Working out your virtual assistant ROI sounds easy. Your independent virtual assistant has been working with you for a month.
The invoice is easy to find. The benefit is harder to pin down. Fewer messages waiting. A bit more breathing room. Maybe a new customer.
But you're still answering questions. And sometimes finishing things yourself.
The tempting calculation goes like this: hours delegated × my hourly rate, minus the VA's bill, equals profit.
It feels right, because your time matters. But a free afternoon doesn't automatically put money in your account. And an afternoon spent fixing work was never really free.
A useful virtual assistant ROI (return on investment) check separates three questions:
- Did the work get done properly?
- Did you get usable time back?
- Did the business gain money, or another benefit you deliberately wanted?
Keeping those separate gives you an honest answer, and tells you exactly what to improve. This guide walks through it with a full worked example, including the version where the maths stops working.
Start with the decision the numbers need to help you make
Before opening a spreadsheet, finish this sentence:
"At the next review, I need to decide whether to..."
- continue as is?
- shrink the task list?
- add more hours?
- stop delegating a particular job?
Pick one arrangement and one review period. Don't mix an established VA's routine work with a new specialist's first week and expect the average to tell you much.
Then write down why you hired:
- If you wanted evenings back, a calmer week counts, even without new sales.
- If the business can only afford the VA through extra income, time alone isn't enough.
Neither goal is more respectable. They just need different evidence.
Compare against what would really have happened without them
The comparison isn't "paying a VA" vs "getting the work done for free." It's your current setup vs a realistic alternative:
- doing it yourself,
- buying a different service,
- doing less of it,
- or leaving some undone.
Find a recent period before delegation. Use your calendar, records and a short time diary to rebuild what the workload looked like. Note the type of work, volume, standard and your time.
If you never tracked time, don't invent a precise historical figure. Time a small sample now and label it "current estimate."
Two traps to avoid:
Work you never used to do. If your VA now updates contact records you previously ignored, that might be valuable. But it hasn't saved you any time, because you weren't spending any. Record it as added coverage and look for its actual effect.
An avoided future hire. That's only a real saving if you were genuinely about to buy it. A cancelled supplier invoice is real. "Maybe we'd have hired someone eventually" is just a planning note.
Count the time that came back, not the hours you paid for
Your VA's hours and your recovered hours are completely different things.
A VA learning your systems might take longer than you did. One with better methods might take less. Neither tells you whether delegating was worthwhile.
What matters is your remaining involvement. For each task, log:
- writing instructions,
- answering questions,
- checking output,
- fixing errors,
- steps you kept for yourself.
Subtract all of that from your original time. Write it down daily while it's fresh.
Don't double count. If a review session includes fixing an error, log it once. But don't hide a Saturday repair session just because it happened outside working hours.
And notice what kind of time came back. Losing a dozen scattered interruptions might improve your focus without giving you a neat empty block. A few scattered spare minutes might not let you start a demanding project. Record both the total and whether it created usable blocks.
If your questions have multiplied, look at why. Your VA might lack a rule, access or a source. Counting questions is a start; treating every question as poor performance skips the real diagnosis.
Keep cash in one column and your own time in another
Cash column (money actually leaving the business this period):
- VA fee,
- agency or platform charges,
- payment fees,
- new software bought for this arrangement.
Record one-off setup purchases separately.
Don't charge software you'd have bought anyway to the VA. And don't leave out a new subscription just because it's on a different card.
Time column (your hours, not money):
- preparation, questions, review and fixes.
You can later put a planning value on your time. Just never call that amount cash profit.
Also note when bills are due. An arrangement can look worthwhile across a quarter while needing more cash in month one than the business can safely spare. Affordability and overall value aren't the same test.
Our VA cost guide helps if you only have an hourly quote to work from.
A full worked example, all the way through
Here's a hypothetical example. All figures are invented to show the method. They're not typical rates or promised results. The period is exactly four weeks.
A solo training business delegates routine workshop enquiries and booking admin.
Before delegation: The owner handled 80 cases at about 12 minutes each. 80 × 12 = 960 minutes (16 hours). "Done" means a correct booking record plus the promised confirmation, not just a reply sent.
