Podcast. I'm your host, Aaron Sheehan. We're still sponsored by OroCommerce, and we have a first-time experience today. 

I'm joined by a longtime personal friend, Ryan Cosby. You might be asking yourselves: why do I want to hear what one of Aaron's personal friends has to say about distribution and manufacturing? Ryan, why don't you introduce yourself and explain why it's relevant?

Ryan: Sure. I'm Ryan Cosby. I'm in financial planning and analysis for a beauty products distribution company based in Springfield — though we cover pretty much all of the Midwest and Southeast United States. I'm a bit of a recovering banker; I spent about 10 years in commercial lending and analytics. I've only been in FP&A for about two years, but I'm bringing a broad knowledge of different industries into a more focused setting. Rather than moving deal to deal like I used to, I'm now focused on one company — and it's really fun.

Aaron: Tell us a little about the company. What do they sell — you mentioned beauty supplies?

Ryan: Mostly hair care, with some skincare as well. And it's very niche — we actually supply to other distributors, and those distributors are smaller. Think about it this way: when someone goes in to get a haircut, the salon has a "back bar." After cutting your hair, they'll shampoo, condition, and apply product — and that's also a sales pitch, because in front of the store they typically have a retail section where clients can purchase. We sell to those salons at distributor prices, and then they sell to their customers. So we're B2B, selling to other businesses.

Those businesses range from large chain salons — we're not talking Great Clips, our products skew higher-end — to maybe a regional chain with 10 locations in Texas offering full spa services, all the way down to a single stylist in a small strip center with one chair and a retail display. We try to be a one-stop shop: combs, shampoo, color, conditioner, the full care line. But we keep our brand offering narrow so our salespeople can be as knowledgeable as possible in those specific areas.

Aaron: Sounds like a very traditional rep-led distribution business. And if you're watching this, you can see that Ryan clearly uses his own product.

Ryan: I do, yes.

Aaron: One of the things I like discussing with folks in the industry is digital orders and eCommerce. You guys have an eCommerce site, and you sell through multiple channels. It's often said that digital orders are cheaper to serve — less infrastructure, fewer staff touchpoints, potentially lower cost of sales. As an FP&A person, you have to prove or disprove that. Walk us through how you model cost-to-serve on a self-serve order versus a rep-assisted one. Without getting into the private financials — is this even a knowable thing?

Ryan: To an extent, yes — we wouldn't be much of an accounting department if we couldn't track those costs. But a lot is intertwined. The data shows that our most valuable customers — on a per-transaction basis and overall — are those who shop across all our channels: brick and mortar, through a rep, and online. So we see it as interconnected. That said, you're right that eCommerce is our most efficient and cheapest channel. We've put a lot of money and research into it, and we have a lot of data to back that up.

Going back to who we serve — our customers have to have a cosmetology license. You can't just walk in off the street and buy from us. When I cracked earlier that you couldn't buy from us, Aaron, that wasn't just a dig at the baldness situation — it's a real supply chain requirement. Every customer needs a cosmetology license and an account number. And they're used to that; if they went to any other beauty supply, they'd have to do the same thing. That gives us a lot of data on individual buying habits, because no one can buy anonymously.

I can definitely do longer chunks! The length isn't a processing limitation on my end — I was being conservative. I'll do bigger sections going forward.

Aaron: So the fact that customers have to give you an account number — that gives you a lot of data on their buying habits. How does that plug into your eCommerce presence specifically?

Ryan: Because customers can't buy anonymously, our eCommerce experience can be really tailored to them — what we want to show them, what they've bought before. Our reps are also very tuned in and know exactly what each customer is looking for. So while eCommerce is the cheapest channel, I don't think we'd have nearly as much data if that were our only channel. It's the rep talking to the customer, pushing certain brands, combined with what that customer is browsing online, that gives us the full picture. Each channel informs the others. Cheaper, yes — but without the rep or the store, we're just not as in tune as we want to be.

Aaron: You implied that reps can actually see what their customers are doing online. Did I understand that correctly?

