Story Terrace Limited v The Commissioners for HMRC

The Appellant appeals against the part of HMRC’s review decision of 21 August 2023 (the ‘Review Decision’), which upheld HMRC’s VAT liability ruling of 6 July 2023 for the Appellant’s 3/23 VAT return (the ‘Liability Decision’).

The Appellant says that it is making zero-rated supplies of books for the purposes of Value Added Tax Act 1994 (‘VATA’). HMRC say that the Appellant is making standard-rated supplies of ghost-writing services for the purposes of VATA.

The issue of quantum was not before us.

For the reasons given below the appeal is upheld.

For the hearing we were provided with a hearing bundle of 1054 pages, a supplementary hearing bundle of 435 pages, an authorities bundle of 856 pages and various samples of the Appellant’s products. We were asked to watch a number of online videos which we did. During the hearing a further authority, another HMRC VAT manual and some corrections to one of the transcripts in the hearing bundle were handed up. Mr Rutger Bruining a director of the Appellant and Officer Angela Seymour of HMRC both provided witness statements and were cross examined.

We are grateful to Mr Brodsky and Mr Hickey for their skeleton arguments, submissions, and willingness to engage with our questions and to the witnesses for their evidence. We set out below our summary of those submissions on the law and the facts. The parties should, however, be assured that when preparing this decision, the terms of the skeletons were reread and our notes of the hearing reviewed. Because we do not deal specifically with any point it does not mean that it was not considered in the round when reaching our decision.

The issue in this appeal is whether, as the Appellant says, its supplies are of books and therefore are zero-rated for the purposes of VATA. Or whether the Appellant is making standard-rated supplies of ghost-writing services for the purposes of VATA.

It was common ground that the Liability Decision affected other periods and that the issue of quantum was not before us.

It is for the Appellant to show that its supplies are zero rated.

The standard of proof is on the balance of probabilities.

If a supply of goods or services is zero-rated no VAT is charged on that supply. However, it is in all other respects treated as a taxable supply, see s30(1) VATA.

A supply of goods or services is zero-rated if it meets the description in Schedule 8 VATA, see s30(2) VATA. Item 1 Group 3 Schedule 8 VATA (‘Item 1’) provides:

s6 European Union (Withdrawal) Act 2018 (‘EUWA’) provided in relevant part at the relevant time:

The Tribunal heard oral evidence from Mr Bruining and Officer Seymour. Mr Bruining’s three written witness statements stood as his evidence-in-chief and were for the most part unchallenged. Mr Bruining was cross-examined, however the focus of the cross examination was not the accuracy of what he said or the veracity of various documents he exhibited to his witness statement. Rather the cross-examination focussed on whether the various documents undermined the Appellant’s case, that is a matter for us to determine primarily by reference to those documents and not the views of Mr Bruining. Officer Seymour’s witness statement stood as her evidence in chief and was largely unchallenged. Officer Seymour was also cross examined. Both Mr Bruining and Officer Seymour answered questions in a straightforward manner and stated clearly what they did not know. We found both witnesses to be honest and credible.

We set out the majority of our findings of fact in this part of our decision. A lot of our findings of fact are from the documents supplied or the largely unchallenged evidence of Mr Bruining and Officer Seymour. Consequently, most of them require no discussion as they were not in dispute. Where they were in dispute we provide the reasons for our findings below. Therefore, we have also incorporated Mr Bruining’s and Officer Seymour’s evidence in the findings of fact below rather than setting out the evidence provided.

The Appellant did not adduce any written or oral evidence from a ghost-writer. Consequently, HMRC invited us to draw an adverse inference as to the low-level role of the ghost-writers contracted by the Appellant in line with the First-tier Tribunal’s recent approach at [259] – [260] in Minstrell Recruitment Ltd & Ors v Revenue & Customs [2024] UKFTT 82 (TC) applying Royal Mail Group Ltd v Efobi [2021] 1 WLR 3863 at [41]. However, we decline to draw such an inference for the reasons put forward by Mr Brodsky: the relevant test in this appeal involves the objective view of a typical consumer and not the subjective view of a particular ghost-writer; also there is no factual dispute about what the ghost-writers provide; and we do not see that such evidence would be relevant to the determination of this appeal. Therefore, we do not consider that the context or particular circumstances of this appeal means that positive significance should be attached to the lack of evidence from any of the ghost-writers the Appellant contracts with.

