Data Select v HMRC: The Five Questions For A Late Tax Appeal

You missed the deadline. HMRC sent you a decision, you had 30 days to appeal, and that window has closed—by weeks, by months, maybe by more. The obvious question is whether the tribunal can still let you in late.

It can. The tribunal has the power to extend the time limit and admit a late appeal. But that power is a discretion, not a right. The tribunal does not have to use it, and it will only do so if you give it good reason. So the real question is: how does a tribunal decide?

The modern answer starts with one Upper Tribunal decision from 2012—Data Select Ltd v HMRC. In it, Mr Justice Morgan set out five plain questions that a court or tribunal asks itself whenever it is asked to extend a time limit. Those five questions are the foundation of the test the First-tier Tribunal applies to your late appeal today. Understand what they are and where they came from, and you can see exactly what the tribunal will be weighing in your own application.

You've Missed The Deadline—Why This Case Matters

For most HMRC decisions, the clock is 30 days from the date of the decision letter. Miss it, and you have not lost your appeal automatically—but you now need the tribunal's permission to bring it late, and permission is not a formality.

What Data Select gives you is the shape of the decision the tribunal is about to make. It is not a lottery and it is not raw sympathy. There is a recognised, principled framework, and once you understand the questions the tribunal will ask, you can answer them directly instead of guessing.

A word of honesty before we go further: in Data Select itself, the taxpayer lost. The late appeal was refused below, and the Upper Tribunal dismissed the company's challenge to that refusal. The value of the case is not its result—it is the framework Morgan J laid down on the way to it.

What Data Select Was About

Data Select Ltd was a VAT case. The company had claimed a large amount of input tax (the VAT it said it was entitled to recover on its purchases). HMRC issued a decision letter dated 28 April 2009 denying input tax of £166,250 on one particular transaction.

Under the VAT appeal rules, an appeal has to reach the tribunal within 30 days of the date of the document notifying the decision. Counting from 28 April 2009, the deadline fell on 27 May 2009. The company did not lodge a valid notice of appeal until on or around 9 April 2010—roughly ten and a half months late.

Its explanation was that it had never received the 28 April 2009 letter at the time. It said it only got hold of a copy in March 2010, and appealed shortly after. The First-tier Tribunal was not persuaded. The only evidence of non-receipt was a brief witness statement from a company representative who did not attend the hearing to be questioned on it, and the tribunal found the company had not acted with proper diligence even once it did have a copy. It refused to extend time.

The company appealed to the Upper Tribunal, where the case came before Mr Justice Morgan—a High Court judge sitting in the Tax and Chancery Chamber. This was a review, not a fresh decision: his task was to decide whether the First-tier Tribunal had made an error of law, not to substitute his own view.

He held it had applied the correct legal test, made findings that were open to it, and disclosed no error of law. As he put it, "My task is to review that decision and not to make my own independent decision." The appeal was dismissed.

So why does a case the taxpayer lost matter so much? Because permission to appeal had been granted on the footing that there was "little case law on the approach to be adopted" to extending time. Morgan J used the opportunity to state the framework clearly—and that statement has governed late tax appeals ever since.

The Five Questions

The heart of the case is a single passage. When a court or tribunal is asked to extend a time limit, Morgan J said, it asks itself five questions. In his words:

"Applications for extensions of time limits of various kinds are commonplace and the approach to be adopted is well established. As a general rule, when a court or tribunal is asked to extend a relevant time limit, the court or tribunal asks itself the following questions: (1) what is the purpose of the time limit? (2) how long was the delay? (3) is there a good explanation for the delay? (4) what will be the consequences for the parties of an extension of time? and (5) what will be the consequences for the parties of a refusal to extend time. The court or tribunal then makes its decision in the light of the answers to those questions."

In plain English, the five questions are:

1. What is the deadline for? Why does the time limit exist at all? Morgan J located the answer in finality—the idea that, after a point, the parties are entitled to treat a matter as settled and move on.
2. How late were you? The bare length of the delay.
3. Is there a good explanation? Was there a good reason the appeal was late?
4. What happens if time is extended? The consequences for both you and HMRC if the late appeal is allowed in.
5. What happens if it is refused? The consequences for both sides if it is not.

Then the tribunal "makes its decision in the light of the answers to those questions". That last clause matters as much as the five questions themselves. This is a balancing exercise: the tribunal weighs all the answers together against the overriding objective—the duty to deal with cases fairly and justly—and reaches a single, overall judgement.

It is not a checklist where four ticks out of five wins. A short delay can still be refused if there is no good explanation; a long delay can occasionally be forgiven if the reasons are compelling and the stakes are high.

