Custom software vs off-the-shelf: a small business owner's guide

A Brooklyn restaurant owner I know once paid a developer $42,000 to build a custom reservation system. It launched. It worked. About four months later, OpenTable rolled out a feature that did 90% of what he’d built — for $129/month.

That same week, a different Brooklyn business owner told me he was paying $700/month across nine different SaaS tools to glue together a workflow that an intern was spending 12 hours a week duct-taping. A custom $18,000 build would have paid itself back in 14 months and given his intern her life back.

Both of these owners thought they made the right call. Both of them got it wrong.

There’s a framework for thinking about this that fits on a napkin. Let me share it.

The three options, honestly

For any business process that takes time and attention, you have three choices:

1. Buy off-the-shelf software. Cheap up front, predictable monthly cost, locked into someone else’s workflow.
2. Build custom software. Expensive up front, fits your workflow exactly, you own it forever.
3. Duct-tape it. Manual processes, spreadsheets, an intern, Zapier glue. Free-ish, but expensive in time and error rate.

Most small businesses default to duct tape and then occasionally pay for a SaaS subscription. They almost never seriously consider custom software because they’ve heard it’s expensive and they don’t know how to scope it.

Custom is the right answer more often than people think.

The 7-question framework

For any process that’s eating time and you’re wondering whether to invest in software, run these seven questions. The score is straightforward.

1. How many hours per week does this process take?

• Under 5 hours: 0 points
• 5–15 hours: 1 point
• 15–30 hours: 2 points
• 30+ hours: 3 points

A custom build needs to save real time to pay back. Below 5 hours/week, almost nothing pencils out.

2. How custom is the process to your business?

• Standard process every business in your industry does (invoicing, accounting): 0 points
• Mostly standard but with a twist that doesn’t fit existing tools: 1 point
• Mostly custom — you’ve built up a way of doing this that’s distinctive: 2 points
• Completely unique — no existing tool comes close: 3 points

The more standard the process, the more likely an off-the-shelf tool exists. The more custom, the more likely you need to build.

3. How long will you be doing this process?

• Less than 1 year: 0 points
• 1–3 years: 1 point
• 3–5 years: 2 points
• 5+ years (forever, basically): 3 points

Custom software is an investment. The payback math fails if the underlying process won’t exist in 18 months.

4. How many people use this process today?

• Just you: 0 points
• 2–5 people: 1 point
• 5–20 people: 2 points
• 20+ people: 3 points

More users means more value per dollar of build. Single-user tools rarely justify custom builds.

5. Is this process directly tied to revenue?

• No — internal admin: 0 points
• Indirectly — supports revenue work: 1 point
• Yes — directly produces revenue (orders, bookings, leads): 2 points
• Yes — and it’s a competitive differentiator: 3 points

The closer the process is to the cash register, the more custom is worth.

6. Are existing tools genuinely close to your process?

• An off-the-shelf tool does 95%+ of what you need with default settings: -3 points
• Off-the-shelf does 70–95% but you can configure it: -1 point
• Off-the-shelf does 50–70% — you’d need workarounds: 1 point
• Nothing on the market gets you above 50%: 3 points

This one can subtract from your score. If the right SaaS already exists, build is the wrong answer.

7. What’s the cost of a mistake in this process?

• Low — annoying but recoverable: 0 points
• Medium — costs an hour or a customer call: 1 point
• High — costs hundreds of dollars or a lost customer: 2 points
• Very high — costs a regulator, a lawsuit, or your reputation: 3 points

Higher-stakes processes deserve more careful, custom tooling.

How to read your score

Total up the points.

• 0–6 points: Buy off-the-shelf. Don’t even consider custom. There’s a SaaS for this.
• 7–11 points: Buy off-the-shelf + minor customization (Zapier, custom forms, etc.). Revisit in 12–18 months if your needs outgrow the tool.
• 12–16 points: Hybrid — buy the platform, build the differentiating piece. Common in retail/restaurants/service businesses.
• 17–21 points: Build custom. The math works. Just be smart about scope.

