Founders arrive at SGI with a virtual assistant business that is roughly two years old, working a full week and taking home less than they did in employment. The numbers explain why. The Society of Virtual Assistants, which runs the only industry-wide survey of the UK VA sector, put the average UK virtual assistant rate at £29.77 an hour in its UK VA Survey [1]. At that rate, 25 genuinely billable hours a week produce a gross figure of around £38,700 before software, insurance, unpaid admin, marketing time and tax. That is not a failure of effort. It is what happens when the pricing model and the positioning are both working against you.
Here is the uncomfortable truth that most guides to starting a virtual assistant business soft-pedal. The barrier to entry is close to zero, which sounds like an advantage and is actually the central commercial problem you have to solve. Anybody with a laptop and a broadband connection can call themselves a VA by lunchtime. That means the market is not short on supply, and any business that positions itself in general administrative support is competing on price against those who will always be willing to charge less. The businesses that work are the ones that solve a specific, expensive problem for a specific type of client.
This guide covers what I actually advise founders on when they come to SGI with this idea: how to choose a niche narrow enough to be defensible, how to price so that your income is not capped by the hours in your week, the legal and tax setup HMRC and the ICO require of you, the method for winning your first five clients without going near a bidding marketplace, and the five mistakes that account for most of the failures I see. I have advised more than 2,000 businesses since founding SGI in 2014, and service businesses of this type follow a consistent pattern.
Why the timing is good, and the competition is worse than you think
The demand side is genuinely strong. The Office for National Statistics recorded UK workforce jobs at 36.8 million in March 2026, with the self-employment component rising by 177,000, or 4.4 per cent, in a single quarter [4]. The Department for Business and Trade puts the UK private sector business population at around 5.7 million, of which 5.64 million are small businesses [7]. Those are your buyers. A five-person consultancy in Bristol cannot justify a full-time administrator but is drowning in invoicing, inbox management and diary conflicts. That gap is real, and it is not closing.
The supply side is where founders misjudge the market. Low barriers to entry mean the number of competitors expands to meet any increase in demand, so demand growth alone will never protect your rates. I have watched founders build careful financial projections based on market growth figures without asking the one question that matters commercially: why would a business owner choose you over the 40 other people who appeared in the same search? If you cannot answer that in one sentence, you do not yet have a business; you have a job with worse security.
What a virtual assistant business actually sells
Clients do not buy administrative hours. They buy the removal of a specific, recurring source of friction that is costing them either money or sleep. This distinction sounds like semantics, and it determines everything downstream: your pricing, your marketing, your client quality and whether you can ever hire.
The common mistake is to build a service list. Almost every new VA website I review lists the same services: inbox management, diary management, travel booking, data entry, social media scheduling, invoicing, and research. The list is comprehensive, and it tells a prospective client nothing except that you will do whatever they ask. It invites the buyer to compare you on the only remaining dimension: price per hour.
The approach I use with service business clients is to work backwards from a costed problem. Instead of asking what you can do, ask what specific failure your target client is currently living with, and what that failure costs them per month. A property management firm that loses two tenancy applications a month because nobody chases references is losing perhaps £2,400 in commission. A private clinic where 15 per cent of appointments are missed because nobody sends reminders is losing considerably more. When you can name the number, you are no longer selling time.
Work the arithmetic on a concrete example. An independent lettings agency that loses two tenancy applications a month because nobody chases references consistently is forgoing perhaps £2,400 in commission. A VA who takes ownership of the entire referencing and pre-tenancy paperwork chain is not selling six hours of administration at the £29.77 survey average, which would be £179. They are selling the recovery of £2,400 a month, and a fee of £450 to £600 is an obvious purchase for the agency at that ratio. The task is identical. The framing determines whether the conversation is about your hourly rate or about their lost revenue.