After delegation, the owner still spends: - 2 minutes reviewing each of the 80 cases = 160 minutes - 120 minutes giving guidance - 60 minutes fixing errors
Total: 340 minutes (5 hours 40 minutes).
Time actually recovered: 960 − 340 = 620 minutes (10 hours 20 minutes).
Cash cost: - VA: 24 hours × £18 = £432 - Software: £28 - Payment fees: £12
Total recurring cost: £472.
Notice the VA worked 24 hours to save the owner about 10. That doesn't invalidate the saving. What matters is what the owner no longer had to do, to the required standard.
What the owner did with the time: They used 8 of the recovered hours to deliver 2 extra workshops that existing customers had been asking for.
- Each workshop brings in £450, with £90 of extra delivery costs.
- Each contributes £360 before the VA's cost.
- Total extra contribution: £720.
(For simplicity, both are delivered and paid within the period.)
The financial return: £720 − £472 = £248 surplus. £248 ÷ £472 × 100 = about 52.5% return.
That result depends entirely on those assumptions. It shows a believable chain: admin done → time recovered → paid work delivered and collected.
Don't sell the same recovered hour twice
Here's an easy mistake in that example.
You could value the 8 hours at an imagined hourly rate and count the £720 workshop contribution. But the workshops are how those hours made money. Counting both is counting the same benefit twice.
So pick one view:
- Cash view: only real extra contribution and genuinely avoided costs. No imaginary value for spare time.
- Time-value view: a separate planning box, clearly labelled.
And remember revenue isn't the same as money available. Deduct the extra costs needed to earn it.
Invoicing isn't the same as getting paid, either. GOV.UK's invoicing guidance clearly separates an invoice from a receipt confirming payment. So track three separate things:
- work agreed,
- work invoiced,
- cash received.
That stops an unpaid invoice from quietly "funding" this month's VA bill. GOV.UK's guidance on payment obligations covers payment terms in a UK context; check the terms that apply to your actual customers.
The core idea: delegation creates capacity first. It only becomes money when something turns that capacity into real income or real savings.
Did the VA cause the improvement, or did something else?
A busier sales month doesn't prove your VA created those sales. Maybe you raised prices, started advertising, or it's just your busy season.
Write down other changes before crediting your VA with everything.
In the workshop example, the evidence is strong: customers had already requested workshops, and the owner simply lacked time to deliver them. Booking records, the time diary and delivered sessions all connect.
For less direct tasks, trace each step:
- A research VA supplies suitable contacts.
- You contact them.
- Some agree to talk.
- One eventually buys.
Record the evidence at each step. Don't give every person the full credit for the final sale.
And check quality before celebrating. If enquiries got answered faster but customers received inaccurate promises, today's time saving might be borrowed against tomorrow's complaints. Our VA KPI guide shows how to check work quality properly.
Setup costs need a visible home, not a permanent excuse
Back to the example. Say the trial also involved £180 of one-off setup costs.
- First-period cost: £472 + £180 = £652
- Surplus: £720 − £652 = £68
- Return: about 10.4%, not 52.5%.
Both numbers are useful:
- 10.4% describes the trial as it actually happened.
- 52.5% describes the ongoing arrangement, if the setup genuinely doesn't repeat.
Report both, rather than picking whichever makes the decision look better.
Treat training time the same way. If your guidance time drops as your VA learns, recovered time should improve in later months. If training keeps coming back because instructions change or access keeps breaking, it's an ongoing cost, whatever you call it.
Set a learning milestone, not just a date. For example: "independently handles ordinary enquiries using the policy, and correctly spots exceptions needing a decision."
Don't forecast a whole year from one great week.
Stress-test it before you add more hours
Change one assumption at a time and see what happens.
What if only one extra workshop happens? £360 contribution − £472 cost = £112 loss on the cash measure. The owner still got time back. But the business case depended on demand.
That doesn't mean the VA was bad. It means demand was the risk. Buying admin capacity can't guarantee customers will buy what you hoped to do with it.
What does each recovered hour cost? £472 ÷ 10.33 hours = about £45.68 per recovered hour. That's not the VA's rate, and it's not your worth. It's the cost of buying back usable time, which you can compare with other ways of getting it.
