Generic Maths Ltd v Revenue and Customs (VAT - Whether online maths product involving provision of "assessments" was a supply of exempt examination services) [2025] UKFTT 460 (TC) (24 April 2025)

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] [DONATE]
	First-tier Tribunal (Tax)
You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Generic Maths Ltd v Revenue and Customs (VAT - Whether online maths product involving provision of "assessments" was a supply of exempt examination services) [2025] UKFTT 460 (TC) (24 April 2025) URL: https://www.bailii.org/uk/cases/UKFTT/TC/2025/TC09497.html Cite as: [2025] UKFTT 460 (TC)

DECISION

1. We decided that the appeal should be dismissed.

introduction and Summary

2. Generic Maths Limited ("GM") appeals against a decision by The Commissioner of His Majesty's Revenue and Customs ("HMRC") to issue an assessment under s73(1) Value Added Tax 1994 ("VATA") for the periods 10/14 to 07/18. The assessment (the "Original Assessment") was originally in the sum of £103,538 and was notified on 12 September 2018. Following a review, on 15 March 2019, the assessment was amended to the sum of £80,118.88. In this judgment the "Assessment" refers to that amended assessment.

3. The essence of GM's appeal was that the "ConquerMaths" product they supply is predominantly a tool that provides assessments enabling those using the product (referred to in this judgment as "pupils") or their parents/teachers to determine what level of maths ability the pupil has reached and identify any gaps in knowledge. As such, GM argued that this product is an exempt supply for VAT purposes, since it falls into the category "examination services" exempt under Item 3 of Group 6 Schedule 9 VATA. HMRC argued that the product was an online mathematical tutorial tool, not a form of examination services; and accordingly the supply made by GM was standard rated. While we accepted that the GM product included diagnostic tests, we did not consider that the product was a supply of examination services within Item 3. We found that the Assessment had been made using best judgment by HMRC and accordingly that the appeal should be dismissed.

4. Dealing with each of the key points in turn we decided:

(1) The correct test for deciding the character of GM's supply was the consumer perception test (see [22]), and based on that test, ConquerMaths was a form of tutorial tool or revision aid (see [45]);

(2) If the correct test for characterisation was not the consumer perception test, but instead some sort of functional test or expert perception test, the result was the same, because objectively ConquerMaths was a form of tutorial tool or revision aid (see [45]);

(3) The term "examination services" in Item 3, is wider than formal public examinations, but not so wide as to include something that is a form of tutorial tool or revision aid (see [49]);

(4) The word "assessments" in Note 4 of Group 6 cannot be read independently of the rest of Note 4 and takes context from the rest of the words of Note 4 (see [52]);

(5) Accordingly, the word "assessments" does not allow for a form of tutorial tool or revision aid to be brought into the definition of "examination services" (although we considered something short of a public exam might be capable of falling within the definition) (see [59]);

(6) Merely because a supply is, or is described as, an "assessment" does not mean it will fall into the definition in Note 4 - rather the nature of the supply needs to be considered to determine what it is and what it is perceived as (see [59]);

(7) The Assessment had been made using best judgment, HMRC having fairly considered all material placed before them and, on that material, come to a decision which was reasonable and not arbitrary (see [69]); and

(8) We did not have jurisdiction to deal with GM's argument that they had a legitimate expectation that the supply of ConquerMaths would be treated as exempt, but even if we had had, the test for showing a legitimate expectation was not remotely met (see [72]).

5. At the hearing GM was represented by Mr Khokhar and HMRC by Ms Donovan.

Evidence

6. We heard oral evidence from Mr Alan Constable (Chair of GM), Mr Richard Hunter (Head of Marketing at GM), Mr Khokar and Officer Lynzie Stephenson of HMRC.

7. In addition, we had the following documents:

(1) A bundle of hearing documents produced by HMRC of 353 pages (including the "Barbados Study" referred to in more detail below);

(2) An authorities bundle produced by HMRC of 281 pages;

(3) A supplemental bundle prepared by GM of 34 pages;

(4) Skeleton arguments from HMRC of 8 pages and from GM of 9 pages; and

(5) Two "screen grabs" from GM's website.

8. We found Officer Stephenson to be a credible and reliable witness. Generally we found the GM witnesses (who were witnesses of fact) to be credible, but at times we considered that their desire to argue that ConquerMaths is an exempt supply of examination services led them to express opinions that were not issues of fact and gave rise to inconsistencies in their evidence which went to the question of its reliability as evidence (as is set out in more detail below).