The implementation step is uncomfortable but simple. Write down the three services you are best at, then delete the other 12 from your marketing entirely. You can still deliver them to existing clients. They should not appear anywhere a prospect can see them, because a long list reads as a lack of expertise.
Choosing a niche before you choose your services
The order matters. Most founders pick their services first, then look for anyone who will buy them. That produces a client base with nothing in common, meaning every engagement is bespoke, nothing is systematised, and your cost per client never falls. Choosing the sector first means every client you win makes the next one easier and cheaper to serve.
The misconception here is that narrowing reduces your addressable market to an unviable size. Run the arithmetic before believing it. If you charge £600 per month and want £48,000 in annual revenue, you need seven clients at a time. There are roughly 5.64 million small businesses in the UK [7]. A niche representing one hundredth of one per cent of that population still contains 564 businesses. You need seven. Narrowing is not the risk. The risk is being invisible in a crowded generalist market.
The three tests a viable niche must pass
I put every proposed niche through the same three tests with clients, and a niche that fails any one of them is not worth pursuing.
- Concentrated pain. The sector must share a specific administrative failure, not merely be busy. Busy is universal and does not motivate a purchase.
- Ability to pay. The problem must cost the client meaningfully more than your fee. Sectors running on thin margins will grind your rate down regardless of how good you are.
- Reachable community. The sector must gather somewhere you can access: a trade body, a LinkedIn group, a conference, or a regional network. If you cannot find where they congregate, your acquisition cost will be prohibitive.
Veterinary practices illustrate all three tests working together. The Royal College of Veterinary Surgeons accredits 3,586 veterinary practice premises under its Practice Standards Scheme [2], so the sector is large enough to sustain a specialist but small enough to be legible. Insurance claim processing is a concentrated, recurring administrative burden across almost all of them. Margins are healthy enough to absorb a professional fee. And practice managers are a well-networked community with regional groups and a national congress, which makes them reachable without paid advertising. A niche that scores well on all three is worth far more than a larger sector that fails one.
To implement this, pick two candidate niches and speak to six people in each before committing. Ask what takes up time they resent, what they have tried, and what it costs when it goes wrong. If six conversations do not surface a consistent, costed problem, the niche fails the first test, and you should move on rather than persist.
Pricing a virtual assistant business without capping your income
Hourly billing is the single most damaging structural decision in this business model, and almost every new VA adopts it by default. It creates a hard ceiling on revenue because there are only so many billable hours in a week, and it creates a perverse incentive structure in which becoming faster and better at your work reduces your income.
It also frames the relationship badly. An hourly rate invites the client to scrutinise your timesheet rather than your outcomes, and it makes every efficiency gain you achieve a gift to them rather than a return to you. I have seen VAs automate a process from four hours to 40 minutes and immediately lose 85 per cent of the revenue from that client.
The alternative I recommend is a monthly retainer priced against the value of the outcome, with a defined scope rather than a defined hour count. You are selling the ongoing removal of a problem, not a block of time. Set the fee by reference to what the problem costs the client, what a part-time employee would cost them once employer National Insurance and holiday are included, and what capacity the work genuinely consumes on your side once systematised.
The employment comparison is the strongest anchor available to you, and most VAs never use it. A part-time administrator on £18,000 costs the employer 15 per cent National Insurance on earnings above the £5,000 secondary threshold for 2026/27, which is £1,950, before pension contributions, holiday, sick pay, equipment, and recruitment costs [3]. The real cost of that hire is comfortably north of £21,000; it is fixed, and it carries employment risk. Set against that, a £750 monthly retainer at £9,000 a year with no employment liability is straightforwardly cheap. Present the comparison that way, and the client is no longer benchmarking you against other VAs.
The practical guidance is to price your second client, not your first. Take whatever you learn about actual time consumed on the first engagement, systematise it, and set the second price against the systematised delivery rather than the exploratory version. Founders who price the first engagement and never revisit it carry an exploratory rate for years.