The Appellant is a limited company and was incorporated in 2014. It is now in administration. The Appellant is VAT registered.

The business of the Appellant is very important personally to Mr Bruining who is a director of the Appellant. Mr Bruining was inspired to set up the Appellant’s business after he found that details of his grandfather’s stories faded with time: Mr Bruining clearly had a remarkable grandfather. It was always a key element of Mr Bruining’s vision that the Appellant would produce books so that there was a physicality to the end-product: something that could be picked up and read at any time and that would become part of family history in a way that text existing on a computer would not.

The Appellant’s press factsheet describes its company mission:

On the Appellant’s website:

The Appellant’s customers engage the Appellant via its website. The website sets out 8 steps for creating one of the Appellant’s products (Footnote: ¹), as follows:

The Appellant’s website also sets out three different levels of package that a customer can buy in ascending price order:

At the bottom of each pricing option the customer is invited to “select book”. The differences in the three packages are the numbers of pages (in total and the number of pages of text), the number of words, the number of photos, the number of hours of interview and the level of writer (junior, senior or premium) that comes as standard as set out below.

All packages have dedicated editorial sessions (not with the ghost-writers which include proof-reading and input into the shape of a customer’s narrative), a project length of 7 months and the customer ultimately receives 4 colour hardback books and a digital copy of the book.

The hearing bundle contained a call script for the Appellant’s customers. The project editor calls the customer and explains the processes entailed in the production of the book: describing the project editor’s role including: management of the ghost-writer, the proofreader, the Bookmaker design team and the printer; how writer matching works; confirmation/clarification of project goals; the basic seven month project plan; and next steps. We accept the Appellant’s point that there is more than simply the ghost-writing that goes into the production of the books.

Mr Bruining also gave uncontested evidence that in various communications with its customers the Appellant focussed on the offerings of books: new leads receive an email saying “Very exciting that you are looking at creating this book”; after purchase the welcome email explains that the project editor is the point of contact for the “book-making process”; an email that explains that a project editor has been assigned explains the elements involved in “delivering the book”.

The Appellant’s website does not provide an option for the text alone or the services of a ghost-writer only. Mr Bruining’s view was that the Appellant did not compete with stand-alone ghost-writing services because in comparison, following some basic research, a freelance ghost-writer’s fees for providing the text that would appear in one of the Appellant’s books would be 3 – 6 times less than the Appellant would charge for one of its books.

The Appellant’s website offers a number of ‘additional services’ distinct from that set out above. For the purposes of this appeal it is sufficient to find that the ‘Editing Services’ and ‘StoryTerrace Pro’ involve ghost-writing.

The Appellant’s customers spend an average of 4.5 times longer on the book than the ghost-writer.

The Appellant spent over £900,000 building its software, ‘Bookmaker’. Bookmaker is important to the Appellant as it allows it to create books efficiently at scale. The Appellant’s customers use Bookmaker to:

The editors are employed by the Appellant, but the ghost-writers are not.

HMRC also invited us to view, which we did, publicly available videos entitled ‘StoryTerrace Swift Webinar’ and ‘Talk: Capture your life story in a book’. HMRC said they supported its position because they emphasised: the Appellant’s access to a large bank of quality and experienced ghost-writers who enjoy what they do; the skill of the ghost-writers in capturing the emotion of the Appellant’s customers; and the biography aspect of the Appellant’s supplies. Further, the second video included in its description ‘#ghostwriter’.

The role of the ghost-writers is clearly an important aspect of the Appellant’s business: in his oral evidence Mr Bruining said that the ghost-writers played an important role in the creation of the books and that was a fact. Mr Bruining did not seek to suggest otherwise.

The Appellant’s contract with its customers is set out in an annex to this decision (the ‘Contract’). No submissions were made that it also applied to the ‘additional services’. We discuss the clauses relied upon by the parties below.

