Retool, Internal, or build it yourself? Pick on a worksheet, not on Hacker News.

Customer-facing software gets the build-vs-buy treatment. Internal tools rarely do. The first admin panel gets thrown together in a weekend on whichever platform an engineer used at their last job, and then every subsequent admin panel gets thrown together on the same platform because the standard has already been picked by accident. Six months later, the platform bill is climbing faster than the team expected, the per-seat costs hit at exactly the worst moment (when ops is expanding), and there is no defensible documentation of why this platform was picked in the first place. Switching costs are now substantial.

This worksheet exists because the build-vs-buy framework that works for customer-facing CRM does not directly translate to the internal-tools category. The per-seat economics are different (you pay for every ops headcount, not every customer), the integration patterns are different (mostly read from your own data warehouse rather than write to external systems), and the audit needs are different (a different surface area for SOC 2 and customer-data exposure). The 10-factor matrix in this worksheet is tuned for those category-specific differences.

The worksheet is targeted at four roles. First, heads of operations at 10-to-500- employee companies whose ops team is about to spin up its first or third internal tool. Second, engineering managers at the same scale who are tired of being the bottleneck for every ops request and need to evaluate whether to give the ops team a no-code platform or build a thin custom system that handles 80% of requests. Third, IT directors evaluating internal-tools platforms as part of a SOC 2 readiness review or a customer security questionnaire response. Fourth, founders making the first call on which platform their company standardizes on for back-office work.

The worksheet is industry-agnostic but most useful in verticals where ops complexity is high — including fintech, healthcare, insurance, and legal services where the audit and data-sensitivity factors weigh heavily on the platform choice.

You will leave with a scored evaluation of the major internal-tools platforms against your specific situation, a 5-year TCO comparison you can present in a finance review, and a one-page decision template suitable for a leadership Slack channel. You will also have the vocabulary to push back on common platform-side sales patterns — including the one where the per-seat price quoted on the marketing page is materially below the actual price after enterprise features are bundled in.

On the build-from-scratch side, you will have a realistic engineering scope (6-to-10 weeks for a first internal tool on a Next.js plus tRPC plus Postgres baseline) and a list of the production-readiness items that distinguish a real tool from a weekend prototype — including RBAC, audit logging, SSO, IP allowlisting, and the disaster-recovery posture you need before the tool touches customer-related data. Many of those items overlap with the broader MVP-to-production checklist.

On the migration side, you will understand what it actually costs to move from a platform like Retool to a custom system 18 months in — which is the single most common failure mode for the "just pick the platform" default. The worked example in Section 8 covers a real migration timeline and the data and integration gotchas to plan for.

A significant share of our custom business software engagements begin as Retool migrations. The pattern is consistent: a small ops team built the first admin panel on Retool, hit a customization or per-seat-cost wall around month 18, and is now evaluating whether to keep paying or build the replacement. The worksheet is what we send before the first scoping call so the conversation can start at the right altitude. It also helps validate web application builds where the back-office surface is half the scope.

For the broader build-vs-buy framework that addresses customer-facing software, see the build-vs-buy playbook and the build-vs-buy calculator. For pricing on the discovery sprint that typically precedes a migration build, see our pricing page or just book a call. Recent migrations are listed on our work page.