Ryan: Yes, though not down to individual clicks — that would be too granular for a rep managing hundreds of customers. What we do give them is lost sales data. If a customer put something in their cart and didn't complete the purchase — whether it went unfulfilled or they just abandoned it — we flag that for the rep. We want to give them every tool to make the sale. And part of that is making sure reps aren't scared of eCommerce thinking it'll cut into their commissions.

Aaron: That's a critical point, and it's something that has come up on this podcast more times than I can count: please, if you're listening, comp your reps on eCommerce orders if you want eCommerce to succeed. It sounds like you do that?

Ryan: Absolutely. We commission reps on eComm sales from their customers. And they're still buying even with that in place. The best use we've found — especially for our app — is having the rep sit with the customer and walk them through placing an order together the first couple of times. There used to be a back-end app for reps; now it's a front-end customer app, and it's so much easier to just sit there with them and do it together. Then the customer starts ordering on their own. It's a natural on-ramp, because they've seen the rep use it, it feels familiar, and it feels safe.

Aaron: That's fantastic. And honestly, we didn't rehearse any of this — you're hitting a lot of the greatest hits for how we position Oro. My content team is going to be thrilled. Let's pivot to something a little more dry: pricing. Pricing comes up constantly as a complexity problem. I was at a B2B trade show a few months ago on a panel, and someone from the audience asked what's the one thing distributors can do to improve their lives. I said: smooth out your pricing. And we ended up talking about it for 20 minutes. In a rep-heavy business, what you often end up with is a lot of very custom pricing sheets — bespoke deals a rep needed approval on to win an account, and then once it's won, that pricing just... stays there forever. Over time you end up with a tangle of one-off agreements, and when you try to model margin, it becomes really complicated because there's no clean, streamlined input. Does that describe SSG? And has that complexity cost you anything — margin leakage, operational overhead?

Ryan: You're describing exactly how we operate, and it's still that way today. Reps are still going out, making deals, asking for approval, and we end up with small individualized agreements per customer. They're not completely piecemeal, but they're pretty customizable per agreement. And tracking that down at the end of a month when your margin is a little off is really hard. Being a distributor, our margins are pretty steady — so if something's off, we're freaking out trying to track it down. We use Qlik as our BI tool, and it does a great job of drilling down into discounts by customer. I can pull up a list of discounts for the month and drill into them, which I love. The problem is, you can drill really deep and maybe find $6,000 of a $180,000 delta. You can make inferences about the rest of the portfolio, but you don't really know. And part of what makes it hard is that a discount is a discount — whether it's baked into a contract or triggered by a large order volume. Tier-based pricing adds another layer. When we have months where margin is up or down, a big culprit is often a large customer order that kicked in a volume discount — but it's hidden inside what just looks like normal sales. Top line looks great, and then you dig into costs.

Aaron: How does that work on the eCommerce side? If a customer has a bespoke pricing agreement, does that get reflected in their online price, or do they just call their rep?

Ryan: Luckily, yes — it is reflected online. We're nimble enough to code those agreements into our ERP, and they push up to the eCommerce site automatically. But there are nuances. Customers have a back bar — product they use on clients — and a retail section where they sell to clients. Some of that product is taxed, some isn't. So giving discounts gets complicated, because it's not as simple as "this customer gets X% off everything." A liter of shampoo, for example, is typically a back bar item, but a client could theoretically buy one too. So not every product is used the same way by every customer, and you can't always tie the discount rule cleanly to the purchase intent. It gets a bit muddy.

Aaron: That sounds like a job for Vertex.

Ryan: Exactly — we use Vertex for sales tax. We used to have a full-time person managing tax rates manually, updating the ERP and reconciling at month-end. I think it would take two people to do that today. There's just no way to do it manually at this point. This is why software exists.

Aaron: Speaking of software — you mentioned your ERP, Qlik as your BI tool, and I also know from knowing you personally that you and your wife are Excel people. It's practically a love language in your household. So: commerce obviously happens in the ERP at some level — every order flows through it. It doesn't happen in spreadsheets or in Qlik. Walk us through the data handoff between all these systems, and how clean — or not — that handoff actually is. I know you're sometimes staying late at month-end dealing with this.