In response to a request from HMRC Mr Bruining provided his second witness statement (dated 24 April 2025). That statement sets out how much the ghost-writers were paid by the Appellant compared to the price paid by the customer for 4 projects chosen by HMRC. The relevant figures are:

Mr Bruining and Mr Theo Branning (the Appellant’s Marketing Director) appeared on the well-known TV programme Dragon’s Den. During that appearance Mr Branning said:

HMRC’s calculations prompted Mr Bruining’s third witness statement (19 May 2025). By reference to the Appellant’s profit and loss figures for 2023 Mr Bruining’s evidence was that the ghost-writers’ fees were 29% of the Appellant’s direct costs (excluding sales and marketing, and non-operating costs). In oral evidence Mr Bruining explained that the profit and loss figures were not retrieved by him, were unaudited and were not exact because of the Appellant being in an insolvency process. He did explain however that he had checked the figures line by line. Mr Hickey did cross-examine Mr Bruining on some of the figures but that did not in our view result in a successful challenge to the headline figures provided by Mr Bruining nor did Mr Hickey submit that it did in closing. We do not consider that Mr Bruining’s evidence regarding percentage of direct costs is necessarily undermined by his earlier evidence (regarding percentage of the cost charged to the Appellant’s customers) or HMRC’s calculations. That is because Mr Bruining’s earlier evidence related to a much smaller sample size. In relation to HMRC’s calculations, again they are based on a small sample size where each sale was in excess of £5,530 and utilises broad comments made in a television programme about average sales of £3,300. Therefore, we accept that during 2023 the ghost-writers’ fees were 29% of the Appellant’s direct costs (excluding sales and marketing, and non-operating costs).

The hearing bundle contained 227 pages of online reviews of the Appellant’s service. Mr Brodsky said that they showed the importance to the Appellant’s customers of the book and Mr Hickey said that they showed the importance of the ghost-writers/ghost-writing services to the Appellant’s customers. If such reviews were relevant, to the view of the typical consumer, we do not consider that we can draw a meaningful conclusion from them that supports either party’s case. That is because on the basis of how they were presented to us it was clear that either party could select the reviews that they said supported their case. Neither party provided any in-depth analysis of the reviews such that we could reach a sensible conclusion on whose case they supported. Therefore, we do not find any facts in relation to them and they do not form part of our decision.

We explained above that Mr Bruining and Mr Branning pitched the Appellant’s business on Dragons’ Den. Mr Brodsky sought to argue that the Dragons’ reactions supported the Appellant’s case. Like the point we make about the online reviews, if it was right to consider these reactions in principle when determining the view of the typical consumer, we do not consider that these are helpful in this appeal. That is because they are made at a point when all the Dragons have seen is the book and not once they had heard the detail of the Appellant’s supply.

The Appellant’s case is that its supplies are within Item 1 and fall to be zero-rated i.e. it supplies books for the reasons that follow.

The Appellant’s end product is a book for the purposes of Item 1.

Applying the relevant test (see below), the predominant element of the Appellant’s supply is of a book. That is because:

The Appellant says that the principal/ancillary test (see below) does not apply to the facts of this appeal.

HMRC say the Appellant makes a single supply of ghost-writing services which falls to be standard-rated because:

HMRC’s Statement of Case put the Appellant to proof as to whether its final outputs were books. We were provided with samples of the Appellant’s end product. They all had a significant number of leaves held together with front and back covers more substantial than the leaves and therefore met the minimum characteristic of a book as required by Customs and Excise Commissioners v Colour Offset Ltd [1995] STC 85. There was no suggestion from HMRC that the samples we had were not representative of the Appellant’s supplies and therefore we find that the Appellant’s final outputs are books.

It is well established that where a transaction comprises a bundle of elements and acts it falls to be determined whether there are two or more distinct supplies or whether there is a single supply for VAT purposes. In this appeal it was common ground that the supplies by the Appellant to its customers are single supplies for VAT purposes. We agree with the Appellant that this position was consistent with the principles set out by the Upper Tribunal in Middle Temple v HMRC [2013] STC 1998, at [60].