Findings of fact

9. We found the following facts which were not disputed:

(1) GM is a taxable person for VAT purposes, registered with VAT number 894 9550 60;

(2) On 11 October 2017 Officer Stephenson undertook a VAT Assurance check by visiting GM's premises with her colleague Mr Rob Illingworth. During the visit Officer Stephenson met with Mr Hunter and Mr Alister Brown (an accountant from Addere Valorem);

(3) During the visit Officer Stephenson viewed GM's product and had a demonstration of it. She subsequently viewed GM's product on their website, although she did not log into the system in the way a paying customer would have done. In considering the nature of the product, she consulted with colleagues and examined a number of sources of HMRC guidance (including VAT notice 701/30) and case law (including The Commissioners for Her Majesty's Revenue and Customs v Metropolitan International Schools Ltd [2017] UKUT 431 (TCC) ("MIS")) to reach a view on the nature of GM's supply;

(4) GM had treated some of their supplies as exempt examination services. Officer Stephenson formed the view that GM's supply was a composite supply the predominant element of which was education and that the supply was not one of examination services. In consequence she considered GM's VAT returns were incorrect and therefore she issued the Original Assessment;

(5) The Original Assessment (and the Assessment) was made in respect of accounting periods 10/14 to 07/18 inclusive and the Original Assessment was notified to GM on 12 September 2018, that being within four years of the prescribed accounting periods.

10. We also found certain facts in relation to the nature of GM's product and these are set out below for convenience under the heading "What is GM's Product?".

11. We were not given convincing evidence as to what proportion of GM's supplies were to eligible bodies and/or to pupils who were receiving exempt education or training and what proportion was to other persons (such as pupils not receiving exempt education or persons outside the UK) so it was not possible for us to make findings of fact on that issue. It was clear that some of GM's supplies are to customers outside the UK (the "Barbados Study" made that plain). HMRC's case was that GM's supplies were not examination services. HMRC did not argue that GM's supplies were not to eligible bodies and/or pupils who were receiving exempt education or training; accordingly, HMRC impliedly accepted that if the GM's supplies dealt with in the Assessment were examination services, they would be exempt from VAT. Therefore it was not necessary for us to consider the issue of whether the customers were eligible bodies and/or to pupils who were receiving exempt education or training and where the consumers of GM's services were located.

Law

Legislation

12. Under s4 VATA, VAT is to be charged on the supply of goods or services made in the United Kingdom by a taxable person in the course or furtherance of a business. Generally supplies are standard rated, but certain supplies are exempt by reason of s31 VATA. Schedule 9 VATA sets out groups of exempt supplies of goods and services. Relevant to this appeal is Group 6 - Education which states:

Item No.

(1) The provision by an eligible body of–

(a) education; [...] or

(2) The supply of private tuition, in a subject ordinarily taught in a school or university, by an individual teacher acting independently of an employer.

(3) The provision of examination services—

(a) by or to an eligible body; or

(b) to a person receiving education or vocational training which is—

(i) exempt by virtue of items 1, 2, 5 or 5A; or

(ii) provided otherwise than in the course or furtherance of a business.

13. Note 4 of Group 6 states:

"Examination services" include the setting and marking of examinations, the setting of educational or training standards, the making of assessments and other services provided with a view to ensuring educational and training standards are maintained.

14. Since GM is a taxable person, it follows that, unless GM's supplies are "examination services" within the meaning of Item 3, the supplies will be standard rated. Even if the supplies are examination services, they will only be exempt if supplied to a person within (3)(a) or (b).

15. Section 73(1) VATA provides:

"Where a person has failed to make any returns required under this Act (or under any provision repealed by this Act) or to keep any documents and afford the facilities necessary to verify such returns or where it appears to the Commissioners that such returns are incomplete or incorrect, they may assess the amount of VAT due [...] to the best of their judgment and notify it to [the person]."

16. Section 77 VATA 1994 provides that a section 73 assessment must not be made more than four years after the end of the prescribed accounting period.

Case law - Best judgment

17. In relation to the meaning of "best judgment", in Van Boeckel v Customs and Excise Commissioners [1981] 211 ER, Woolf J stated:

"What the words 'best of their judgment' envisage, in my view, is that the Commissioners will fairly consider all material placed before them and, on that material, come to a decision which is reasonable and not arbitrary as to the amount of tax which is due. As long as there is some material on which the Commissioners can reasonably act then they are not required to carry out investigations which may or may not result in further material being placed before them."

Case law - education exemption - Consumer perspective and nature of supplies consisting of more than one element

18. In Finance and Business Training Ltd v Revenue and Customs Commissioners [2016] EWCA Civ 7, the Court of Appeal observed that "the question whether a person is entitled to the education exemption is objective, to be determined if necessary by the court."

19. In The Commissioners for Her Majesty's Revenue and Customs v Metropolitan International Schools Ltd [2017] UKUT 431 (TCC) ("MIS"), the Upper Tribunal considered the characterisation of supplies consisting of more than one element in the context of whether distance learning services to customers should be treated as a single zero rated supply of books or as standard rated for VAT purposes. The court considered the relevant case law, including in particular the European case Mesto Zamberk v Financni reditelstvi v Hradci Kralove [2014] STC 1703 ("Mesto"). The UT concluded that:

(1) the Mesto "predominance test" should be the primary test to be applied in characterising a supply for VAT purposes;

(2) if that can be done, the predominant element characterises the supply;

(3) the characterisation is to be assessed objectively;

(4) matters must be viewed through the eyes of the typical consumer.

20. We understood from MIS that the Mesto predominance test requires all the circumstances in which the supply takes place to be taken into account in order to ascertain the characteristics of the supply and the identification of its predominant elements. This required an overall assessment of the nature of the supplies, taking into account the qualitative and quantitative of the elements of the supply.

