Every service business owner has lived this version of the story.
You start with one tool. It mostly works. A few months in, you realize it doesn’t quite handle the thing you actually need it to do, so you add another tool. That one handles the gap. But now you have two systems that don’t talk to each other, so you find a third tool to connect them. Then the second tool’s pricing changes, and the third tool gets acquired, and somewhere in the middle of all that you discover that nobody on the team is sure which system is the source of truth anymore.
A year later, you’re paying for nine subscriptions, using maybe five of them well, and the business still feels like it’s held together with email.
The math behind this isn’t subtle. Small businesses now run an average of 72 SaaS applications. The average company wastes nearly half its SaaS budget on unused licenses — and per-employee SaaS spend for small companies averages $400 to $600 per month, or roughly $5,000 to $7,000 per employee per year. On a ten-person business, that’s a $50,000–$70,000-a-year line item, mostly invisible because it gets paid in $49-and-$199 monthly increments.
That’s the visible cost. The deeper cost is what off-the-shelf SaaS does to how the business runs.
SaaS forces businesses to fit the software’s worldview
Every off-the-shelf SaaS product is built around a set of assumptions about how your business operates. This is what a project looks like. This is what a customer is. This is how invoicing works. This is the order of these steps.
When those assumptions match how your business actually operates, the software is a gift. When they don’t, the software becomes a low-grade tax on every workflow.
The mismatches manifest in predictable ways:
- Workflows that require workarounds every time. “When a new client signs, we put them in System A, but then we also have to manually enter them in System B because the integration doesn’t pull custom fields. And then we tag them in Spreadsheet C for the reporting.” None of these are bugs in any individual tool. They’re failures of fit.
- Fields and statuses that don’t match your business. Your CRM lets you mark a deal as “won” or “lost.” Your business has six meaningful stages between those two states. You either lose the data or shoehorn it into a comment field nobody reads.
- Reports that almost but don’t quite answer the question. You can pull “deals closed this month” from one system, “revenue collected” from another, and “projects in progress” from a third — but no tool gives you the cross-cut view you actually need to run the business.
- Workflows that depend on someone remembering to update something. This is the most common one. The data flows aren’t connected, so someone has to be the human integration layer. That someone is usually the owner.
None of these problems are anyone’s fault. They’re structural. Service businesses are heterogeneous. Off-the-shelf software has to be homogeneous to be profitable. The mismatch is built into the relationship.
Why service businesses suffer the worst from this
The SaaS-fit problem hits service businesses harder than e-commerce, software, or most other business types, for a few specific reasons:
-
Service businesses have idiosyncratic workflows. A fence builder’s project lifecycle is genuinely different from a roofer’s, which is genuinely different from a landscaper’s. The mass-market software has to choose: serve “contractors” generically and miss the specifics, or specialize so narrowly that the addressable market gets too small to fund the product.
-
Service businesses cross too many functional boundaries. A single client interaction touches sales, scheduling, project management, communication, invoicing, payment, and reporting. There’s no single SaaS that handles all of that. So you end up with one tool per slice — and the slices don’t add up to a working business.
-
Service businesses have to coordinate physical work with digital systems. Crews in the field, jobs that move, materials that get delivered, customers who text questions. Off-the-shelf SaaS is built for desk work. The handoff between the digital and the physical is where most workflows actually break.
-
The “long tail” matters. A retail or e-commerce business can survive on a stack that handles 95% of their cases well, because the remaining 5% is statistical. A service business lives in the edge cases — the unusual scope, the picky client, the weather delay, the rush job. SaaS that’s optimized for the common case fails specifically where service businesses make their margins.
The “Frankenstein stack” tax
When SaaS doesn’t fit, what most service businesses end up with is what we’d call a Frankenstein stack: a chain of products held together with manual processes, exported CSVs, copy-paste, and a Zap or two. It works, in the sense that the business runs. But the cost of running it is enormous.
Concretely, the Frankenstein stack costs you:
- The integration tax. Every tool integration is something that can break, and silently. When it breaks, someone notices three days later that the data didn’t sync. Then someone (usually the owner) has to figure out which day’s worth of data needs to be reconciled by hand.
