Understanding the build vs buy calculator — concept and principle
This calculator compares the total cost of building software in-house with buying a subscription over the same years, including cost per user per month and the year building catches up, if ever. IT managers and founders use it as a first financial check.
How it is calculated
- Build cost = development cost + years × (maintenance per year + hosting per year)
- Buy cost = implementation + Σ over each year t of (subscription × users × 12 × (1 + price increase)^(t−1) + other yearly costs)
- Cost per user per month = total cost ÷ (users × 12 × years)
- First year building is cheaper = first year in which cumulative build cost ≤ cumulative buy cost; "—" if it never happens within the period
Amounts are not discounted.
The principle
This is a make-or-buy comparison using relevant costs: only costs that differ between the two choices count. Building has a large up-front cost and then steady yearly maintenance, often estimated at a share of the original build. Buying has a small up-front cost and a recurring fee that grows with users and price rises. Building wins only when the period is long enough, or the user count high enough, for the subscription stream to overtake the build cost.
Worked example
Defaults: build 180,000 plus 36,000 maintenance and 6,000 hosting a year; buy 45 per user per month for 60 users, 25,000 implementation, 3,000 other costs a year, prices rising 3% a year; 5 years.
- Build = 180,000 + 5 × 42,000 = $390,000.
- Year-1 subscription = 45 × 60 × 12 = 32,400; over five years with 3% rises = 32,400 × 5.3091 = 172,016.
- Buy = 25,000 + 172,016 + 5 × 3,000 = $212,016.
- Saving = 390,000 − 212,016 = $177,984; cheaper option: Buy.
- Per user per month: build 390,000 ÷ 3,600 = $108.33; buy 212,016 ÷ 3,600 = $58.89.
- Cumulative build cost never falls below buy cost within 5 years, so the first year building is cheaper shows —.
| Build: development cost up front | Saving with the cheaper option |
|---|---|
| 45,000 | $42,984 |
| 90,000 | $87,984 |
| 135,000 | $132,984 |
| 180,000 | $177,984 |
| 225,000 | $222,984 |
| 270,000 | $267,984 |
| 360,000 | $357,984 |
How to read the result
A large saving for buying, as here, means building only makes sense for reasons not priced here, such as unique features or data control. If building wins only near the end of the period, a delay or overrun can erase the gain.
Common mistakes
- Underestimating maintenance; software needs ongoing fixes, security updates and features.
- Leaving out implementation and data migration for the bought product.
- Ignoring user growth, which raises subscription cost.
Frequently asked questions
Why is the result not discounted?
It keeps the comparison simple. For long periods, discounting future costs makes buying look somewhat cheaper relative to an up-front build.
How much does software maintenance usually cost?
The page hint suggests 15–25% of the build cost per year as a common rule of thumb; use your own estimate where you have one.
When does building make sense?
When the software is core to your competitive advantage, no product fits, or the user count is large enough that subscriptions become very expensive.
Sources: OpenStax, Principles of Accounting, Volume 2: Managerial Accounting, section 10.3 "Evaluate and Determine Whether to Make or Buy a Component"