21. We also noted that in MIS, having concluded that the supplies in that case did not fall to within the category of zero-rating supplies, the Tribunal stated:

"It is unnecessary to go further and identify another characteristic which does characterise the supply...but if we had to do so we would characterise the supply, in the eyes of the typical consumer, as the supply of educational services."

22. Our conclusion was that the test we should use for determining the character of GM's supplies was an objective test, determining how their supplies would be characterised by the typical consumer.

Burden of proof

23. We approached the issues on the basis that it was for HMRC to show that the Assessment was valid, in time and to their best judgment. We so found, see [9(5)] and [72]. The burden is then on GM to show the correct amount of tax due (in other words that their supplies are examination services).

What is GM's product?

24. We were given a considerable amount of written and oral evidence about what exactly GM's product was. Some of the detail is set out below, but summarising and aggregating the evidence from all three of the GM witnesses we found that:

(1) The GM product is an online offering;

(2) It does not lead to any particular qualifications.

(3) The pupils using it can drop in and out of the offering (unlike the way they might have to proceed if following a course leading to a qualification);

(4) The product includes many hundreds of available diagnostic tests that test students' knowledge of the principles that will be taught on the various subjects relevant to age profile of the pupils undertaking the diagnostic tests;

(5) The product also includes a number of short videos, although the number of videos is small in comparison to the number of diagnostic tests. These videos provide tutoring;

(6) The videos were described as providing "remedial" teaching but we did not think that this was a helpful adjective. The videos provide teaching and the only sense in which the teaching is "remedial" is that the pupil might seek out the video upon encountering a question to which they did not know the right answer. Possibly the word "remedial" might be significant in that it implies some users of ConquerMaths might be using it for revision purposes in the way that a person who has been studying for an exam might do a past paper and then do remedial work to fill gaps that were shown. This would perhaps tend to show ConquerMaths was a revision aid;

(7) The average user spends 75 minutes on diagnostic tests compared to 5 minutes on videos;

(8) GM's witnesses described the product as diagnostic assessments, formative assessments, and summative assessments; and

(9) In addition to online diagnostic tests, the product includes worksheets in an exam format. Pupils are encouraged to complete these offline and then feed the results into the system.

25. GM argued that their product was an assessment led resource which enabled a pupil's level of attainment, and any shortcomings, to be identified so that the pupil could then get any necessary teaching from other sources (for example the ConquerMaths remedial videos, other online sources of teaching or their school teacher or private tutor). On that basis GM argued that their product was exempt as a form of "examination service" within Item 3(b) of Group 6.

26. HMRC argued that GM's supplies are not exempt examination services on the basis the supply is of an online learning or teaching product. HMRC asserted that it is a supply of educational services rather than an examination service.

27. One perhaps noteworthy point is that none of the evidence we were given suggested ConquerMaths was a form of examination. We considered this was unhelpful to GM's argument that ConquerMaths was a supply of examination services since the phrase "examination services" is the one used in Item 3.

28. Because the word "assessments" is used in Note 4 we have taken care in this judgment when we describe ConquerMaths generally to avoid the use of that word and instead to use the neutral term "diagnostic tests" (which happens to be the term used by GM themselves on the web page referred to at [40]). The GM witnesses used the word "assessments" to describe ConquerMaths; however, one of the issues we had to consider was the meaning of the word "assessments" in Note 4 so it was preferable to avoid using that word when we describe ConquerMaths since the meaning of that word needed to be considered independently of establishing the facts about what ConquerMaths is. Nothing should be read one way or the other into our use of a neutral term to describe ConquerMaths.

Mr Constable's evidence

29. Mr Constable's evidence was that in the educational landscape, it is essential to distinguish between a "course" that leads to a formal qualification and "supplementary resources" that support the learning process. Mr Constable's evidence was that a course typically includes a structured curriculum, instructional content, and formal assessments, leading to a recognised qualification. In contrast, supplementary assessment resources are designed to support and enhance learning by helping students identify gaps in their understanding, thereby guiding further study. His evidence was that ConquerMaths is a supplementary resource. This was in contrast to another venture he had been involved in called OnLine Education Ltd, which involved the development and delivery of hybrid and fully online degree programmes in collaboration with UK universities.

30. We accepted his evidence that there is a difference between courses that lead to formal qualifications and supplementary resources that support the learning process and that ConquerMaths is a supplementary resource. However, it did not seem to us that the distinction Mr Constable sought to draw between courses that lead to formal qualifications and supplementary resources assisted GM in their argument that ConquerMaths is an exempt examination service. Indeed, we considered that the absence of any formal qualification being provided by GM as part of its offering tended if anything to point away from its product being an examination service because in our judgment a typical examination service will lead to a qualification which the ConquerMaths product did not.

