Platform reuse savings calculator
Weigh a shared platform against standalone designs: up-front cost, reuse share, savings per derivative, time saved and the number of variants to break even.
Cost and businessFree to useHow it worksSourcesWhere it helps
Fills the inputs below with typical figures. Replace any of them with your own.
- Saving in present value
- $13.9 m
- The platform route costs less
- Total cost, standalone
- $120 m
- $101 m in present value
- Total cost, platform
- $104 m
- $86.8 m in present value
- Saving, undiscounted
- $16.1 m
- Break-even number of variants
- 4
- 3.06 exactly; 3 in present value
- A derivative costs
- $12.8 m
- 42.8% of a standalone variant; saves $17.2 m
- Time saved per derivative
- 10.3 months
- 7.7 months against 18
- Value of changed launch dates
- -$7.89 m
- Negative: the platform delays launches
- Saving plus launch value
- $6.03 m
- Present value
Check these
- The platform takes longer than a standalone variant, so the first product on it launches later. The value of earlier launches can be negative for the first variants.
- Standalone (solid line, circles)
- On the platform (dashed line, squares)
Show the data as a table
| Variants built | Standalone | On the platform |
|---|---|---|
| 0 | $0 | $52.5 m |
| 1 | $30 m | $65.3 m |
| 2 | $60 m | $78.2 m |
| 3 | $90 m | $91 m |
| 4 | $120 m | $104 m |
| 5 | $150 m | $117 m |
| 6 | $180 m | $130 m |
Cost of each route, and the value of changed launch dates. All variants weighted by the chance they go ahead.
- Standalone cost$101 m
- Platform cost$86.8 m
- Saving$13.9 m
- Launch timing value-$7.89 m
Show the data as a table
| Item | Present value (USD) |
|---|---|
| Standalone cost | $101 m |
| Platform cost | $86.8 m |
| Saving | $13.9 m |
| Launch timing value | -$7.89 m |
Sign in to save your inputs and come back to them later.
| Variant | Standalone launch (month) | Platform launch (month) | Months earlier | Standalone cost | Derivative cost | Cumulative standalone | Cumulative platform | Launch value, PV | Cumulative saving, PV |
|---|---|---|---|---|---|---|---|---|---|
| Variant 1 | 18.0 | 31.7 | -13.7 | $30 m | $12.8 m | $30 m | $65.3 m | -$26.6 m | -$28.2 m |
| Variant 2 | 27.0 | 31.7 | -4.7 | $30 m | $12.8 m | $60 m | $78.2 m | -$9.14 m | -$12.5 m |
| Variant 3 | 36.0 | 31.7 | 4.3 | $30 m | $12.8 m | $90 m | $91 m | $8.35 m | $1.43 m |
| Variant 4 | 45.0 | 34.7 | 10.3 | $30 m | $12.8 m | $120 m | $104 m | $19.5 m | $13.9 m |
The table and chart show every variant built. The totals above weight each later variant by the chance it goes ahead. The cumulative platform column includes the up-front cost and all the maintenance.
How it works
A product platform is a shared body of design (silicon blocks, software, verification, packaging) that several variants are built on. This tool compares building a set of variants one by one with building a platform first and then a derivative for each, and gives the cost both ways, the number of variants at which the platform pays back, the development time saved and the value of the launch dates that change.
- Price a derivative. The platform covers a share of the work in a variant, the reuse share. That part is not redone, but each derivative pays an integration overhead on it (adaptation, extra verification, packaging). The rest is new work. The same share scales the development time.
derivative share = (1 − reuse) + reuse x integration overhead; derivative cost = standalone cost x derivative share; derivative time = standalone time x derivative share - Total the costs both ways. The standalone route pays full price for each variant. The platform route pays the up-front cost once, a derivative cost for each variant, and the yearly maintenance for the years it is supported. If later variants may be cancelled, each variant after the first counts with the chance that it is built, and the platform cost stays.
expected variants = 1 + (N − 1) x (1 − cancel chance); standalone total = standalone cost x expected variants; platform total = up-front cost + derivative cost x expected variants + maintenance x years; saving = standalone total − platform total - Schedule the variants. Variant k is planned to start (k − 1) x spacing months in. Standalone, it starts then and launches after the standalone time. On the platform, the platform is built first, so a derivative starts at its planned date or when the platform is finished, whichever is later, and launches after the derivative time.
standalone launch = (k − 1) x spacing + standalone time; platform launch = max((k − 1) x spacing, platform time) + derivative time - Discount to present value. Each cost is paid in the middle of its development period and the yearly maintenance in the middle of each year after the platform is done, discounted at the rate you enter. Present value gives the platform's early spending its proper weight.
present value = cost / (1 + rate)^(months / 12) - Value earlier launches. A launch that moves earlier by some months earns that many more months of revenue (or contribution) at the rate you enter. The value is discounted to the platform launch date. It is negative where the platform route launches later, which it does for the first variants when the platform takes a long time.
launch value of variant k = (standalone launch − platform launch) x revenue per month / (1 + rate)^(platform launch / 12) - Find the break-even. The platform pays back when the savings on derivatives cover the up-front cost and the maintenance. The break-even number of variants is that cost divided by the saving on each derivative. The tool also finds the first number of variants at which the present value of the platform route is no higher.
break-even variants = ceiling((up-front cost + maintenance x years) / (standalone cost − derivative cost)); time saved per derivative = standalone time − derivative time
Assumptions
- The platform is built first and every variant is a derivative of it; the first variant does not carry the platform. If your first product is built together with the platform, treat its cost as part of the up-front figure and enter one fewer variant.
- A derivative costs and takes a share of a standalone variant, set by the reuse share and the integration overhead. Cost and time scale the same way.
- Costs are paid in the middle of their development period. Variants cancelled before they start cost nothing.
- Revenue per month is the same for every variant and starts on launch; enter contribution margin rather than revenue if you want profit.
- The presets are editorial figures for the type of programme. They are not benchmarks; the published work below argues for the method, not for these numbers.
Limits and warnings
- Reuse share and integration overhead are estimates, and they drive the answer. Test the result at a lower reuse share and a higher overhead.
- Platform lock-in, lower performance from shared design choices and the cost of a platform that is later replaced are not priced.
- Variant revenue is not affected by the platform. A platform often broadens the product range, which this tool does not credit.
- Cancellation is a simple chance on each later variant. Real programmes cancel in clusters and part-way through.
- Currency, tax, inflation and staff reallocation are not modelled. Maintenance is a flat yearly cost.
Sources
The methods and default figures rest on these public sources. The wording here is our own; read the source for its full text.
- The Power of Product Platforms: Building Value and Cost Leadership (opens in a new tab), Free Press (Marc H. Meyer and Alvin P. Lehnerd), 1997. The case for developing families of products from shared components and technology, and for deriving products from a platform.
- Product platform design: method and application (opens in a new tab), Research in Engineering Design, volume 13 (Simpson, Maier and Mistree), 2001. A method for designing a product family around a scalable platform, with the trade-off between commonality and performance.
- Derivative ICs: A Look At The Options (opens in a new tab), Semiconductor Engineering, 23 February 2012. Trade-press discussion of derivative chip cost and time against a new design, which shows why reuse share and overhead need testing.
Where it helps
The lifecycle stages this tool serves: 2 stages in 1 lifecycle.
More cost and business tools
- Mask set cost estimator
- Wafer cost model
- Test time and cost estimator
- Die area estimator
- MPW shuttle cost planner
- Last-time buy calculator
- IP make or buy comparison
- FPGA, structured ASIC or ASIC crossover calculator
- Memory bit cost estimator
- Bill of materials cost roll-up
- Equipment cost of ownership calculator
Method notev1Formula, ranges, sources and limits
Compares developing several product variants standalone with building them on a shared platform: total cost, present value, savings, break-even number of variants, time saved and the value of changed launch dates.