And one answer sits in the background of every application: time limits exist for a reason. Morgan J stressed that the desirability of not reopening matters that the other side is entitled to regard as closed applies to a challenge against an HMRC decision just as it does to an appeal against a court judgment. The deadline is not red tape. It is the point at which HMRC, and the system, are entitled to assume the matter is over.

Where The Test Came From—And Where It Went

Data Select was decided in 2012, and it is important to be honest about its age, because the law around late applications shifted significantly just after it.

Morgan J drew his approach partly from the old version of a civil-court rule on "relief from sanctions"—the rule that governs when a court forgives a party's failure to comply with a deadline or direction. In 2012, that rule contained a long checklist of factors. Morgan J cross-referred to it and to the overriding objective, and treated the five questions as consistent with it.

Then the ground moved. The relief-from-sanctions rule was rewritten in 2013 into a much stricter, stripped-down form, and two Court of Appeal decisions interpreted it: Mitchell v News Group Newspapers Ltd [2013] EWCA Civ 1537 (notoriously harsh), and then Denton v TH White Ltd [2014] EWCA Civ 906 in 2014, which softened Mitchell and recast the exercise as a three-stage test. All of this happened after Data Select. So Data Select did not, and could not, apply the Denton three-stage test—it predates it.

The case that brought the modern thinking into the late-tax-appeal context is Martland v HMRC [2018] UKUT 178 (TCC). Martland quoted Morgan J's five questions in full and approved them. It then mapped them onto the Denton framework, recasting the exercise into three stages:

1. Establish the length of the delay. If it was very short, the tribunal "is unlikely to need to spend much time on the second and third stages."
2. Establish the reason (or reasons) the default occurred.
3. Evaluate all the circumstances of the case—a balancing exercise that weighs the merits of the reasons for the delay against the prejudice to both parties of granting or refusing permission.

You can see the five questions living inside those three stages. The "length of delay" question is stage one. The "good explanation" question is stage two. The "purpose of the time limit" and the two "consequences" questions are folded into the stage-three balancing exercise. Martland said so itself: approached this way, "all the factors raised in Aberdeen and Data Select will be covered, without the need to refer back explicitly to those cases". It added the line that anchors the whole exercise: "The FTT's role is to exercise judicial discretion taking account of all relevant factors, not to follow a checklist."

That is the precise relationship. Data Select is the foundational statement of the underlying questions. Martland is the operative test the First-tier Tribunal applies today. The five questions live inside Martland's three stages. If you want the three-stage test unpacked in detail, with the supporting case law, our Martland analysis is the companion to this one.

Is Data Select Still Good Law?

Yes. Martland did not overrule Data Select—it absorbed it. The five questions were endorsed, not discarded, and Data Select remains the source statement that tribunals cite when they explain where the modern approach comes from. It is foundational and still good law.

What changed is the structure you will see applied to your own application. Today the tribunal will work through the three Martland stages rather than reciting Data Select's five questions one by one. But the factors are the same, because Martland built directly on them.

And Martland itself is secure. It was challenged in 2025 but reaffirmed by the Court of Appeal in January 2026, which described its three-stage approach as an unimpeachable one. The full story of that challenge—and what it means—is in our Martland analysis; for our purposes here, the short point is that the framework resting on Data Select is firmly the law.

What This Means For Your Late Appeal

The five questions are not just legal theory. They are the structure of the story you need to tell the tribunal. Here is how each one translates into what goes into a late-appeal application—described, so you can see how the tribunal weighs things, not prescribed.

Be precise and honest about how late you were. The tribunal dates the delay from the deadline, not from when you got round to dealing with it. Understating it does not help; the tribunal will work it out anyway. A short delay may need little justification. A long one—Data Select's ten and a half months, say—needs a compelling account that covers the whole period, not just how it started.

Explain why—and back it up. The "good explanation" question is where most late appeals are won or lost. A reason that is specific, evidenced, and covers the entire delay carries far more weight than a vague one. Data Select is the cautionary tale: the company's only proof that it never received the letter was a brief witness statement from someone who did not turn up to be questioned on it, and that evidence did not hold. If your reasons rest on facts—illness, a missed letter, an adviser's failure—the tribunal will look for evidence behind them, not just assertion.

Address the consequences both ways. The fourth and fifth questions ask what happens to each side if time is, or is not, extended. Refusal usually means you lose the right to challenge the tax at all—that is real prejudice to you. But the tribunal also weighs HMRC's legitimate interest in finality, and the importance of statutory time limits being respected. A realistic application acknowledges both sides of that ledger rather than pretending the deadline does not matter.

The more serious the delay, the better the explanation must be. These limbs interact. A modest delay with a solid reason is a very different application from a long delay with a thin one. The longer you were late, the higher the bar your explanation has to clear.