31. Mr. Constable gave evidence in relation to the question of the standard to which ConquerMaths assessed. It was put to Mr Constable that the essence of an examination was an assessment to an objective third party standard. Mr Constable's evidence was that ConquerMaths assessed to the standard of the answer being right (and in this respect he asserted that maths differed from other subjects in that there was always a right answer). Mr Constable's evidence was that for each age group ConquerMaths provided age appropriate assessments of relevant topics and by doing assessments the pupil could determine if they were getting the answers for those topics correct and identify any areas requiring teaching. Mr. Constable accepted that this was not an assessment to a third-party standard. We accepted that ConquerMaths's diagnostic test questions all have a right answer. But it did not seem to us that this bore helpfully on whether ConquerMaths was an examination service. In our judgment an examination assesses to an objective (often third party) standard and simply getting a correct answer in a maths diagnostic test does not show a pupil has taken part in an examination.

The Barbados Study

32. One part of the dispute between HMRC and GM was what conclusions should be drawn from a report of a pilot of ConquerMaths in Barbados, which had been put into evidence by GM. It was noteworthy that the feedback made almost no reference to assessments but rather included phrases that suggested the teachers and pupils giving feedback believed the product was providing teaching. The feedback included phrases such as "the work that they give to help us learn", "I think ConquerMaths was fun and educating", "I think ... ConquerMaths ... teaches me more", "ConquerMaths helps when your teacher was not around", "the video tutorials on ConquerMaths have been very insightful and helpful", "I think that is a helpful programme and I really like the way the tutorials are set up". In the teacher's questionnaire the overwhelming majority of the questions focused on ConquerMaths as a teaching solution, e.g. parts of the questionnaire headed "quality of the lessons". HMRC argued that this showed the perception of students and teachers was that the product was teaching. When taken to this feedback Mr. Constable's evidence was that the reason the feedback did not mention assessments but instead focused on learning was because much of the feedback was from primary school children who would not understand the nature of the product. That did not explain why the feedback from teachers focussed on the product as a means of teaching. We did not find Mr Constable's evidence on this issue to be convincing. In our judgment, the reason the feedback focused on teaching was because ConquerMaths is a teaching product and consequently the feedback that was sought and provided was as to how effective it was as a teaching product in the eyes of pupils and teachers.

33. Mr Khokhar also gave evidence about this issue. Towards the end of his evidence for the first time he asserted that although the product itself was the same, the way ConquerMaths was used in the Barbados pilot was not the same as the way in which ConquerMaths is typically used. Mr Khokhar asserted that:

(1) the pilot was looking at whether ConquerMaths could be used with a greater scope than typically used;

(2) the typical use is for assessment, followed by a limited number of remedial lessons, followed by further assessment;

(3) the product was chosen for this pilot because of similarities between the UK and Barbadian education system and a desire to drive up the standard of maths education in Barbados;

(4) the pilot had sought to provide data as to whether ConquerMaths could be used as a resource for all schools in Barbados; and

(5) the pilot was exploratory and was not eventually rolled out more widely.

34. We accepted that the data for the use of ConquerMaths in Barbados did relate to a pilot but we were not convinced by Mr Khokhar's evidence that the pilot proposed the use of ConquerMaths in a way that was fundamentally different to the way the product is marketed and used elsewhere. We noted that this alleged difference put forward by Mr Khokhar was not mentioned in his witness statement or by the other witnesses and was only brought up by Mr Khokhar towards the end of his evidence as he sought to explain why the feedback from the pilot appeared to focus on teaching. We found as a fact that the product used in Barbados was the same as that used elsewhere and we were not persuaded that there was a material difference in the use of the product which would account for the feedback from the pilot focusing on teaching aspects of the product rather than assessments.

Mr Hunter's Evidence

35. Mr Hunter evidence was:

(1) ConquerMaths was designed to be an assessment tool, helping students and educators identify areas of an individual's strengths and weaknesses in mathematics through structured, automated assessments;

(2) The product's core functionality lies in its ability to provide detailed, data-driven insights into a student's starting position and ongoing performance, allowing for targeted intervention and self-directed learning based on assessment outcomes;

(3) while certain features were prominently showcased on the landing pages to attract a broad audience, the assessment capabilities of ConquerMaths have always been at the heart of its design and functionality. The automated testing, progress tracking, and detailed reporting features are integral components that serve as the backbone of the product, differentiating it from traditional tuition services;

(4) each GM module was self-standing and age-appropriate, aimed at school age children. Each topic had resources available for it which included diagnostic assessments to set a baseline and identify strengths and weaknesses which the school or tutor could then address through tuition. A mastery report provided further information to the tutor;

(5) the GM product did include remedial teaching resources but these were short, typically 3-5 minute videos, and very much a part of the assessment process; and

(6) the GM offering included "worksheets" in an exam format which could be worked through in paper form by the pupil and then fed into the system and used to generate feedback. A worksheet is a type of assessment in this context.

36. We accepted these aspects of Mr Hunter's evidence, save to the extent that his use of the terms "assessments" (rather than say "diagnostic tests") to describe Conquermaths did not provide any evidence as to whether the product was within Note 4 or not since that depended among others things on the meaning of "assessments" in Note 4.

37. Mr Hunter also gave evidence that that:

(1) sometimes the language used by GM on their website was somewhat "softer" than might be expected of an assessment product because words such as "test" and "assessment" might be off putting for pupils so they were avoided;

(2) teachers looked at the various assessment results;

(3) a parent would buy ConquerMaths to identify gaps in their child's knowledge;

(4) a pupil would use the ConquerMaths product because without assessment the pupil and their tutor would not know what the pupil did not know. ConquerMaths therefore helped identify gaps in knowledge; and

(5) a pupil using ConquerMaths would have a teacher at school and typically take the results of the ConquerMaths assessments to the teacher in order to get additional teaching on topics identified as needing attention.