What's the break-even? If 8 recovered hours go into paid work, they need to generate £59 of contribution per hour to cover £472. That's a break-even point within these assumptions, not a recommended price.
Run your own low-demand and high-rework versions. If a small rise in checking time wipes out most of the benefit, improve the handover before adding hours.
And don't automatically respond to a weak result by squeezing your VA's fee. The expensive part might be the task design, unnecessary work, or nowhere useful to put the time you got back.
Some benefits don't need a price tag
Maybe the owner uses the recovered evening to rest instead of running another workshop.
That's a perfectly valid thing to buy. Just record it honestly: cost, and time outcome. Don't dress rest up as invented sales.
Make it observable:
"No routine booking work after 6pm."
That's clearer than "less stress." A diary can show whether it held and what broke it.
Coverage can matter too: work carrying on while you're at an appointment or on holiday. Test it. Can your VA handle the defined cases, reach a backup decision-maker and leave a usable record?
Write the trade-off in plain English:
"This costs us £X per period, gives back Y usable hours, and is affordable from existing income."
That's often more useful than a big percentage built on guesses.
Three quick checks you can do this week
Don't have a month of data yet? These three checks give you a rough read in a few days.
1. The quiet-fix count. For five working days, put a tally mark on a sticky note every time you redo, tidy or double-check something your VA did. At the end of the week, you'll know whether hidden rework is eating your saving.
2. The question audit. Copy the last ten questions your VA asked into a list. Mark each one: missing rule, missing access, genuine exception, or something they should have known. If most are missing rules, writing three rules down may be worth more than any extra hours.
3. The "where did the time go?" test. Look at your calendar for the hours you expected to free up. What actually happened in them? If they filled up with more admin, the problem isn't your VA. It's that nothing was planned for the time you bought back.
None of these is precise. But together they'll tell you whether to dig deeper, adjust the handover, or carry on as you are.
Your VA ROI review sheet
Keep evidence links beside each number, and leave unknowns blank rather than treating them as zero.
VA ROI REVIEW
Decision needed and review date:
Period and comparable task scope:
What I'd do without delegating:
Standard for "done" and sampled evidence:
Baseline owner minutes, and how measured:
Current owner minutes:
guidance / review / fixes / steps I kept
Net owner minutes recovered:
Usable time blocks created, and what I used them for:
Recurring cash cost (with bills):
One-off cash cost:
Extra contribution collected (with evidence):
Genuinely avoided costs (with evidence):
Other changes that might explain the result:
Non-financial outcome I wanted:
Weakest assumption, and the downside version:
Decision: continue / narrow / redesign / expand / end
One change, who owns it, next review date:
Go through the sheet with your VA where it makes sense. They might spot duplicate checks or missing information you've accepted as normal. Keep commercial details private if they're not needed.
A good final decision is narrow:
"Keep enquiry handling, clarify the refund rule, and measure again next month."
Expand once correct work, recovered time and affordable cost are all proven. The invoice will still be easy to find. Now the benefit has a paper trail too.
Related: is a virtual assistant worth hiring yet? and how much a virtual assistant costs.
Frequently asked questions
How do I calculate virtual assistant ROI?
Measure how long a task used to take you, subtract the time you still spend preparing, reviewing and fixing, and record the cash cost. Count only real extra income or genuinely avoided costs as financial return, and treat recovered time separately.
Is a virtual assistant worth it if I don't make more money?
It can be, if what you wanted was time back, fewer interruptions or reliable coverage. Just don't describe that benefit as cash profit. Judge it against the goal you hired for.
Why does my VA not seem to be saving me time?
Often because of hidden owner time: preparing inputs, answering questions, quiet fixes and rework. Log those for a few weeks and look at why they happen.
Should I count training time in VA ROI?
Yes, but keep one-off setup separate from ongoing costs. Report both the first-period result and the ongoing result, so neither hides the other.
Can I just multiply hours saved by my hourly rate?
Not as cash profit. A free hour only becomes money if you use it for paid work that actually gets delivered and paid. Use the hourly-rate view only as a separate planning tool.