Some Brooklyn-specific examples

Example 1: Park Slope dentist scheduling

• Hours per week scheduling: 12 (1 pt)
• Customness: standard (0 pt)
• Duration: indefinite (3 pts)
• Users: 4 office staff (1 pt)
• Tied to revenue: direct (2 pts)
• Existing tools: Dentrix and Curve do 95%+ (-3 pts)
• Cost of mistake: medium (1 pt)

Score: 5. Buy off-the-shelf. Specifically, Curve Dental.

Example 2: Brooklyn HVAC dispatching

• Hours per week: 35 (3 pts)
• Customness: medium (1 pt)
• Duration: 10+ years (3 pts)
• Users: 8 (techs + dispatch) (2 pts)
• Tied to revenue: direct + differentiator (3 pts)
• Existing tools: ServiceTitan does 70%, but the workarounds are ugly (1 pt)
• Cost of mistake: high (lost service call = lost customer) (2 pts)

Score: 15. Hybrid. Buy ServiceTitan (or a competitor), build a thin custom layer for the Brooklyn-specific routing and the parking realities. Total cost: $14K build + $400/mo. Saves about $4,500/month vs. the current chaos.

Example 3: Boutique Brooklyn agency client portal

• Hours per week: 8 (1 pt)
• Customness: high — your whole methodology is your differentiator (3 pts)
• Duration: 5+ years (3 pts)
• Users: every client + every staff (15+) (2 pts)
• Tied to revenue: direct + differentiator (3 pts)
• Existing tools: Notion / ClickUp get to 50% (1 pt)
• Cost of mistake: medium (1 pt)

Score: 14. Hybrid → custom. Buy nothing, build a focused portal in 8–12 weeks. $22K, owned forever.

How much custom software actually costs in 2026

The honest pricing:

Scope	Cost	Timeline
Internal dashboard (single user, basic CRUD)	$6K–$12K	2–4 weeks
Customer portal (multi-user, auth, payments)	$15K–$35K	6–10 weeks
Internal tool that replaces 2-3 SaaS subscriptions	$25K–$60K	8–14 weeks
Operations platform (multi-team, real workflow logic)	$60K–$150K	4–8 months

If anyone quotes you under $5K for “custom software,” they’re either selling you a no-code template or they’re going to ghost you halfway through.

If anyone quotes you over $150K for a small business operations tool, they’re either burning a Big Agency overhead or they’re scoping for a different business than yours.

What to look for in a custom software partner

Building custom software for a small business is a different game than enterprise software. The bar:

• Fixed-fee on discovery. Two weeks, fixed price, you get a real scope document at the end. If they want T&M on discovery, walk away.
• They use boring, durable tech. Postgres, a mature framework, a hosting platform that’ll still exist in 8 years. Not the trendy framework of the month.
• They write tests. Ask if their codebase has tests. The answer “we’ll add them later” means “we won’t.”
• They demo every week. Two-week sprints, weekly demos. You see actual working software, not slides.
• You own the code from day one. Repo in your GitHub. Your domain. Your hosting credentials. If they balk, walk away.
• They have a hand-off plan. What happens if you fire them? Can the next developer pick it up in two days? Good code says yes.

How to keep the budget honest

The two ways custom software builds blow up:

1. Scope creep mid-build. “While you’re in there, can you also add…” turns a 6-week build into a 14-week build. The fix: enforce a change-order process. New features = new scope, with a written estimate, signed before work starts.
2. Over-engineering. Junior engineers build for imaginary scale. Senior engineers build for the next 18 months and refactor later. Pick people who’ve shipped before. Reject anything that doesn’t directly serve your current scope.

A good partner will push back on your scope. They’ll say “you don’t need that yet — let’s launch and revisit in 90 days.” Trust them when they do.

How we approach custom software

At Outlast Digital we run software builds in this shape:

• Week 1–2: Discovery, fixed fee. We meet your team, watch the actual workflow, write a scope document with budget and timeline.
• Week 3+: Build in 2-week sprints. Demo every Friday. Staging link live by week 4.
• Week 8–12: Launch. Owner training, documentation, hand-off ready.
• Ongoing: Maintenance retainer or clean exit. You decide. You own the code either way.

If you’ve been duct-taping a workflow that’s costing you hours every week, book a 30-minute call. We’ll walk through the framework above with you and tell you honestly whether you should build, buy, or just keep duct-taping for now.

Built to outlast. Ready when you are.