38. We found these elements of Mr Hunter's evidence unconvincing. The evidence seemed to be unsupported assertion. Mr. Hunter could not explain how many schools used ConquerMaths; how many teachers looked at the assessment results; or why a parent would buy it and encourage their child to use it, other than in expectation of the learning effect described on the ConquerMaths landing page. We were not convinced the presentation of ConquerMaths on the GM website was "softer" so as not to put off pupils; rather we thought the language reflected what the product really was.

Consumer Perception

39. Mr Constable's evidence was that the principal marketing of the GM product was through online sources which through search engine optimisation would lead potential customers to the ConquerMaths landing page. We were shown screenshots of various landing pages. We were told that some of these were landing pages which the pupils would encounter rather than those which their parents would encounter. We were also shown an example of the landing page which the parents would encounter. Mr Constable and Mr Hunter's evidence was that the landing pages for the pupils did not mention assessments because using words such as assessment or test would be off putting to pupil. Hence GM accepted that the landing pages for pupils might be perceived as downplaying the assessment elements of the ConquerMaths offering. The two screenshots we were shown that we understood to be the pages aimed at pupils indeed did not mention assessments. The wording of these pages included phrases such as "making maths fun and understandable", "welcome to ConquerMaths the premier online maths tutor", "our innovative online tutoring system allows children to learn at their own speed removing the pressure often associated with traditional classrooms and tuition", "the award-winning way to learn and revise maths at any age". All of this was suggestive that the product was a teaching tool.

40. We were also taken to a screenshot of a page which we understood was intended to attract parents to purchase the product for their children. This also seemed to us to predominantly present ConquerMaths as a learning or tutoring product. Mr Constable confirmed the screenshot was a summary of what the GM offering was, designed to attract parents and teachers to understand the full benefits of the ConquerMaths programme. The page included wording such as "the online maths tutor for students of all ages", "diagnoses gaps in maths knowledge - teaches the right lessons - tests understanding", "ConquerMaths helps to find those gaps in knowledge then teaches the right animated and narrated maths lessons to correct any problems", "an excellent maths teacher. ConquerMaths have teamed up with Pat Murray, a truly excellent maths teacher who presents each lesson in a friendly step-by-step manner. The lessons are concise and logical making them really easy to retain which gives students that important confidence in their ability", "an expert maths teacher available day or night to help with homework, revision or to recap missed or misunderstood lessons". Although the screenshot does include words and phrases such as "diagnoses" and "free diagnostic tests", we did not find there to be a material difference between the phraseology of this page (which GM asserted was designed for parents and teachers) and the pages designed for pupils. In both cases the description of the ConquerMaths product included phraseology which indicated it was, or substantially included, teaching and tutoring elements. The page we were shown does not use the word "assessments".

41. Therefore our conclusion was that the consumer perception of the ConquerMaths product was that it was a teaching tool designed to improve the understanding of maths of those using it. While we accepted that the product included a very large number of diagnostic tests, we were not persuaded that consumers, whether they were teachers, parents, or pupils, would perceive it as a product that primarily provided an assessment of maths abilities, even if it did consist mainly of diagnostic tests. We found that consumers (parents, pupils, or teachers) would perceive ConquerMaths as a teaching product designed to enhance the users' understanding of maths.

42. Although it did not form part of his evidence, Mr Khokhar accepted in his closing submissions that one way in which a pupil's maths proficiency might improve was simply by doing assessments. If that proposition is correct then, even if consumers did regard the GM offering as consisting primarily of assessments (which we did not think was likely to be the consumer's perception), the product would still be regarded by the consumer as being designed to improve the understanding of maths of those using it. It seemed to us that, if consumers had been told the product consisted of assessments and that by doing these tests the pupils understanding of maths would be enhanced, then the consumers would still regard it as essentially a teaching tool/revision aid albeit one that involved doing tests. When it was pointed out that this was unhelpful to GM's case, Mr Khokhar rather unconvincingly attempted to row back from his acceptance that doing tests might improve maths proficiency. We considered that his initial acceptance was in fact closer to the truth.

43. Mr Khokhar encouraged us to take the view that parents would do considerable due diligence on a product like ConquerMaths because of the importance of their child passing maths exams at various stages, the consequence of failure potentially being life changing. This was unsupported assertion. In any event we considered that if a parent looked "under the bonnet" to see what ConquerMaths consisted of, they would see that ConquerMaths included many diagnostic tests but in the context of a desire for their child to pass exams and the expectations set by the ConquerMaths web page, we considered that the well informed parent would (correctly in our view) believe that the diagnostic tests leading to the resulting analysis of the pupils strengths and weaknesses were a means of teaching not an end in themselves.