For the purposes of this appeal the there are two relevant circumstances where one or more supplies may be considered a single supply. In this case the parties proceeded on the basis that the relevant type of single supply was a Levob Verzekeringen BV v Staatssecretaris van Financiën (Case C-41/04) [2006] STC 766(‘Levob’) type of single supply. We do not see any reason to disturb that approach. Consequently, we do not consider it necessary to address HMRC’s criticisms of the part of the Appellant’s skeleton argument that analysed why the Appellant’s supply was a Levob type of supply.

Levob concerned the supply of customised software. The cost of the basic software was distinct in the contract from the cost of the customisation. The issue before the ECJ (as it then was) was whether there had been separate supplies of the basic software and the customisation. At [30] the ECJ stated:

The application of Levob and subsequent CJEU case law was recently considered in Gray & Farrar International LLP v HMRC [2023] EWCA Civ 121 (‘Gray & Farrar’). Gray & Farrar concerned, in summary, the VAT treatment of the supplies made by ‘matchmakers’. Following its conclusion at [47] that the predominant test was the primary test for characterising a single supply, the Court of Appeal at [49] – [51], clarified the hierarchy of tests:

Discussing the application of the predominance test the Court of Appeal, in Gray & Farrar at [41] – [42], also usefully sets out the pertinent extract from Město Zamberk v Financni reditelstvi v Hradci Kralove (Case C-18/12) (‘Město’) and its conclusions on how to apply the predominance test:

Having applied the predominance test to the facts of the case and commented on the First-tier Tribunal’s and the Upper Tribunal’s approach at [60] and [61], at [64] the Court of Appeal warns against artificially splitting single supplies when applying the predominance test:

To emphasise the Court of Appeal’s warning Mr Brodsky took us to the both the Attorney General’s opinion at [25] – [28] and the CJEU’s decision at [39] – [40] in Case C-581/19 Frenetikexito – Unipessoal Lda v Autoridade Tributária e Aduaneira (‘Frenetikexito’). Determining whether we wish to accept Mr Brodsky’s submission would require some detailed discussion of Frenetikexito. We decline to undertake that discussion in this appeal for a number of reasons. First, Frenetikexito was decided in 2021 which was after IP completion day. This means that the CJEU’s decision is not binding on us although we may take it into account, see s6 EUWA. We also note that the CJEU’s endorsement of the Attorney General’s analysis clearly relates to the comments made about the classification of supplies as single or multiple and not the characterisation of the supply. That is not surprising as that was the issue before the Court but that is not the issue before us. Finally, we have the above binding guidance from the Court of Appeal which follows from a discussion of the relevant characterisation test in Město (which in any event the Advocate General’s opinion did not disagree with).

We do not understand that the overall assessment required by Město precludes an awareness/identification of all aspects of a supply. However, such an assessment should not entail identifying distinct components of a supply (which may bundle together aspects of a supply) and then deciding which predominates.

HMRC say that the majority of the elements of the transaction at issue (such as Steps 3 – 6 above) are all closely linked to form a single supply of a ghost-writing service. Further, that from the perspective of a typical consumer they receive ghost-writing services and a book. HMRC conclude that the supply of the ghost-writing service is the predominant element. In our view this approach starts with the dissection of the Appellant’s supply in the way that the Court of Appeal warned against (although that does not preclude us from considering the factors that HMRC put forward for the overall assessment). Instead, our approach is to answer the question what is the predominant element in what the typical consumer thinks he or she is acquiring? By making an overall assessment of all elements of the supply to determine their importance to the typical consumer, both qualitatively and quantitatively, we can then decide which predominates.

Gray & Farrar at [56] – [57] rehearses the authority for the proposition that normally the starting point for the characterisation of a transaction/supply for VAT purposes which is governed by a written agreement is the written agreement itself. Therefore, we start by considering the Contract.