For a fuller treatment of how to structure this, see our guide to the art and science of pricing strategies, which covers value anchoring and the psychology of fee presentation in more depth.
The legal, tax and data protection setup HMRC and the ICO require
This section is where founders either save themselves several thousand pounds or create problems that surface 18 months later. None of it is complicated, and all of it is time-bound, which is why it gets missed.
Business structure
Most virtual assistant businesses start as sole traders, and that is usually correct at the outset. The administrative load is lighter, and losses in the first year are more easily offset against other income. The case for incorporation strengthens once profits reach a level where the difference between income tax and corporation tax outweighs the additional compliance costs, and it strengthens further if you intend to hire or take on clients whose procurement processes prefer a limited company. Our guide comparing sole trader and limited company structures sets out the threshold arithmetic in detail.
Registering with HMRC
If your gross trading income exceeds the £1,000 trading allowance in a tax year, you must register for Self Assessment. The deadline is 5 October following the end of the tax year in which you started trading, so a business that began in June 2026 must register by 5 October 2027 [8]. Note that the test is based on gross income before expenses, not profit, which catches founders who assume that a loss-making first year removes the obligation.
Making Tax Digital
From 6 April 2026, Making Tax Digital for Income Tax applies to individuals with combined self-employment and property income above £50,000, requiring digital record-keeping and quarterly updates through compatible software [8]. If you are building toward that threshold, adopt compliant bookkeeping software from day one rather than migrating from spreadsheets under time pressure later.
VAT
The VAT registration threshold is £90,000 of taxable turnover for the 2026/27 tax year, tested on any rolling 12-month period rather than your financial year [9]. Cross it, and you must notify HMRC within 30 days of the end of the month in which you crossed, with registration effective from the first day of the second month afterwards. Voluntary registration below the threshold is worth considering if your clients are VAT-registered, because they can reclaim the VAT, and you can recover input tax on your own costs.
ICO registration
This is the requirement most VAs miss entirely. If you process personal data as a data controller, which includes holding client contact records, managing an inbox containing third-party personal data, or running a CRM, you must pay the data protection fee under the Data Protection (Charges and Information) Regulations 2018. Tier 1 covers micro-organisations and costs £52 a year, reduced to £47 by direct debit [6]. Non-payment is a breach with fixed penalties attached, and it is the first thing a well-run corporate client will check during onboarding.
The ICO enforces this actively rather than theoretically. Fixed penalties for failure to pay run at £400 for tier 1 micro-organisations, £600 for tier 2, and £4,000 for tier 3, rising to a statutory maximum of £4,350 in cases with aggravating factors such as non-cooperation [5]. The ICO issued 126 monetary penalties for non-payment in a single seven-month period between May 2021 and January 2022 [6]. Set a £400 minimum penalty against a £47 annual fee, and the arithmetic makes itself. The commercial argument is stronger still: the register of fee payers is public, and a corporate client’s procurement team will check it before they check anything else about you.
Insurance and contracts
Professional indemnity insurance is not legally mandatory for most VA work but is increasingly a contractual requirement from corporate clients, and cyber liability cover is worth carrying once you hold client credentials. Every engagement should be governed by a written contract covering scope, response times, payment terms, confidentiality, data processing responsibilities, and notice. Working without one is the most common cause of the scope disputes covered later in this guide.
Winning your first five clients without entering a price race
Freelance marketplaces are the obvious starting point and the one that concerns me most. They solve the visibility problem by placing you in a global auction where buyers sort by price, and your competitors include people whose cost of living is a fraction of yours. Founders who start there frequently anchor their own rates so low that they spend two years climbing back out.
The method works slower for the first six weeks and considerably faster thereafter. It rests on the observation that, in a defined niche, buyers already know each other, so the first client is expensive to acquire, and every subsequent one is cheap.