- The license tax. You pay for seats in tools where half the team needs read-only access but the only available plan is full-seat. You pay for “premium” tiers because one feature you need is locked behind it.
- The training tax. Every new employee has to learn five or six systems before they can do their job. The systems each have their own quirks and their own logins. Ramp time stretches from weeks to months.
- The owner-attention tax. Because no system has the full picture, the owner becomes the integration layer in their own head. They’re the one who knows that what’s marked as “scheduled” in System A is sometimes actually “complete” because the crew updated System B but not A.
- The decision tax. Because no system gives a clean read on the business, decisions get made on intuition rather than data. Profitability per job, real utilization rates, true revenue pipeline — all there in pieces, none of it visible at the level of a single dashboard.
The Frankenstein stack works until it doesn’t. Then it doesn’t, in the worst possible way: at the moment of scale.
When custom actually wins
Custom isn’t always the right answer. To be honest: there are entire categories where you should buy SaaS, not build:
- Accounting and bookkeeping. QuickBooks, Xero, FreshBooks. Don’t build this. Tax law alone is reason enough.
- Payroll. Gusto, Rippling, ADP. Same logic.
- Email and document collaboration. Google Workspace, Microsoft 365. Battle-tested infrastructure.
- Cybersecurity and identity. Not your business to maintain.
- Generic horizontal categories. If a well-built tool exists for exactly your problem and the problem is the same as everyone else’s, buy it.
Custom wins specifically when:
- The work is specific to how your business operates. Quoting against your service catalog. Scheduling crews based on your specific routing rules. Tracking project costs with your particular method.
- The workflow crosses multiple functional areas. Sales-to-operations handoff. Project-to-invoice flow. Client communication across stages. The handoffs are where the Frankenstein stack falls apart and a single custom system pays back fast.
- The data needs to be consolidated for decisions. When you need a single dashboard pulling from sales, ops, and finance to actually understand the business, the integration cost of doing that with five SaaS products is often higher than the build cost of a unified system.
- You’re large enough that the SaaS line item is meaningful. When you’re paying $50K+ a year across a sprawling SaaS stack, the math on a one-time custom build with modest hosting changes considerably.
What “custom” actually looks like in practice
Custom doesn’t mean rebuilding QuickBooks. It means building the connective tissue between the tools you should keep — and the workflows that don’t fit any off-the-shelf product.
A typical service-business custom platform might:
- Keep the accounting SaaS, Google Workspace, and payroll
- Replace the four overlapping tools handling CRM, project management, quoting, and client communication with a single custom system
- Connect to the SaaS that stays via APIs, so data flows cleanly without manual intervention
- Surface a single operational dashboard that shows the actual state of the business in real time
The result isn’t more software. It’s less software, doing more, fitting your business instead of fighting it.
We recently built exactly this kind of platform for a fence and deck building company. Before the build, the business was running on a CRM, a separate project management tool, two estimating spreadsheets, a quoting plugin, an SMS service, a scheduling app, and an accounting system — none of which fully talked to each other. After the build, the operational work happens in one custom platform that handles employee management, client management, quoting (with AI assist against their service catalog), scheduling, time tracking, project cost tracking, e-signatures, and client texting. The accounting stayed in QuickBooks. The Frankenstein stack went away.
The reframe
Off-the-shelf SaaS isn’t bad. It’s just not infinitely scalable to your specific business. There’s a point in every service business’s growth where the cost of fitting your business to the software exceeds the cost of building software that fits your business.
The owners who recognize that point early stop stacking subscriptions and start building the connective platform their business actually needs. The ones who don’t keep adding tools, keep adding manual workarounds, and keep paying the Frankenstein-stack tax until something breaks badly enough to force the conversation.
If your SaaS bill keeps growing and your business keeps feeling like it’s held together with email, the question isn’t which tool to add next. It’s whether the stack still fits the business — and whether it ever did.
If you’d like a clear-eyed look at where your current SaaS stack is helping vs. where it’s becoming a tax on growth — and whether a custom platform would change that math — book a 30-minute call and we’ll work through it together.