Internal tools are the unloved workhorses of every company — the approval portal that five people use daily, the dashboard nobody can find in Jira, the spreadsheet that four engineers manually update each Monday. When the manual process finally breaks, someone asks: should we build a tool or buy one?
This post gives you cost numbers and the decision framework to answer that question without wasting a month in vendor demos.
Internal Tool Cost Tiers
| Tool type | Cost | Timeline | Examples |
|---|---|---|---|
| Simple form / view | $8K–15K | 1–3 weeks | Employee request portal, data entry form, read-only dashboard |
| Ops workflow tool | $12K–30K | 3–6 weeks | Approval flow, onboarding checklist, incident tracker |
| Multi-role admin portal | $25K–55K | 6–12 weeks | Customer portal, finance dashboard, team ops center |
| Platform extension | $40K–80K+ | 3–6 months | Custom Jira app, Salesforce extension, data warehouse UI |
Build vs Buy: The Decision Framework
The build vs buy decision comes down to four signals. Two in favor of building, two in favor of buying.
Build when: the SaaS cost compounds faster than the build cost
For a 50-person team, Retool costs $500–2,500/month depending on plan. A custom tool built for $20K pays for itself in 8–40 months. If the team grows to 200 people, the per-seat cost crosses the break-even in 6–12 months. Run the 3-year total cost of ownership calculation, not just the upfront numbers.
Build when: your data cannot leave your infrastructure
Financial data, medical records, proprietary algorithms, unreleased product data — any data with regulatory, contractual, or competitive sensitivity may rule out SaaS entirely. Some SaaS tools offer private cloud deployments but charge enterprise premiums. A custom tool on your own infrastructure sidesteps this entirely.
Buy when: the use case is genuinely standard
Approval workflows, employee self-service portals, CRM pipelines, expense tracking — these have been solved by dozens of products. If the process can be configured into an existing product without significant workarounds, buy. The $1,000/month SaaS tool is cheaper than $20K of custom development plus ongoing maintenance.
Buy when: you need it in less than two weeks
Custom development takes weeks. If the pain is acute and the timeline is immediate, use an off-the-shelf tool as a stopgap while you evaluate whether a custom build is worth it. Retool, Notion, or even a well-structured spreadsheet can solve 80% of the problem in two days. The remaining 20% may not be worth building for.
What Drives Cost Within Each Tier
User roles and permissions
A tool used by one type of user (all admins, or all viewers) is significantly simpler than a tool with multiple roles. Each additional role requires separate permission logic, separate UI states, and separate testing paths. A two-role tool (admin + viewer) is 30–50% more expensive to build than a single-role tool. A five-role tool may cost 2x the single-role equivalent.
External integrations
Reading from an external API: $2K–4K per integration. Bidirectional sync (read + write + handle conflicts): $4K–8K per integration. Real-time sync (webhook-driven, immediate): $6K–12K per integration. Every integration you add multiplies the testing surface and the ongoing maintenance load.
Audit trail and compliance requirements
If every action needs to be logged with user, timestamp, before/after values, and exportable reports — add $5K–10K to any estimate. This is common in finance, healthcare, and legal workflows. It is not difficult to build but requires a disciplined data model from day one and cannot be bolted on later cheaply.
Free resource
Build vs Buy Decision Template
A one-page spreadsheet template with the 3-year TCO calculation for internal tools. Plug in your team size, SaaS pricing, and estimated build cost to see the break-even month.
The Retool vs Custom Development Decision
Retool is the most common alternative to custom development for internal tools. Here is an honest comparison:
| Dimension | Retool | Custom build |
|---|---|---|
| Upfront cost | $0–$5K (build time) | $8K–50K |
| Ongoing cost | $10–50/user/month | Hosting $20–100/mo + maintenance |
| Customization | Within component library | Unlimited |
| Data residency | SaaS (EU option available) | Your infrastructure |
| Speed to first version | Days to 2 weeks | 3–8 weeks |
| Performance (large datasets) | Struggles past 50K rows | Scales to your database |
The crossover point for most teams: Retool makes sense until you have more than 30–50 users or until the customization limits start requiring workarounds that are themselves expensive to maintain. Custom development makes sense when you cross either threshold.
Related service
AI Ops Sprint — Internal Automations in 5 Days
3 production automations for your ops team in 5 days. Fixed scope, fixed price, working software — not a design sprint or a prototype.
See the Sprint →Common Internal Tool Mistakes
- Building before the process is stable. If the workflow changes every month, any tool built around it will be outdated before it is finished. Stabilize the manual process first, then automate it.
- Over-scoping the first version. Version 1 of an internal tool should cover the critical path only. Features added in v1 that turn out to be unnecessary are not free — they add test surface, documentation overhead, and future maintenance. Build the smallest version that solves the primary problem and add based on actual usage.
- No owner after launch. An internal tool with no designated owner degrades within 6–12 months. Someone needs to be responsible for keeping it running, handling user requests, and coordinating updates. Without this, the tool becomes a support liability instead of a productivity asset.
- Buying a license for a tool that won't be adopted. Procurement signing a contract for a SaaS tool that frontline users do not understand and will not switch to is a common and expensive mistake. Validate that real users would actually use the tool before buying. A two-week trial on a real workflow is the minimum bar.