Start by producing something genuinely useful for your chosen sector rather than an advertisement. A checklist, a template, a short diagnostic that addresses the exact problem you identified during niche validation. Take it to the places that sector gathers: the trade association, the LinkedIn group, the regional meet-up. You are not selling at this stage; you are demonstrating that you understand their operating reality better than a generalist does.
The asset does not need to be elaborate. A one-page compliance tracker, a handover template, a costed checklist of the failure points in a process that sector runs badly: anything visibly built by someone who has done the work, rather than by a marketer. Post it where the sector already gathers rather than on your own channels, because distribution beats production at this stage. The measure of whether it worked is not downloads; it is how many conversations it starts with people who describe the problem back to you in their own words.
Once you have two clients in a niche, ask each of them directly for one introduction to a peer. In a concentrated sector, this single request is more productive than any paid channel, and it arrives pre-qualified. If you want a broader view of acquisition channels and how to sequence them, our guide to how to get your first sales and customers covers the wider framework.
Scope, boundaries and the problem that ends most VA relationships
Scope creep is not an occasional irritation in this business; it is the default trajectory of every unmanaged client relationship. A retainer is agreed for inbox and diary management, which includes inbox, diary, invoicing, social media, travel and the occasional personal errand, at the same fee, over about seven months. The client is rarely acting in bad faith. Each individual request is small and reasonable, but their cumulative effect erodes your margin.
The mistake is to treat this as a relationship problem to be managed with goodwill. It is a documentation problem. Where scope is written down in specific, countable terms, creep is visible and correctable. Where it is described loosely, it is invisible until you are working 14 hours a month beyond what you priced.
Define scope by volume and category, not by hours. Rather than “up to 20 hours of administrative support”, specify “inbox management for two mailboxes, up to 40 outbound client emails per week, diary management for one calendar, and monthly invoice raising up to 30 invoices”. Add an explicit line stating that work outside these categories is quoted separately. Then review actual volumes against the agreed figures every quarter, in writing, with the client.
The cost of not doing this compounds quietly. A £700 monthly retainer for 18 hours of work yields an effective rate of £38.89 per hour, comfortably above the £29.77 survey average [1]. Let the same account drift to 33 hours through unpriced additions, and the effective rate falls to £21.21, below the rate the same survey records associate VAs being paid for subcontracted work [1]. Nothing visible has changed. The invoice is identical, the client is happy, and you are now working at associate wages on your own account.
Moving from solo virtual assistant to an agency
There is a natural revenue ceiling to solo delivery somewhere between £45,000 and £70,000, depending on your rate and how much of your week you are willing to sell. Growing beyond it means either raising rates substantially, which requires deeper specialisation, or delivering through other people. Both are legitimate. Only one thing changes what the business is.
The common failure is subcontracting before systematising. A founder wins more work than she can deliver, brings in an associate, hands over a client, and discovers that everything she knew about that account lived in her head. Quality drops, the client notices, and the founder ends up doing the work twice. Delegation fails not because the associate is weak but because there was nothing documented to delegate.
Systematise first. Every recurring process should have a written procedure specific enough that a competent stranger could follow it without asking you any questions. This is tedious, and it is the entire foundation of a service business that can grow. Only once the procedures are in place should you bring in associate capacity, and the first handover should be your least complex client, not your largest.
The associate economics only work on top of documented processes. The UK VA Survey records associate VAs being paid an average of £21.97 an hour against a client-facing average of £29.77 [1], so the gross margin on delegated work is roughly £7.80 an hour before you account for your own time checking it. Where procedures are documented, checking time is minimal, and the margin holds. Where they are not, you spend two hours reviewing and correcting every eight hours of associate work, and the margin disappears entirely. Documentation is not administrative tidiness in this model; it is the entire source of profit on delegated delivery.
If growth of this kind is the direction you intend to take, it is worth building a proper financial model before you commit to fixed associate costs. Our guide to preparing a cash flow forecast sets out the method, as associated costs are incurred monthly while client payment terms are not.
