Custom vs off-the-shelf retail software: a decision framework from someone who runs stores

Most build-vs-buy advice is written by people selling one of the two options. Here's the view from someone who runs a multi-location retail operation — and who replaced most of its SaaS stack with software we built after doing this math for real.

Start with one question

Is this workflow your edge, or is it a commodity?

Email is a commodity. Accounting is a commodity. Buy those — Google and Intuit have spent billions polishing them, and your version of "sending email" is not special.

But how you buy inventory, how you price, how you count, how you pay commission, how you spot a dying SKU before it dies — in retail, that's the business. When the workflow is your edge, renting a template for it means sanding your edge down to fit everyone else's average.

The 70% fit trap

Every off-the-shelf tool fits about 70% of your operation. The trap is what the missing 30% does:

• You bolt on another tool to cover part of it (new subscription, new sync problems)
• You bend your process to match the software (your edge, sanded)
• You cover the rest with spreadsheets and tribal knowledge (the real system is now undocumented)

We lived all three at once. The stack "worked," in the sense that a pile of workarounds can be said to work. Every decision started with an export. Every count was stale on arrival.

Do the three-year math

The comparison is never "SaaS subscription vs. build cost." It's the whole stack over three years:

• Per-seat fees × headcount growth × every tool in the pile
• The hours spent re-keying data between tools that don't talk (price your own time honestly)
• Decisions made late or wrong because the numbers were a week old

For a small operation, SaaS wins this math. Somewhere around multi-location, multi-channel complexity, it flips hard. Our own flip: one custom platform replaced enough subscriptions and manual hours that it pays for itself on an 18–24 month horizon — and then keeps paying.

The scorecard

Rate each statement 1–5. Over 20 total? You're a build candidate.

1. Our workflow differs meaningfully from how our tools assume we work
2. We maintain spreadsheets about our software's data
3. Someone re-keys the same information into two systems weekly
4. Per-seat costs are rising faster than revenue
5. We've asked a vendor for a feature and been told to vote on a roadmap
6. Real-time numbers would change actual decisions (buying, staffing, pricing)

If you build: three rules

1. Keep the storefront boring. Shopify is excellent at being a storefront. Don't rebuild checkout — build the operational brain behind it and sync through the APIs. That's the pattern behind both Command Center and Baseify.

2. Ship one painful workflow first. Not the platform — the single worst manual process. Prove the pattern in weeks, then expand. Big-bang operations rebuilds die in month four.

3. Own it outright. Code, data, infrastructure. The entire point is escaping rented software; don't rebuild the landlord relationship with your dev shop.