Microsoft 365 Copilot only pays off if the time it saves is worth more than the seats, the metered credits and the rollout. Model the return for your team, in dollars, with Australian wage defaults and a three-year payback view.
Beyond the estimate
We'll turn the business case into a rollout that actually lands the savings.
Headcount, salary and the hours Copilot saves each person per week. Defaults are Australian SMB averages - edit any cell, or add your own role.
RolePeopleSalary / yrHrs / wk
Fully-loaded overhead
An employee hour costs more than the salary rate once super, on-costs and overhead are counted. Super, on-costs, overhead.
Step 2 · How fast it lands
Steady rollout with some enablement. About 85% of full value by month 9.
Step 3 · What it costs
Estimated 3-year return
192%
3-year ROI
5 mo
payback
Cumulative valueCumulative costBreak-even (mo 5)
$66,315
Year-1 net gain
$9,869
Net / month (steady)
$493
Net / user / month
$303,173
3-year net gain
Over 3 years
Value returned$461,140
Total cost$157,967
2.92x value per A$1 of cost · seats $659/mo · credits up to $4,567/mo
Value by role (at full adoption)
Corporate knowledge workers × 10$96,842/yr
$53/hr loaded · 48% of value
Customer-facing knowledge workers × 5$51,069/yr
$59/hr loaded · 25% of value
Technical workers × 3$29,961/yr
$72/hr loaded · 15% of value
Managers & senior leaders × 2$25,572/yr
$86/hr loaded · 13% of value
Directional estimate, not a guarantee. Value assumes saved time is reinvested in productive work and is realised on the adoption curve you chose. Costs reuse our Cowork pricing - seats at partner AUD pricing, Copilot Credits converted at the live rate above.
Get this ROI breakdown in your inbox, with the assumptions written out, ready to put in front of a budget holder. No gate on the calculator - this is optional.
How to read it
What every number means.
The model builds value from the bottom up - per role, from hours saved and a fully-loaded rate - then nets off the cost month by month. Here is each field, step by step.
Step 1
Your team, by role
Role
Each row is a type of worker. Four are pre-filled with Australian SMB averages; rename them, edit any number, remove rows, or add your own.
People
How many in that role will have Copilot switched on.
Salary / yr
Annual base salary, before the overhead loading. Defaults are AU SMB mid-points.
Hrs / wk
Hours Copilot saves a typical person in that role each week. Defaults blend to about 45 min a day, between the UK Government and NHS findings.
Fully-loaded overhead
An employee hour costs more than the salary rate once super, on-costs and overhead are counted. +30% is standard.
Step 2
How fast it lands
Adoption scenario
How quickly the team reaches steady-state usage. Conservative reaches ~60% of full value, moderate ~85%, aggressive 100%.
Why it matters
A licence nobody opens returns nothing. Seats are paid in full from day one, but value and credit usage both ramp on this curve.
Step 3
What it costs
Copilot licence
The seat each user needs - Copilot Business (~AU$33) or the enterprise add-on (~AU$47).
Usage intensity
How heavily the team uses Cowork's agentic tasks, which sets the metered Copilot Credits. For the full credit breakdown, use the cost calculator.
One-off implementation
Optional rollout cost - training per user, enablement and data governance - applied once in month one.
How the estimate works
Value, minus the reality of adoption, minus the cost.
Value (the upside)
Hours Copilot saves each person, in dollars, at their fully-loaded rate. Roles heavy on email, meetings, drafting and search see the most.
Adoption (the reality)
Savings ramp as people build the habit. Pick a conservative, moderate or aggressive curve - this is the single biggest swing in the result.
Cost (the netting)
Copilot seats, metered Copilot Credits and a one-off rollout cost, subtracted month by month. The same AUD pricing as our cost calculator.
Where the numbers come from
Defaults pitched as a floor, not a sales figure.
The time-savings defaults sit near the lower end of the published evidence. Here is the research behind them, and what actually moves the return up or down.
UK Government Copilot experiment
Around 26 minutes a day saved on average across 20,000 civil servants.
NHS England AI pilots
Up to ~43 minutes a day for some clinical-admin cohorts.
Microsoft Research / WorkLab
About 29% faster on common drafting, summarising and search tasks.
Forrester Total Economic Impact
Commissioned studies report three-year ROI from roughly 100% to 450%.
01
Adoption, not licences
A seat that nobody opens returns nothing. The ramp scenario is the single biggest swing in the model - enablement is where the ROI is won or lost.
02
Which work it touches
Email, meetings, drafting and search are where the minutes are. Roles heavy on those see the most return; deep specialist work, less.
03
Whether saved time is reinvested
Minutes returned only become value if they go into billable, revenue or higher-order work, rather than quietly evaporating.
04
Keeping the variable cost capped
Cowork's metered credits can erode the return if usage runs hot. Spending caps keep the cost side predictable - see the cost calculator.
Modelling the spend in detail? Use our Copilot Cowork cost calculator for the per-persona Copilot Credits breakdown and the spending caps that keep it predictable.
FAQ
Questions about Copilot ROI.
How is the ROI actually calculated?
Value is hours saved per week times 46 working weeks times each role's fully-loaded hourly rate (salary times the overhead multiplier, divided by 1,976 paid hours a year). That value ramps over 36 months on the adoption curve you pick, and we subtract Copilot seats, metered Copilot Credits and your one-off rollout cost month by month. Three-year ROI is (total value minus total cost) divided by total cost.
Where do the default hours-saved numbers come from?
They sit between the published studies. The UK Government's Copilot experiment found about 26 minutes a day saved on average; some NHS England pilots reported up to 43 minutes; Microsoft Research reports roughly 29% faster on common tasks. Our moderate defaults blend to about 45 minutes a day, weighted by role - and every figure is editable.
Why are the salary defaults what they are?
They are Australian SMB mid-points for June 2026 - the national full-time average is about $100k and the median about $90k, with SMB pay sitting a little below large-corporate. We default corporate knowledge workers to $80k, customer-facing to $90k, technical to $110k and managers to $130k. Adjust each to your actual payroll.
Isn't this just a best-case sales pitch?
It is built to be the opposite. The time-savings defaults sit near the lower end of the evidence, the working-time basis is leave-adjusted, and the model only counts value if you assume saved time is reinvested in productive work. Switch the scenario to Conservative and the payback period stretches honestly.
How does this relate to your cost calculator?
They are two halves of the same decision. This tool models the value side - what Copilot returns. The Copilot Cowork cost calculator models the spend side in detail - seats plus metered Copilot Credits, with the full per-persona credit breakdown. Use both: this one for the business case, that one for the budget and the spending caps.
Does the cost side include Cowork's metered credits?
Yes. On top of the fixed Copilot seat, the model adds metered Copilot Credits based on the usage intensity you choose, billed in USD by Microsoft and converted to AUD at the live rate shown on the tool. For a precise, per-persona credit estimate, use the cost calculator.
Is anything I enter here sent to you or stored?
No. The calculator runs entirely in your browser and nothing is sent anywhere. The optional email-me-this field is the only thing that would ever submit data, and it is opt-in.
Is this an official Microsoft figure?
No. It is a directional business-case tool calibrated to public research and our own Cowork pricing, not an official Microsoft quote. Treat it as a defensible floor to start the conversation, then we will pressure-test it against your real workflows.