The five mistakes I see most often
These are not theoretical risks. They are the specific failures that account for most of the struggling VA businesses that reach me, usually 18 to 24 months in.
- Positioning as a generalist. Offering everything to everyone produces price competition against a limitless supply of alternatives. It is the root cause from which most of the others follow.
- Pricing by the hour. It caps revenue, penalises efficiency and invites timesheet scrutiny instead of outcome discussion.
- Starting on bidding marketplaces. The rates you accept there become the rates you believe you are worth, and unwinding that anchor takes years.
- Operating without a written scope. Creep is invisible without documentation, and by the time it is felt, it has usually been running for months.
- Skipping ICO registration and data processing agreements. It costs £47 a year, and its absence disqualifies you from exactly the corporate clients you want.
Your first 90 days: a phased implementation plan
Work through these in order. The sequence matters more than the speed, because each phase produces the input the next one needs.
Days 1 to 30: validate and position
- Select two candidate niches and complete six discovery conversations in each. Record the specific costed problem that recurs.
- Test both against the three criteria: concentrated pain, ability to pay, and reachable community. Commit to one.
- Write a single positioning sentence naming the sector, the problem and the outcome. Every marketing asset derives from this sentence.
- Reduce your public service list to three services aligned with that problem.
Days 31 to 60: build the compliant foundation
- Decide whether to operate as a sole trader or a limited company, and register accordingly. Register for Self Assessment with HMRC if trading income will exceed £1,000.
- Complete the ICO registration self-assessment and pay the data protection fee if required.
- Open a separate business bank account and set up Making Tax Digital-compatible bookkeeping software.
- Obtain quotes for professional indemnity and cyber liability. Put a written client contract and a data processing agreement in place before you sign anybody.
- Set your retainer pricing against the cost of the problem, not against your hourly rate.
Days 61 to 90: acquire and systematise
- Produce one genuinely useful asset for your sector and distribute it through the two or three places that sector gathers.
- Convert conversations into two paying retainer clients. Ask each for one peer introduction within the first 60 days of the engagement.
- Document every recurring process from the first engagement as you deliver it, not afterwards.
- Set a quarterly scope review date with every client at the point of signing, not when creep becomes a problem.
The principle underneath all of this
A virtual assistant business succeeds or fails based on one variable, and it is not the quality of the administrative work. It is whether the buyer perceives a real alternative to you. Where the buyer sees 40 interchangeable options, they will optimise for price, and you will spend your career defending a rate. Where the buyer sees one person who has clearly done this exact work for businesses exactly like theirs, price becomes a secondary consideration, and the conversation moves to availability.
Everything practical in this guide—the niche selection, the retainer pricing, the documented scope, the sector-specific asset—is in service of that single objective. Narrow enough to be the obvious choice, systematised enough to deliver profitably, and documented enough to grow beyond yourself.
The virtual assistant businesses that thrive are not the ones that can do the most. They are the ones a specific buyer cannot easily replace.
Take the next step
If you are building a virtual assistant business and want to pressure-test the niche, the pricing model and the financial projections before you commit, our startup consulting service works through exactly this with founders at pre-launch and early trading stage. We have advised more than 2,000 businesses across 47 or more industries since 2014, and service businesses of this type are among the most common briefs we handle.
You can book a consultation directly at https://sgi.youcanbook.me/, or, if you would prefer a lower-commitment starting point, our free business resources include a business plan template pack and a startup launch toolkit you can work through on your own.
Frequently asked questions
How much can a UK virtual assistant realistically earn?
Generalist VAs billing hourly typically charge between £22 and £35 per hour, which yields £25,000 to £45,000 per year once unbillable time is accounted for. Specialists working on monthly retainers within a defined niche commonly reach an effective rate of £45 to £70 an hour because delivery is systematised and the fee is set against the value of the outcome rather than the time consumed. The difference is positioning and pricing structure, not skill.
Do I need qualifications to become a virtual assistant?
No formal qualification is required, and no professional body regulates the role. What buyers actually assess is demonstrable sector experience and evidence that you understand their specific operating context. A former practice manager selling to clinics carries more credibility than a certificate, which is one reason choosing a niche adjacent to your previous career is usually the fastest route to a viable rate.
Should I register as a sole trader or a limited company?
Most VA businesses should start as sole traders because the compliance burden is lighter and early losses are more easily offset. Incorporating becomes worth the additional cost once profits are high enough that the corporate tax and dividend route outweighs the extra administration, or when you intend to hire, or when target clients prefer contracting with a company. It is a decision worth revisiting at the end of each trading year rather than settling once.
Do I need to register with the ICO?
If you process personal data as a data controller—which includes holding client records, managing inboxes containing third-party data, or operating a CRM—then yes. The Tier 1 fee for micro-organisations is £52 a year, or £47 by direct debit [6]. Use the ICO self-assessment tool to confirm your position and treat it as a prerequisite, not an afterthought, because corporate clients check.
When do I need to register for VAT?
Registration becomes compulsory once your taxable turnover exceeds £90,000 in any rolling 12-month period, which is not the same as your financial year [9]. You must notify HMRC within 30 days of the end of the month in which you crossed the threshold. Voluntary registration below can be advantageous if your clients are VAT-registered, since they reclaim the VAT and you recover input tax on your own costs.
How long does it take to replace a full-time salary?
For founders working in a defined niche on retainer pricing, six to nine months to replace a £35,000 salary is a realistic planning assumption, based on five to seven retained clients. Generalists competing on hourly rates in open marketplaces frequently take twice as long and plateau at a lower level. The variable is not effort; it is whether the offer is differentiated enough to command a retainer.
Is it worth using freelance marketplaces at all?
They can be useful for a small number of early engagements that build testimonials and let you test delivery processes, provided you treat them as a temporary proving ground with a fixed exit date. The danger is anchoring: rates accepted there tend to become the rates you believe the market will bear. Set a limit, for example, three engagements or three months, and build your direct acquisition channel in parallel from day one.
References
[1] Society of Virtual Assistants, UK VA Survey. https://societyofvirtualassistants.co.uk/home/uk-va-survey/
[2] Royal College of Veterinary Surgeons, RCVS Facts, Practice Standards Scheme accredited premises. https://www.rcvs.org.uk/news-and-views/publications/rcvs-facts-2024/
[3] HM Revenue and Customs, Rates and thresholds for employers 2026 to 2027. https://www.gov.uk/guidance/rates-and-thresholds-for-employers-2026-to-2027
[4] Office for National Statistics, Labour market overview, UK: June 2026. https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/bulletins/uklabourmarket/june2026
[5] Information Commissioner’s Office: Fixed penalties for failure to pay the data protection charge. https://ico.org.uk/about-the-ico/our-information/fixed-penalties-for-failure-to-pay-the-data-protection-charge/
[6] Information Commissioner’s Office, Data protection fee and registration FAQs. https://ico.org.uk/for-organisations/data-protection-fee/
[7] Department for Business and Trade, Business Population Estimates for the UK and Regions, 2025. https://www.gov.uk/government/collections/business-population-estimates
[8] HM Revenue and Customs, Set up as a sole trader, and Making Tax Digital for Income Tax. https://www.gov.uk/set-up-sole-trader
[9] HM Revenue and Customs, VAT registration thresholds. https://www.gov.uk/vat-registration/thresholds
Related Posts
Kurt Graver is the founder and CEO of SGI Consultants, a business consultancy that has helped over 2,000 entrepreneurs establish successful startups using systematic business development methodologies. An accountant with an MBA and 25 years of commerce and consultancy experience, Kurt specialises in strategic planning, market analysis, and sustainable business growth