Turning to the Contract as set out in the Annex to our decision (all references to clauses that follow are to those in the Contract), we find as fact that these were the terms and conditions at the material time. We also find that Steps 1 – 8 above and the three different levels of package described above formed part of the Contract between the Appellant and its customers by virtue of Clause 12 (the definition of goods and services). The parties drew our attention to which clauses in the Contract they relied on.

In relation to the Contract the Appellant says its purpose is the provision of a book:

In relation to the Contract HMRC say:

HMRC are right to say that aspects of the Appellant’s supply would be standard-rated if they were supplied on their own but that is not a barrier to a conclusion that the Appellant’s supply is zero-rated.

It is clear and not in dispute that the books produced under the terms of the Contract are bespoke, see Clause 20. The parties made a number of arguments in light of this fact.

Mr Brodsky took us to the First-tier Tribunal’s decision in Harrier v HMRC [2011] UKFTT 725 (TC) (‘Harrier’). Harrier considered whether the supply of personalised photobooks should be zero-rated pursuant to Item 1. Mr Brodsky’s point was that simply because a product is customised or bespoke does not mean that the supply of the customisation service is predominant. Mr Hickey took us to the facts in Harrier at [8] – [15] to make clear that the facts of Harrier are clearly distinguishable from the facts of this appeal. In particular, there was no assistance from Harrier in laying out the photobooks nor the provision of anything akin to the ghost-writing service provided by the Appellant. Having identified at [32] that Levob was the relevant authority as to how to classify the supply in the case the First-tier Tribunal concluded at [34]:

This is a useful juncture to note that we were referred to a number of other cases that considered the determination of whether there was a single or multiple supply (which is not the issue before us) and/or the characterisation of single supplies (including Paragon Customer Communications v HMRC [2018] UKFTT 162, Faaborg-Gelting Linien v Finanzamt Flensburg (C-231/94) (‘Faaborg’), David Baxendale v HMRC [2009] EWCA Civ 831, HMRC v Weight Watchers (UK) Limited [2008] EWCA Civ 715, HMRC v Metropolitan International Schools Ltd [2017] UKUT 431 and College of Estate Management v HM Customs and Excise [2005] UKHL 62). We were asked to draw analogies between these decided case and the facts of this appeal. We take from Faaborg the point that the existence of an end product in a supply that involves a number of elements does not necessarily render the characterisation of that supply as a supply of that product. However, generally we make the same point as above: in terms of applications of the relevant principles to the facts, these cases very much turn on their facts and the present case is clearly distinguishable.

A point of dispute between the parties arose in relation to HMRC’s notice Zero-rating books and printed matter (VAT Notice 701/10) which provides at 7.2 that original or specialist items are standard-rated. The dispute flowed from what was meant in that notice by original or specialist items. However, it is uncontroversial to say that such a notice is not the law (which we have set out above) it is HMRC’s view of the law. Accordingly, and with respect to the submissions made by both parties, we do not derive any assistance from the notice in reaching our decision and so do not address the parties’ detailed arguments on it.

Following on from the copyright position, HMRC invited us to consider what the Appellant’s customers would receive if they had not received the ghost-writing service. Again, with respect that is not the question prescribed by the law set out above and so we do not find it necessary to consider it.

Both parties took us to Clause 12 (the definition of goods and services) which encompasses all aspects of the Appellant’s supply and is buttressed by Clause 8 (the definition of the contract). In our view the inclusion of ‘the number’ in Clause 12 emphasises the importance of the books: the words could be removed from Clause 12 and the Contract would still have the same effect given the definition of order in Clause 13. This emphasis is repeated by the fact that book has its own definition in Clause 5 which also describes the book as the ‘final output’, there is no mirror provision in relation to any other part of the Appellant’s supplies. We also agree with the Appellant’s submissions above that, in the order they were made by the Appellant, Clause 26 (sending to print within a certain time), Clauses 71 – 73 (the books will meet a certain standard), Clause 28 (the importance of proof-reading) and Clause 31 (replacement of ghost-writer) demonstrate that the Contract’s purpose is to ensure that a book is produced. This conclusion is also supported by the clear production schedule/production plan which ensures the timely production of a book, see Clause 15 (this view disposes of HMRC’s point about the length of the production process). