MTD ITSA Cost Per Client: What It Really Costs a Practice
The MTD ITSA cost per client for a practice has three parts: compatible software, a one-off sign-up and digital-record set-up, and the recurring work of four quarterly updates plus a final declaration where one annual return used to sit. Software is the small line and the one everyone quotes; the recurring coordination is the large one, and no vendor can price it for you because it depends on how late your clients send things in.
The first cohort is now two quarters into it, so this is no longer a preparedness question. Below is what the obligation actually costs a small practice per client, how to work the number out from your own records, and where automation genuinely removes cost rather than moving it.
What is the MTD ITSA cost per client for a practice?
Making Tax Digital for Income Tax Self Assessment (MTD ITSA) is the requirement for qualifying sole traders and landlords to keep digital records and send HMRC quarterly summaries of income and expenses through compatible software, followed by a final declaration after the tax year ends. It replaced one filing event per client with five.
| Cost layer | What it covers | What it looks like | What moves it |
|---|---|---|---|
| Compatible software | The ledger or bridging tool that actually transmits to HMRC | List prices sampled 8 September 2026 ran from roughly £30 a year for bridging tools to £15–£50 a month per client for a full cloud ledger | Whether the client keeps their own records, and whether you need a full ledger or only a submission route |
| Sign-up and set-up | Agent authorisation, signing the client up, opening balances, chart-of-accounts tidy-up, showing them how to capture records | One-off, per client, front-loaded into the first year | The state of the records you inherit. A shoebox client and a Xero client are different products |
| Quarterly updates | Four submissions a year, each preceded by collecting, categorising and chasing | Recurring, four times a year, per client | Record quality, not client complexity. This is the line that surprises practices |
| Final declaration | The year-end adjustments and the declaration itself, due 31 January | Recurring, once a year, per client | Broadly the old annual job, minus some of the catch-up bookkeeping |
| Exception handling | Corrections, missed deadlines, clients who go quiet, HMRC service issues | Unpredictable and unbillable unless you priced for it | How many clients you have, not how complicated any one of them is |
Two of those five lines scale with the number of qualifying clients rather than with the difficulty of their affairs. That is the structural change, and it is why a practice can add MTD ITSA fees, work more hours, and still be less profitable per client than it was.
How much extra work is it, precisely?
Quarterly updates are summaries, not tax returns. They report totals by income and expense category rather than individual transactions, and they are submitted through compatible software. For standard quarters, they are due one month and seven days after each quarter end — 7 August, 7 November, 7 February and 7 May — and the final declaration is due by 31 January following the end of the tax year (ICAEW TAXguide 01/25: MTD income tax, checked 8 September 2026).
Mandation began on 6 April 2026 for qualifying income above £50,000. Qualifying income is gross self-employment turnover plus gross property income combined, before expenses. The threshold falls to £30,000 from April 2027 and £20,000 from April 2028 (GOV.UK, Find out if and when you need to use Making Tax Digital for Income Tax, checked 8 September 2026). Partnerships remain deferred with no start date announced, and partnership income does not count towards the qualifying-income threshold — though a partner with their own rental or sole-trade income may still be in scope.
There is one number worth holding onto. HMRC's newsroom reminded more than 864,000 sole traders and landlords that the rules started in April 2026 (HMRC press release, "Act now", checked 8 September 2026). After the first deadline, HMRC announced that more than 436,000 had submitted their first quarterly update (GOV.UK news, 12 August 2026, checked 8 September 2026). Those two figures are not a compliance rate — exemptions, deferrals and non-standard quarter dates all sit between them — but the gap is large enough that a practice should not assume every qualifying client on its book has filed.
How do you work out your own cost per client?
Four numbers, all of which are in your own systems rather than in anyone's buyer's guide.
- Your qualifying client count. Not your sole trader and landlord count — the ones whose gross self-employment and property income combined exceeded the threshold on the return HMRC used to test them.
- Recorded time on one quarterly update, at your current record quality. Take the median across a real quarter, not the best case. Include the chasing, not just the submission.
- Your recovery rate on this work. Time recorded against time billed. If you do not measure it, that is the finding.
- Your charge-out rate for whoever actually does it. Usually not the partner rate, and that matters to the answer.
Then: annual cost per client is roughly (time per update × 4 × rate) + final declaration time × rate + software. Compare it against the fee you are charging that client today, not against the fee you charged before April 2026.
The worked example below uses assumed inputs to show the shape of the arithmetic. They are assumptions for illustration, not benchmarks — replace every one with your own figure before drawing any conclusion.
| Line | Assumed input | Annual effect per client |
|---|---|---|
| Time per quarterly update | 45 minutes, including the chase | 3 hours a year |
| Charge-out rate for the person doing it | £45 an hour | £135 a year |
| Final declaration | Broadly the old annual job | Unchanged from before |
| Software | £180 a year | £180 a year |
| Recovery rate on the quarterly work | 70% | £40 a year of the above is unbilled |
| Indicative added cost if the fee did not move | — | Around £315 a year per client |
Run it at 30 minutes and at 90 minutes per update and the answer moves by more than the entire software line. That is the point: the software price everybody publishes is the least important number in the calculation, and the one nobody publishes — your median time per update at your actual record quality — decides everything.
