The model behind this site's free ROI calculator, as a spreadsheet where every formula is visible and every assumption is an editable cell — plus a one-page exec summary of the model's default run. Half an hour of your own numbers, and the case is yours to defend in the meeting where the budget gets decided.
Agrees to the cent
The free calculator at /tools/roi-calculator runs the same model: same inputs, same annual saving, ROI, payback and NPV.
Designsystems.one · Tools
The business case, as a model you can open and edit.
Counted, not rounded
Five sheets, 142 live formulas, 20 editable inputs, and a one-page exec summary — read off the files themselves.
or write to kir.zhukovsky@gmail.com — subject prefilled
XLSX + PDF · 5 sheets · 142 live formulas · instant download · one-time purchase
The same arithmetic in a browser tab — no email, no account.
Kiryl Zhukouskimaintainer of the Agent-Ready Index — 37 systems, audited quarterly.
The ROI Kit
or write to kir.zhukovsky@gmail.com — subject prefilled
01 · What you get
5 sheets — The workbook
Inputs, Model, Scenarios, Summary and Read me, in design-system-roi.xlsx. No macros, no protected cells.
142 live formulas — Nothing is pasted
Every row on the Model sheet is a formula over named cells on Inputs, with a plain-English column beside it saying what it does.
20 editable inputs — Where your numbers go
Seven about your team and spend, six levers for how much you think the system helps, seven baselines the web calculator keeps hidden.
13 assumptions — The Summary sheet, printed
Three scenario outcomes, every assumption with an evidence-source cell and a High/Medium/Low confidence dropdown, a sensitivity pair and a sign-off block — one portrait page.
1 A4 page — The exec one-pager
The headline saving, payback in months, a before/after on hours, a payback timeline, three supporting figures, and the inputs every one of them came from.
1 licence — Team-shareable
One purchase covers internal use across your team, and LICENSE.md ships in the download.
02 · Look inside
The sheet from design-system-roi.xlsx as it ships — the real formulas, in named references, with a plain-English reading of each one in the column beside it. If you are going to defend a number in a budget meeting, you should be able to see the arithmetic before you pay for it.
Step 1 prices what a year of projects costs you today. Step 2 is the six savings lines — design time, developer time, quality, consistency, onboarding, maintenance — each one a formula over the Inputs sheet, summed into a single annual figure. Step 3 is what leadership asks for: total spend, net total, ROI, payback in years and in months. Step 4 is the discount factor for each year of the horizon.
The damping is on the page rather than buried: quality and consistency are multiplied by 0.10, maintenance by 0.15, and the notes column says so in words. Payback is (ImplementationCost+MaintenanceCost)/AnnualSavings — the same expression the site calculator evaluates, which is why the two agree to the cent.
03 · Free sample
Free counterpart
The calculator runs the same model in your browser and returns the same annual saving, ROI, payback and NPV. The $29 buys that model as a file: formulas you can click into, cells you can change, an artifact you can save beside your budget documents and send to someone who will not open a browser tab in a planning meeting.
Open the free calculatorBrowser · seven inputs, six levers · no email required
04 · Fit
The same model runs in your browser at /tools/roi-calculator — same inputs, same six savings lines, same ROI, payback and NPV. It is free and stays free. Spend the $29 for the model in a file you can edit, audit and forward.
You have a number to defend in a budget meeting
The formula stays on screen when it gets challenged, and the cell changes in front of the room.
The case has to travel as an attachment
A file to save beside the budget documents and forward to someone who will not open a browser tab during a planning meeting.
You want your own hours in the baseline
The 40 and 80 hours per project are editable cells. Put hours from three real projects in them and the case stops being someone else's.
You need a conservative column and one printed page
The Scenarios sheet scales the levers down to numbers you would still stand behind under questioning. The Summary sheet prints the result for leadership.
Skip it if
05 · Price
The boring terms
The ROI Kit · $29
The files are finished and being packaged for checkout. Get notified when it opens — until then, the free calculator runs the same math for nothing.
or write to kir.zhukovsky@gmail.com — subject prefilled
This kit makes the case; the Migrate Playbook sequences the work it argues for.
06 · Details
Every sheet in full, the worked example, three scenario runs with every input printed, the working method, and the questions worth asking before you pay.
A browser tab is fine for exploring. A budget conversation needs an artifact: something leadership can open, poke at, and forward — and something you can put in front of them that fits on one page.
design-system-roi.xlsx
roi-one-pager.pdf
The headline saving and the payback in months, a before/after on design and engineering hours, a payback timeline, and three supporting figures — net gain, what a year of projects costs, and NPV. Every figure on it is the model run on its default inputs; the workbook is where your numbers go.