The role of assessments in learning

44. In relation to the role of assessments in education, we were taken to a paper entitled "the assessment for learning strategy" published by the Department for Children, Schools and Families and to some "assessment principles" published by the Department for Education. These papers did highlight the importance of effective assessment systems within schools as a means of giving reliable information to parents and helping to drive improvements for pupils and teachers. However, it did not seem to us that these bore particularly on the nature of the supplies being made by GM. We also noted that the papers themselves referred to assessment being a way of raising pupils' achievement and helping learners to know how to improve. This seemed to us consistent with the possibility that the process of assessment in education was to help to promote learning.

Conclusion on the nature of ConquerMaths

45. Our conclusion was that a typical consumer would perceive ConquerMaths as a teaching product (or revision aid) designed to improve the understanding of maths. If we had to go further, we concluded that objectively it was indeed a teaching product.

Meaning of "examination services"

46. We considered what meaning should be given to the expression "examination services" in Note 4. We understood the ordinary meaning of examination to be a formal test to assess knowledge or ability.

47. Our attention was taken by Mr Khokhar to VATEDU45100 in HMRC's internal manuals which includes the following words:

"Group 6 Item 3 Examination services including school inspections: what are examination services?

This item covers a wide range of activities and includes those carried out at the instigation of an educational institution itself. The term examination services includes the following supplies in the education and training sector:

(1) GCSE examinations etc;

(2) National Vocational Qualification (NVQ) assessments;

(3) accreditation;

(4) validation;

(5) certification;

(6) assessment and

(7) registration services.

Although item 3 of Group 6 of Schedule 9 of the VAT Act 1994 refers to examination services, it is clear from the legal definition contained in Note 4 to the Group that the exemption is intended to apply to a far broader area of services than examinations alone.

This deliberately reflects the trend away from examinations towards continual assessment and monitoring of workplace performance, personal portfolios, accreditation and so on as the means of recording attainments and achievements and also of awarding qualifications."

48. Our conclusion was that this was not particularly helpful to GM's case. It indicates that HMRC consider examination services is wider than examinations, but it does not state or imply that a product that uses diagnostic tests to teach maths is an examination service.

49. We considered that the words examination services in item 3 should be given their ordinary and natural meaning and that that meaning requires some element of a formal test to assess knowledge or ability. Our conclusion was that "examination services" was a term wider than formal public exams, but not so wide as to include something that was essentially a teaching product, or an integral part of a teaching product.

Is an assessment a self-standing concept in Schedule 9 VATA?

50. We heard arguments at some length from Mr Khokhar and Ms Donovan about the correct construction of Note 4 of Group 6. On the one hand Mr Khokhar argued that there were several separate limbs to this Note. He emphasised two elements. The first being the use of the word "include" which meant that the Note was not exhaustive but simply gave examples of items that were included within the meaning of examination services; items not specifically listed might still amount to examination services. The second being that the punctuation of the Note suggested that examination services encompassed separately (a) the setting and marking of exams, (b) the setting of educational or training standards, (c) the making of assessments, (d) other services provided with a view to ensuring educational and training standards are maintained. That construction seemed to be consistent with VAT notice 700/30 where at paragraph 7.1 the separate limbs of Note 4 of group 6 are set out as individual bullet points.

51. Ms Donovan sought to persuade us that Note 4 should be read as a whole and in context. She argued that the words "the making of assessments" in context sandwiched between "the setting of .... standards" and "ensuring ... standards are maintained" meant the words "the making of assessments" should not be treated as a separate item at all but rather was a phrase describing things like setting and judging the standards to which schools teach.

52. We did not find either Mr Khokar or Ms Donovan's submissions entirely compelling. We accept that (1) the use of the word "include" means that the list in Note 4 is not exhaustive and could include for example a "test" that led to a formal qualification even if it was described as a test not an exam or assessment; and (2) the punctuation of Note 4 means that the word "assessments" can mean more than assessments provided with a view to ensuring educational and training standards are maintained. However, in our judgment Ms Donovan's argument that Note 4 had to be read as a whole was correct. We considered that context had to be taken from the words "examination services" and that a service which was in essence a form of teaching/education could not be an examination service by reason of this Note merely because it involved, or indeed consisted of, a considerable number of tests and regardless of whether those tests were correctly called "assessments" or something else.

Meaning of "Assessments"

53. Mr Khokhar argued that HMRC had not offered a clear definition for "assessments" and that the word should be given a special educational meaning in Note 4. Mr Khokhar referred us to:

(1) a publication titled "The Assessment for Learning Strategy" from the Department for Education ("DfE") which highlights the importance to pupils' learning when they and their teachers have a really good understanding of where pupils are in their learning;

(2) another DfE publication on the assessment principles that underpin effective assessment;

(3) an academic publication titled "Assessment Summative and Formative - some theoretical reflections" written by Maddalena Taras which sought to show that all assessment begins with summative assessment (which is a judgment) and that formative assessment is summative assessment plus feedback which is used by the learner; and

(4) an analysis of HMRC's arguments from Mr Aydin Onac (a former Head of Education at GM) which sought to explain how the ConquerMaths product gives reliable information to parents about how their child, ... is performing and help drives improvement for pupils and teachers.