Be sceptical of the fee benchmarks circulating in this category, including any range quoted with confidence and no method behind it. A single afternoon's sampling on 8 September 2026 turned up per-quarter fee guidance, per-month fee guidance and annual-uplift guidance that did not reconcile with one another, all published by parties with something to sell. None of it is repeated here as fact.
What does automation actually remove?
Not the filing. Compatible software already does that, and swapping one submission tool for another rarely changes the economics. What automation removes is the coordination layer sitting on top: knowing which of your qualifying clients have supplied records for the quarter that is closing, which have gone quiet, which are three days from a deadline, and who needs chasing today rather than next week.
That work scales with client count multiplied by four quarters. It is repetitive, it is low-judgement, and it is exactly what a tracked, automated deadline and chase workflow is for — which is what our Deadline Autopilot product does for HMRC and Companies House dates, at £1,800 setup plus £50 a month, fully managed. A practice wanting the coordination built into its own CRM and workflows is in Practice Autopilot territory, from £4,000.
The honest sequence, though, is that neither is where you start. Ihsan Ops is a UK AI automation agency based in Bedford, and every engagement begins with an AI Opportunity Audit from £1,000 — a structured review that establishes which processes are worth automating before anything is built. A single scoped bespoke workflow runs £2,000–£5,000 over two to four weeks, with £100–£600 per workflow per month in platform and model costs at production volume. Quoting a fixed price for work that has not been scoped would be guessing, and the audit is how the guessing stops.
If you want to test the coordination problem before spending anything, our free Companies House deadline tracker covers the same mechanic on a different deadline set.
When is this the wrong spend?
Four situations where a practice should not buy automation for MTD ITSA yet.
- A small qualifying book. Below roughly fifteen qualifying clients, a shared spreadsheet and calendar reminders genuinely still work, and the build cost will not clear the saving.
- Mid-migration. If you are moving practice-management systems this year, automate after the move, not across it.
- The real problem is client behaviour. If records arrive three weeks late because clients are not capturing them, the fix is onboarding, expectations and possibly a different client, not a workflow.
- The fee conversation has not happened. Automating an unpriced service makes an unprofitable client cheaper to serve while leaving it unprofitable. Price first, then automate.
What changes in April 2027 and April 2028?
Two things, and the second is easy to miss. First, the threshold drops to £30,000 in April 2027 and £20,000 in April 2028, which for most small practices means the qualifying book grows substantially rather than incrementally.
Second, the penalty easement does not travel with it. HMRC confirmed that taxpayers joining in April 2026 will not receive penalty points for late quarterly updates in their first year, but this does not cover the 2026/27 final return due 31 January 2028, does not cover late payment, and does not apply to the April 2027 and April 2028 cohorts. Under the points regime, a quarterly filer reaching four points receives a £200 penalty (ICAEW, MTD for income tax penalties, checked 8 September 2026). Clients who are genuinely digitally excluded can apply to HMRC for exemption, which is assessed case by case (ATT, Making Tax Digital: exemption cases, when and how to apply, checked 8 September 2026).
In practice that means a process which coped in 2026/27 because nothing was penalised may not cope in 2027/28 on a larger book with points attached. This post describes the obligations and cites the sources; how they apply to a particular client is a matter for your own professional judgement and, where needed, your own adviser.
Frequently asked questions
What does MTD ITSA cost a practice per client? Three lines make up the bill: compatible software, the one-off sign-up and digital-record set-up, and the recurring work of four quarterly updates plus a final declaration. Software list prices sampled on 8 September 2026 ran from about £30 a year for bridging tools to £15–£50 a month for a full cloud ledger. The recurring work is the larger cost and only your own time records can price it.
How many submissions does MTD ITSA add per client per year? Four quarterly updates plus a final declaration, where annual self assessment needed one return. Quarterly updates are due one month and seven days after each quarter end — 7 August, 7 November, 7 February and 7 May for standard quarters — and the final declaration is due by 31 January after the tax year ends.
Are there penalties for a late quarterly update in 2026/27? HMRC has confirmed an easement: taxpayers joining in April 2026 do not receive penalty points for late quarterly updates in their first year. It does not extend to the 2026/27 final return due 31 January 2028, it does not cover late payment, and it does not apply to the April 2027 and April 2028 cohorts. Confirm your client's position with HMRC guidance.
Should a small practice charge more for MTD ITSA clients? That is a commercial decision, not a rule, but the arithmetic is unavoidable: five submission events replace one, and the coordination around them scales with client count rather than complexity. Work out your cost per qualifying client from your own time records before quoting, and price the chasing separately from the filing if record quality varies across the book.
When is automating MTD ITSA coordination not worth it? When the qualifying client count is small enough that a spreadsheet and a calendar still work, when the practice is mid-migration between practice-management systems, or when the real problem is that clients supply records late and no software fixes that on its own. Automation removes coordination overhead; it does not remove the underlying bookkeeping work.
Where to start
Pull the recorded time on one full quarter of updates before you price anything or buy anything. If the median is close to your assumption, your fees are probably about right and the coordination is the only thing worth automating. If it is double, you have a pricing problem that no software will solve, and that conversation comes first.
If you want a second opinion on which part of the MTD ITSA workload is actually worth automating in your practice, book a 30-minute discovery call.