Ryan: Our ERP is pretty tailored to what we do. It's an older system — System 2000, very niche — but it does exactly what we need it to do. The challenge is getting data out of it, because reporting just wasn't a priority when it was built. That's where our BI tool comes in. We upload from System 2000 into a data warehouse, which then pushes tables into Qlik. That makes it easy to view and easy to build dashboards from. That handoff is actually pretty clean. Every morning I'm getting reconciliation statements between the data warehouse and the BI tool, comparing them to the dollar. If those match, we're good.

The harder part is on the accounting side, which we still pull directly out of the ERP as spreadsheets. So now you've got your chart of accounts operating in Excel, your BI tool also exporting to Excel, and your sub-ledger — which I'm using to reconcile at month-end — is nicely structured and detailed. The problem is they don't always map cleanly to each other. The chart of accounts can be vague where the BI tool is very granular. You can filter out one little subset and accidentally exclude a bunch of relevant data. So you really have to know both systems and understand what each one is actually telling you. A big part of my job is knowing that "this account" on one side really means "these 8 sub-categories of income" on the other.

Aaron: And I'm sure you have customers that show up under 6 nested DBAs because of acquisitions or rebranding over the years.

Ryan: A hundred percent. COVID made that especially messy. We've been around since the eighties, and a lot of salons that shut down during COVID came back under totally different names. They want to feel like you know them and can pull their full purchase history — but bridging that gap in the data just isn't always possible.

Aaron: Walk us through what month-end actually looks like for you, and any advice you'd have for someone trying to figure out how to reconcile transactional data against a chart of accounts — especially when there's compliance and tax reporting layered on top. Is that a solo effort, or is it a team sport at SSG?

Ryan: It's a team sport, and honestly it's gotten so much better. Our BI tool has only been in place for about as long as I've been here — a little before — so everyone is still learning it as we go. I just had the advantage of coming in when it was already there. I see it as part of my job to get people comfortable with it, because I don't want to be the only one who knows how everything works. We don't need a wizard's tower situation. We want everyone involved.

Sometimes I'm sitting in meetings just to catch a keyword and say, "Hey, when you say that, what you're really talking about is this." That's a real part of my role — translating between business language and data language. And it requires the whole team to stay curious. You can't just assume you know what something means; you have to ask. It's not always me they need to ask — sometimes it's our business analysts, sometimes our IT director. But we have a great team across accounting, purchasing, and other departments who are curious enough to want to understand how their work connects to the BI tool. Once everyone is operating in the same system and speaking the same language, it becomes a lot easier to communicate across functions.

At month-end, my job is to translate what the data is saying for the broader team. If we're five margin points off, I need to be able to tell accounting what that means, and then go to the sales team and ask: why did we give that discount? Was it a large order that's going to repeat next quarter, or was it a one-time thing? That back-and-forth between finance and sales is a big part of the close process. And we are getting better at getting accounting to be able to go directly into the BI tool and interpret it themselves, rather than always coming to me.

Aaron: So at month-end you're exporting data out of the ERP into the data warehouse, pushing it into Qlik, matching things up, chasing discrepancies, and forensically reconstructing why margin moved the way it did — whether that's at the product line level or the customer account level. One thing we talk about a lot at Oro right now is where AI fits into the reporting and reconciliation piece. You and your team are essentially doing pattern matching — taking two different forms of structured data and trying to uncover the truth. Are there any AI pilots happening at SSG, specifically around FP&A workflows?

Ryan: The value is definitely there, and actually we just came out of a meeting on this with Qlik. The thing is, we have so much data in that tool — we're talking invoice-level, SKU-level detail, everything about a customer and a transaction. The challenge is that it's hard for someone to fully grasp what they can get out of it, especially a sales team member or a manager who needs to stay at 30,000 feet. Everyone in the company has Qlik access — all the way down. That's great, but not everyone can build a financial statement from it. What someone in purchasing can do is see exactly what Sally's Salon bought yesterday versus what they bought in 2023.

That's where I think AI can really help. Right now, if someone wants a custom dashboard, they come to me or our business analyst or IT director and put in a request. With AI, someone could just ask: "Is there any correlation between sales of this brand and these 20 customers?" — and get an answer immediately, which then cues us to build a dashboard, or eventually has AI build it directly. Because once you're in the custom dashboard business, you're always in the custom dashboard business. And people's willingness to keep using a custom dashboard five minutes after it answered their one question is very low. You end up with a graveyard of abandoned dashboards.