It prints the inputs it used along the bottom, and a “what to challenge” paragraph naming the three numbers the estimate is most sensitive to. Nothing on it is sourced from an industry benchmark, and it says so in the first line — because a business case that hides its assumptions dies in Q&A.
Typeset in the site's own type and colour, so it looks like it came from somewhere rather than out of a template gallery.
Every figure here is the model run on the calculator's default inputs: a team of 10, 12 projects a year, $75 and $85 an hour, $50,000 to build the system and $15,000 a year to run it, over three years. They are the defaults, not your numbers and not benchmarks. Change one cell and all of it moves.
$60,430
Annual saving
Six savings lines, priced at those hourly rates
12.9 mo
Payback
Build cost plus one year of running cost, over annual saving
90.8%
ROI over 3 years
Net $86,290 against everything spent
$62,978
NPV at 10%
Three-year horizon, the site's discount rate
1,440 hours of design and engineering across 12 projects, against 1,056 with the system — 384 hours back on the roadmap, same headcount. It all rests on the 40-and-80-hour baseline, which is the first cell to overwrite with your own numbers.
78% of the total is hours — developer time, onboarding, design time. The quality, consistency and maintenance lines are damped to 10–15% of face value before they count as money, so they stay small. The spreadsheet ships this same breakdown as a chart on the Model sheet.
The middle one is the calculator's defaults. Either side of it: a six-person team with a cheaper system, and a thirty-person team with a more expensive one. Every input is printed with its scenario and the six levers are left at their defaults, so the arithmetic is yours to check — run any row in the free calculator and you should land on the same figures.
960 h → 704 h of design and engineering
$36,800
Annual saving
The six savings lines, at these rates
12.7 mo
Payback
Build plus one year of running cost, over the annual saving
93.7%
ROI over 3 years
Net $53,400 against everything spent
$39,134
NPV at 10%
Three-year horizon, the site's discount rate
1,440 h → 1,056 h of design and engineering
$60,430
Annual saving
The six savings lines, at these rates
12.9 mo
Payback
Build plus one year of running cost, over the annual saving
90.8%
ROI over 3 years
Net $86,290 against everything spent
$62,978
NPV at 10%
Three-year horizon, the site's discount rate
2,880 h → 2,112 h of design and engineering
$150,300
Annual saving
The six savings lines, at these rates
12.8 mo
Payback
Build plus one year of running cost, over the annual saving
87.9%
ROI over 3 years
Net $210,900 against everything spent
$154,300
NPV at 10%
Three-year horizon, the site's discount rate
Four times the annual saving between the smallest and the largest team here — driven by eight projects a year becoming twenty-four, and by onboarding, which scales straight off headcount. This is not a claim about what your team would save. It shows the model responding to its inputs; there is no stored benchmark underneath.
At the model's 40-and-80-hour baselines with the default 30% and 25% reductions: 256, 384 and 768 hours. This is the line your own timesheets can contradict, and that is the argument worth having. The baseline is an editable cell; put measured hours in it.
What the three rows show
Payback lands at 12.7, 12.9 and 12.8 months and ROI at 93.7%, 90.8% and 87.9% — which looks like a law and is not one. It is an artefact of these three scenarios scaling build and run costs roughly in step with team size. Halve the build cost in the middle row — $25,000 rather than $50,000, everything else untouched — and payback falls to 7.9 months while ROI rises to 159%.
That is the reason to want the spreadsheet: the ratios move with the inputs, and the only inputs that matter are yours. The Scenarios sheet exists so the pessimistic version of your row is one you built yourself, on levers you can defend.
Half an hour with the file, most of it spent on step two. The order matters: the case is only worth presenting once the inputs behind it are yours.
Five sheets, no macros, no protected cells. The pink cells are yours; everything else is a formula you can click into.
Your rates, your project count, your build and run costs. Then replace the 40 and 80 hour baselines with hours from three real projects — that single edit is what separates this from a benchmark someone else made up.
Scale the levers down until you would still defend them under questioning. If the conservative column clears your company's hurdle rate, the case is already made.
The PDF gives leadership the shape of the case — headline, payback, and the three assumptions to challenge — printed from the model's default run. Your case is the workbook: when a number gets challenged, you change the cell in front of them.
This is an estimate, not evidence
The model multiplies your beliefs by your rates. If the beliefs are wrong, the output is wrong to the nearest dollar, which is worse than being obviously vague. Every number it produces is only as good as the inputs you type in — and the default inputs are the calculator's placeholders, not measurements of your organisation.
The Read me sheet asks for one more thing: write down the date and the inputs you used, then check them against reality in a year. That is what turns an estimate into a receipt.