54. We noted that Mr Onac had not given a witness statement and was not available for cross examination.

55. While the publications and documents which Mr Khokhar referred us too were obviously important in his mind as to the argument he was advancing, we did not find them particularly helpful. We accepted GM's argument that assessment may be an important part of a child's learning, especially in mathematics. We also accepted that ConquerMaths product includes many diagnostic tests that could be undertaken by the pupil. But we could see no authority for giving "assessment" a special meaning in Note 4.

56. Accordingly we considered the right approach was to adopt the ordinary and natural meaning of the word assessment, i.e. the process of testing, and making a judgment about, someone's knowledge, ability, or skills.

57. However, we considered that in the context of Note 4 which commences with the words "examination services", that all the words of the Note take colour from the introductory words "examination services" and, in context, the word "assessments" although wider than examinations that lead to qualifications, is not so wide as to encompass "assessments" that are simply part of a learning/teaching/revision process.

58. In our judgment, the key question to consider was whether the ConquerMaths product was a form of teaching/revision product or not. We considered that it would be perceived as such and that it was indeed such a product. Accordingly, in our judgment it cannot be shoehorned into item 3 as an examination service merely because it involves (using our neutral words) diagnostic tests.

59. We considered whether anything turned on the description of ConquerMaths by the CM witnesses as providing "assessments". We noted that the webpage we had been taken to (described by Mr Constable as an accurate summary of the product) used the words "diagnostic tests" not "assessments", but the GM witnesses had consistently described the product as providing "assessments". Officer Stephenson also accepted that the product included "assessments", albeit she regarded that as an ancillary part of the supply. In our judgment nothing very much turned on this point, although we have been careful in this judgment to use a neutral term to describe the product to avoid any confusion that might arise by using the word assessments. If ConquerMaths can correctly be described as providing "assessments" then we did not think that would of itself bring it into the category of "examination services". On the other hand, it was not fatal to whether the product was within Note 4 that the word "assessments" seemed not to be used in its description to consumers. As set out in [61] and the following paragraphs, in our judgment the key issue was how the product was perceived by consumers and/or the actual nature of the product. It made no difference what word was used to describe it. Accordingly, we did not find it necessary to form a definitive conclusion on the meaning of "assessments" in Note 4.

Test to be applied

60. Having determined what GMs supplies consisted of, we considered what test we should apply to determine the nature of these supplies for VAT purposes.

Consumer perception test

61. Based on MIS, we considered that the test we should use for determining the character of GM's supplies was an objective test, determining how their supplies would be characterised by the typical consumer.

62. Based on MIS we considered all the circumstances in which the supply takes place and concluded that:

(1) The supply by GM was a single supply, the predomination element of which is a supply of educational services;

(2) the perception of a typical consumer would be that the supply was a form of teaching rather than a form of examination; and

(3) accordingly, the supply was not an examination service within the meaning of Item 3.

63. In our judgment what GM is supplying is a form of education and not a form of examination services. We reached that conclusion on the basis that:

(1) A consumer of the ConquerMaths product would regard it as a teaching product primarily designed to improve the understanding of maths rather than something designed to assess maths skills as a self-standing end on its own. In our judgment, that would be the case regardless of whether the consumer is the pupil using the product (whose perception would be shaped by the landing page for pupils which clearly presents the products as a teaching product) or the parent buying the product (whose perception would be shaped by the landing page for parents, which again emphasises the product as involving teaching).

(2) We were not presented with any satisfactory evidence of why a consumer would buy the product purely as a means of accessing assessments. It was argued that a parent would want to get access to the library of assessments so as to determine what standard their child had reached and seek appropriate remedial maths tutoring if necessary, because of the importance of pupils understanding maths and the way in which it is necessary to master topics cumulatively so as to be able to make progress. But we were presented with no evidence to support that assertion and it seemed to us much more likely that a parent reading the landing page would buy the product in expectation it was a teaching and/or revision resource that would improve their child's understanding of maths. In this context, we observe that we could have been given evidence that consumers were buying the product in order to get access to assessments to determine the level of the child's understanding in maths and what teaching was required. For example we could have been presented with evidence of advertising that that was the benefit of ConquerMaths, or a survey of consumers seeking to show why they had bought ConquerMaths. The only evidence we had of what benefits parents were expecting from ConquerMaths was what we could deduce from the landing page they saw; the benefits described on that landing page were predominantly teaching and learning rather than assessment.

(3) The admission, albeit that he subsequently tried to retract it, by Mr Khokhar that one way of making progress in maths was to do the tests was telling in our view. In our judgment and based on our own experience, doing tests is itself a way of learning about a topic since, once an incorrect answer is identified, it is often possible to reverse engineer why the answer is incorrect so that the point can be understood and learned. This pointed to ConquerMaths being a learning/tutoring/revision product rather than an examination service.

64. As explained in [22], based on MIS, we considered that the appropriate test to apply to the characterisation of GM's supplies was to determine the nature of the supplies viewed objectively from the point of view of the typical consumer. Having established (see [45]) that the typical consumer would regard GM supplies as a supply of educational services to which the diagnostic tests were an ancillary part, our conclusion was that the supplies did not fall within the category of examination services. We considered that the typical consumer would regard the supply as a single supply of educational services. Strictly we did not think the elements that consisted of diagnostic tests were ancillary to that single supply; rather the diagnostic tests were an integral part of a single product designed to increase the understanding of maths.