Aaron: That's so true. Are you sure you don't work where I work?

Ryan: The vision is more of what I'm already doing in those meetings — catching keywords, connecting dots — but democratized. People querying an AI that knows our data and asking, "Is this in here? Is there a correlation?" And it either confirms it, rules it out, or points somewhere else in the data. That's where we see the value. And I think it removes the last barrier to buy-in, because people won't have to feel nervous approaching whoever holds the keys to the BI tool. It's a faster, more approachable path into the data.

Aaron: And this is coming through Qlik specifically? Or are you looking at a standalone AI tool — a corporate Claude or Gemini account?

Ryan: It'll be through Qlik — we'll be using their agentic AI for that process. It's essentially turning on a feature. Margins are thin in distribution, so we're keeping licenses lean and building it out gradually rather than just flipping everything on at once.

Aaron: That makes complete sense. So, thinking about this from a vendor perspective — we serve a lot of companies with the same business model as yours, and I'm curious: what does a credible business case need to include to survive an FP&A review? You've adopted Qlik, Vertex, eCommerce — there are vendors in your world whose job is to solve problems for you. Finance is often one of the sign-offs, even on purchases that don't directly involve your team. What does a vendor need to bring to that conversation?

Ryan: I think it comes down to three things: price, solution, and partnership. And price and solution have a push-pull relationship. If you're highly specialized and you're the only one with the solution to my specific problem, you have more leverage on price — I'm not pushing back hard, because the value is clear. At a mid-market company like ours, somebody is doing that work by hand right now. By hand means time, errors, and everything falling apart when that person goes on vacation. So if the solution is real, price becomes secondary.

The key is listening to where the customer actually is. If I already know what the solution is and I'm just evaluating vendors, I'm more price-sensitive. If I only know I have a problem and I've heard you can solve it, you have more room on price. And that's where partnership becomes critical. The most valuable thing I've seen vendors do is say: "We solved this problem for you today — but companies who've had this problem often find that the underlying issue is actually this. And either we can help with that too, or we know someone who can." That kind of relationship means I'm coming to you first for the next problem, even if it's outside your core product.

Aaron: That's a really important point. So if I'm a director of digital at a company like yours — margin-constrained, not exactly swimming in venture capital — and I want to be the champion for a new piece of software, what's your advice for making that case up the chain?

Ryan: Know their business. Not just their surface-level problem, but their actual business. It's not enough to say "we solve this problem" — management might already have a solution for that, or might not think that problem applies to them. But when you walk in and demonstrate that you genuinely understand how their business works, that creates trust. And trust creates partnership. If I feel like you know my business, I'm coming to you first when the next problem comes up — even if it's in an adjacent area you don't directly cover.

And practically speaking — start with vendors already in your portfolio. I cannot stress this enough. If someone already in my stack can flip a switch and solve the problem, that is always my first call. No Google search, no RFP, no demos with 10 new vendors. Just: can you do this? Here's a couple thousand more a month. Done. And even when a vendor can't solve it themselves, the ones I trust most will say, "We don't do that, but here's who plays well with us." That level of honesty and ecosystem awareness is incredibly reassuring. As someone who does not want to change their ERP, their BI tool, or their payment solution, knowing that a new vendor integrates cleanly with what I already have makes the decision so much easier. I'm not reinventing the wheel — I'm adding to it.

Aaron: And I'd imagine that post-demo internal selling matters a lot too. When someone on your team comes out of a product demo energized and starts connecting dots for other people in the organization — that can be more powerful than anything the vendor said in the room.

Ryan: Absolutely. Getting as many people as possible into those demo calls — not just the decision-maker — is huge. The person at 30,000 feet only cares about the one headline solution. But the people underneath them who are in the weeds every month? They're the ones who will come out of that demo saying, "Oh my gosh, we could use this for five other things I've been struggling with." That internal advocacy is often what closes the deal, not the vendor pitch itself.