"I didn't know about the deadline" rarely succeeds on its own. HMRC's appealable decisions generally set out the appeal rights and the time limit in plain terms. Not knowing, by itself, is seldom treated as a good explanation. It is the kind of answer the framework is designed to test rather than accept.

Mention the merits, but only briefly. The strength of your underlying case carries little weight at this stage unless it is obviously very strong or very weak on a quick look. Do not turn the late-appeal application into a full trial of the tax dispute. A short paragraph on why HMRC's decision looks wrong is appropriate; a lengthy merits argument is counterproductive.

The route differs by tax, but the test is the same. For direct taxes (income tax, capital gains tax, and the like), the first step is to ask HMRC to accept a late appeal under section 49 of the Taxes Management Act 1970—HMRC must accept it if you had a reasonable excuse for the delay and applied without unreasonable delay once that excuse ended. If HMRC refuses, you apply to the tribunal. For VAT and other indirect taxes, that HMRC route does not exist—only the tribunal can admit a late appeal. Either way, the tribunal applies the Martland/Data Select framework. The mechanics are in our late appeals guide, and our how to appeal guide covers the wider process.

You make the application inside the appeal itself—there is no separate form. Under Rule 20(4) of the Tribunal's rules, when you give a notice of appeal after the deadline, the notice must say that it is late and must include a request for permission to make it late. In practice that means completing the ordinary notice of appeal—the GOV.UK appeal form or the online service—as normal, then adding a section that states the appeal is late, says exactly how late, and explains why. That explanation is where you answer the five questions in your own words. You do not wait to be asked; you put the case for admitting the appeal in the document that starts it.

And if the tribunal refuses? A refusal to admit a late appeal is itself a decision, and you can challenge it—but only on a point of law, and only with permission, by an onward appeal to the Upper Tribunal. That is a high bar. The Upper Tribunal will not step in simply because it might have exercised the discretion differently; it asks whether the First-tier Tribunal made an error of law, in the same way Morgan J reviewed the decision in Data Select itself. The realistic position is that the application you make first time is the one that decides the outcome—which is exactly why it is worth answering all five questions properly. What happens next is covered in after your tax tribunal decision.

None of this guarantees a late appeal will be admitted. The discretion is structured and principled, and "it's not fair" on its own is not an argument the framework recognises. What it does mean is that you can answer the tribunal's questions deliberately—and an application that engages with all five, honestly and with evidence, gives itself the best chance the framework allows.

Key Legislation And Resources

The Judgment

• Data Select Ltd v HMRC [2012] UKUT 187 (TCC) (GOV.UK tribunal decisions) — Upper Tribunal (Tax and Chancery Chamber); Mr Justice Morgan; the five-questions framework is at paragraph [34].

Legislation

• Section 49, Taxes Management Act 1970 — late appeals for direct taxes (reasonable excuse; without unreasonable delay; tribunal route if HMRC refuses)
• Section 83G, Value Added Tax Act 1994 — the 30 days VAT appeal limit and the power to extend time at issue in Data Select
• Rule 20, Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 — what a late notice of appeal must say (that it is late, and why permission should be given)
• Rule 5(3)(a), Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 — the tribunal's general power to extend time for complying with a rule or direction, unless another enactment sets the limit

Key Cases

• Data Select Ltd v HMRC [2012] UKUT 187 (TCC) — the foundational statement of the five questions a tribunal asks before extending a time limit
• Martland v HMRC [2018] UKUT 178 (TCC) — the operative test today; recasts the five questions into the three-stage Denton framework the First-tier Tribunal applies
• Mitchell v News Group Newspapers Ltd [2013] EWCA Civ 1537 — the first, notoriously strict interpretation of the rewritten relief-from-sanctions rule, later softened by Denton
• Denton v TH White Ltd [2014] EWCA Civ 906 — the Court of Appeal's three-stage relief-from-sanctions test that Martland adopted
• BPP Holdings v HMRC [2017] UKSC 55 — Supreme Court approval of structured guidance on respecting tribunal time limits

On This Site

• Martland v HMRC: the three-stage late appeal test — the successor case and the test applied today; read this alongside Data Select
• Late appeals to the tax tribunal — the step-by-step guide to filing a late appeal
• How to appeal to the tax tribunal — the wider filing process, direct and indirect tax routes
• Writing grounds of appeal — how to put the merits of your underlying case briefly at stage three
• Understanding your HMRC appeal rights — what the 30 days deadline is and where it comes from
• HMRC internal review — the 45 days review route, often missed alongside the appeal deadline
• BPP Holdings v HMRC — relief from sanctions and tribunal directions in the tax context
• After your tax tribunal decision — what happens if your late-appeal application is refused

This article is for informational purposes only and does not constitute legal or tax advice. For advice specific to your situation, consult a qualified tax adviser, accountant, or solicitor.