Alternative test - functional analysis

65. Mr Khokar submitted that in the context of GM's supplies the consumer perception test was not the appropriate test and instead a different test should be used to characterise the supplies. He submitted that instead of the consumer perception test, the characterisation of these supplies should be made using a functional analysis, citing as authority Procter & Gamble UK v Revenue and Customs Commissioners [2008] EWHC 1558 (Ch) and Customs and Excise Commissioners v Ferrero UK Ltd [1997] STC 881.

66. We were not persuaded that these cases were authority for that proposition. But in case we were wrong on that point we considered what would the outcome be if a functional test were to be applied. We considered that the function of the ConquerMaths product was to increase the understanding of the user's knowledge of maths through carrying out [the diagnostic tests and reviewing the results. Accordingly, even if we were wrong in applying the consumer perception test and, as submitted by Mr Khokar, the correct test to apply was some sort of functional test, or test based on the perception of a maths tutor, we would still have found that the supply was not an examination service.

(1) From the evidence we were presented, the function of ConquerMaths is to increase the understanding of maths and therefore we concluded its function was a form of teaching rather than a form of examination;

(2) The evidence of the Barbados study showed that teachers regarded the product as a form of teaching and, even if the intention was to use the product in a different way in Barbados, since the product was identical, logically, it would be just as much a teaching product when used outside Barbados as it was a teaching product used there.

67. Accordingly, on any of the tests we were encouraged to use, the product was not in our judgment an examination service.

Best Judgment

68. On behalf of GM, Mr Khokhar sought to argue that Officer Stephenson had not exercised best judgment as required by s73 VATA when issuing the Assessment. Mr Khokhar's argument was that Officer Stephenson had not gone through an objective fact-finding exercise, had not carried out a reasonable analysis, considered all relevant material or had a transparent basis for the Assessment.

69. In considering this issue, we applied the test set out in the decision of the High Court in Van Boeckel. We considered whether Officer Stephenson had fairly considered all material placed before her and, on that material, come to a decision which is reasonable and not arbitrary as to the amount of tax due. Officer Stephenson's evidence on this issue was clear and compelling. She had taken into account the information she had been given by GM (including a letter sent to her by GM dated 15 May 2018) and the other sources available to her, and reached the conclusion that the supplies were of education. Officer Stephenson formed her view on the basis that she regarded GM's supply as consisting of various elements which included assessments. She considered that the assessments were an ancillary part of a single supply and that the overall supply was not one of examination services but rather was a supply of mathematics tuition. That was a reasonable conclusion. She had then applied an appropriate method to calculate the VAT due.

70. On this point, we regarded Mr Khokhar's argument to be essentially an alternative way of putting his principal argument, that GM was supplying examination services rather than educational services. It was clear from Officer Stephenson's evidence that she had considered all the evidence before her, provided both by GM and other sources (see [9(3)]) she had identified herself. These were relevant sources for Officer Stephenson to have considered. The conclusion she had come to was not arbitrary. Mr Khokhar's objection to her conclusion was at its heart not founded upon her having taken into account irrelevant matters or not taken into account relevant matters but rather that he did not agree with the conclusion that she had come to because he considered the supply by GM to be a supply of examination services. That is an issue which is examined fully elsewhere in this judgment.

71. Accordingly, we considered that the Assessment did meet the best judgment requirements of s73 VATA and had been notified to GM within the time limits in s77 VATA.

other matters

72. Mr Khokhar repeatedly sought to argue that GM had a legitimate expectation that its supplies would be treated as exempt. This argument was put forward with various formulations, but the common thread was that an enquiry had been made via HMRC's online portal on behalf of GM in 2014 or 2015 asking for confirmation that "supplies of examination/assessment services are exempt from VAT under item 3(a) of Group 6 of Schedule 9 of the VAT Act 1994". We took the view that we did not have jurisdiction to hear this argument on the basis of the authority of the Upper Tribunal in The Commissioners for Her Majesty's Revenue and Customs v Abdul Noor [2013] UKUT 71 (TCC) where it was held that the First-tier Tribunal does not have jurisdiction to give effect to legitimate expectation claims that go beyond an entitlement ascertained in accordance with the relevant VAT legislation; any such legitimate expectation is a matter for remedy by judicial review. In any event, the enquiry made on behalf of GM and the reply from HMRC did not appear to us to even begin to meet the requirements for actionable legitimate expectation set out in R v Inland Revenue Commissioners, ex parte MFK Underwriting Agencies Ltd [1990] 1 WLR 1545. The enquiry asked for confirmation that examination services were exempt (which they self-evidently are), rather than explaining the nature of GM's supplies and asking whether those were correctly to be characterised as examination services. Using the analogy of MFK, GM's cards were not even put on the table, let alone face up.

Decision

73. Accordingly we concluded that we should dismiss the appeal.

Right to apply for permission to appeal

74. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to "Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)" which accompanies and forms part of this decision notice.

Release date: 24^th APRIL 2025

First-tier Tribunal (Tax)