Aaron: Last serious question before we wrap up. You mentioned early on that your most valuable customers are the ones touching the business across multiple channels — rep relationship, online, in-store. From a payments standpoint, how does that typically work? What we usually see in the industry is that payment tends to be very channel-specific — someone pays one way online and a completely different way through a rep. Does that describe SSG? And what does reconciliation look like on the back end — credit memos, chargebacks, returns, wire transfers, paper checks? How do you make sense of what your bank account looks like at the end of the month when your best customers are putting money in through all kinds of different methods?

Ryan: Great question. Going back to something fundamental about our business — we know every customer when they walk in. No one is buying anonymously. You have to give us an account number, and part of that account setup includes identification, so we can keep a card on file. That card on file can be used for in-store purchases, online orders, and rep-assisted orders. That gives us a lot of flexibility and removes a lot of friction from the ordering process — it genuinely feels seamless across channels.

Aaron: That's fantastic. But you mentioned a bit of a hot-button word there — chargebacks.

Ryan: Right. Card-not-present is not something credit card companies love to hear, and you're not going to win a lot of those disputes. You pay more in processing fees for a reason. And when you're trying to provide a frictionless checkout experience — because these are business owners who are busy and just want product in their hands fast — you're not always getting a signature. You want that new product out of the warehouse and into their salon as quickly as possible. The tradeoff is that when someone says "I didn't authorize that," you don't have a lot to stand on. Maybe you have a text, a phone call, a verbal agreement. It's hard.

But we feel the volume and loyalty we gain from making the process frictionless outweighs the chargeback risk — at least for now. That calculus could change, but right now it's worth it.

On the reconciliation side, most customers tend to use one form of payment consistently. They're either on net terms with a card or ACH on file, paying cash at the counter, or using a card at the time of purchase. So while they're ordering across multiple channels, the payment method itself is usually pretty consistent per customer, which keeps reconciliation manageable.

Aaron: That's a relief to hear. What about partial fulfillment? If someone places an order and you can only fulfill 80% of it, are you holding a charge on the full amount in the meantime?

Ryan: That's actually something we've had to be very thoughtful about. A lot of our customers are small business owners using debit cards, not credit cards. If we place a hold on the full order amount and can only ship 80%, we're tying up their money — sometimes for a couple of weeks. A double charge or an unexpected hold of $700 might not be a big deal to a large salon chain, but to a one- or two-chair operator, that's a significant cash flow issue. So we've had to really understand how our payment system works at a mechanical level, be proactive about tracking those situations down, and give both the customer and their rep the reassurance that we're on top of it and it'll be resolved. We're not perfect at it, but I think we're at a pretty good place. And the fact that our payment infrastructure is essentially centralized across channels — rather than siloed by channel — is a huge advantage. That's often not the case in this industry, and when it isn't, it creates a whole separate set of problems when you're trying to figure out where an order came from and how to handle it after the fact.

Aaron: Couldn't agree more. Great conversation, Ryan. I really appreciate it. Now — the question all longtime listeners know is coming. What is one piece of media you've consumed in the last 30 to 60 days that you'd genuinely recommend to someone else? Book, magazine, podcast, YouTube series, movie, TV show — anything goes.

Ryan: I know this audience, so I'll go with a magazine called Imbibe. It's primarily about cocktails, but over the last couple of years they've quietly been folding in tea and coffee coverage — not making a big announcement about it, just gradually expanding until it's almost a third of the magazine now. They're also building out a podcast. It's been a really interesting evolution to watch. And it's made me realize my media consumption habits are a little different than I thought — I'll actually sit down and read a physical magazine just because it's on my coffee table. Something about that format changes how I engage with it. If you're into the broader drinks world, I'd highly recommend checking out both the magazine and what they're building around it.

Aaron: I hope they give you an affiliate code at this point. If I sign up, I'll be sure to use promo code Ryan for 15% off.

Ryan: I hope so too.

Aaron: Ryan, thank you so much for being here and sharing all of this with our audience. I hope this looks and sounds as good to everyone watching and listening as it does to us here in the studio. Really appreciate it, and have a wonderful rest of your day.

Ryan: Thanks so much.