Formula
derivative share = (1 − reuse) + reuse x integration overhead; derivative cost = standalone cost x derivative share; derivative time = standalone time x derivative share; expected variants = 1 + (N − 1) x (1 − cancel chance); standalone total = standalone cost x expected variants; platform total = up-front cost + derivative cost x expected variants + maintenance x years; saving = standalone total − platform total; standalone launch = (k − 1) x spacing + standalone time; platform launch = max((k − 1) x spacing, platform time) + derivative time; present value = cost / (1 + rate)^(months / 12); launch value of variant k = (standalone launch − platform launch) x revenue per month / (1 + rate)^(platform launch / 12); break-even variants = ceiling((up-front cost + maintenance x years) / (standalone cost − derivative cost)); time saved per derivative = standalone time − derivative time
| Input | Unit | Range it holds for | Comes from |
|---|---|---|---|
| Number of variants | variants | 1 to 30 | Your figure |
| Standalone development cost per variant | USD | from 0 | Your figure |
| Standalone development time per variant | months | from 0 | Your figure |
| Platform up-front cost | USD | from 0 | Your figure |
| Platform development time | months | from 0 | Your figure |
| Reuse share | % | 0 to 100 | Your figure |
| Integration overhead on the reused part | % | 0 to 100 | Your figure |
| Platform maintenance | USD a year | from 0 | Your figure |
| Years supported | years | 0 to 40 | Your figure |
| Variant launch spacing | months | from 0 | Your figure |
| Revenue or contribution per month per variant | USD | from 0 | Your figure |
| Discount rate | % a year | 0 to 100 | Your figure |
| Chance each later variant is cancelled | % | 0 to 100 | Your figure |
Outputs: Total cost standalone (USD); Total cost on the platform (USD); Saving (USD); Saving in present value (USD); Break-even number of variants (variants); Development time saved per derivative (months); Value of changed launch dates (USD); Present value saving plus launch value (USD).
Assumptions
- The platform is built first and every variant is a derivative of it; the first variant does not carry the platform. If your first product is built together with the platform, treat its cost as part of the up-front figure and enter one fewer variant.
- A derivative costs and takes a share of a standalone variant, set by the reuse share and the integration overhead. Cost and time scale the same way.
- Costs are paid in the middle of their development period. Variants cancelled before they start cost nothing.
- Revenue per month is the same for every variant and starts on launch; enter contribution margin rather than revenue if you want profit.
- The presets are editorial figures for the type of programme. They are not benchmarks; the published work below argues for the method, not for these numbers.
Uncertainty: The arithmetic is exact. The result is driven by the reuse share and the integration overhead, which are estimates; a ten point change in reuse share can move the break-even by a variant or more.
Known limits
- Reuse share and integration overhead are estimates, and they drive the answer. Test the result at a lower reuse share and a higher overhead.
- Platform lock-in, lower performance from shared design choices and the cost of a platform that is later replaced are not priced.
- Variant revenue is not affected by the platform. A platform often broadens the product range, which this tool does not credit.
- Cancellation is a simple chance on each later variant. Real programmes cancel in clusters and part-way through.
- Currency, tax, inflation and staff reallocation are not modelled. Maintenance is a flat yearly cost.
Where the method comes from
The Hub's own comparison arithmetic: derivative cost as a share of a standalone variant, schedule and present value on mid-period payments, and the break-even from fixed platform cost over the saving per derivative. It applies the platform idea in the published work named in the sources and prescribes no formula from it.
Reference example, checked by test: Worked example in this method note: 4 variants at $10m and 18 months standalone; platform $14m and 24 months; 60% reuse with 15% integration overhead; $0.5m a year of maintenance for 4 years; variants planned 12 months apart; $0.4m a month each; no discounting or cancellation. Derivative share 0.49, so $4.9m and 8.82 months. Standalone $40m, platform $14m + 4 x $4.9m + $2m = $35.6m, saving $4.4m, break-even 16 / 5.1 = 3.14 so 4 variants, time saved 9.18 months, launch value 0.72 months x $0.4m = $288,000. Total cost standalone 40,000,000; Total cost on the platform 35,600,000; Saving 4,400,000; Saving in present value 4,400,000; Break-even number of variants 4; Development time saved per derivative 9.18; Value of changed launch dates 288,000; Present value saving plus launch value 4,688,000.
Versions
- v1 (2026-10-25): First version, built from the tool idea.
Terms on this page
- Platform
- A reusable hardware and software base, comprising a core architecture, IP set, software stack and toolchain, from which multiple products are derived.
- Silicon
- The group IV element that carries almost all commercial semiconductor volume, valued for its abundance, its stable native oxide and a manufacturing base refined over sixty years.
Tell us what you're after and leave your email: we'll connect you with the right person on the Hub.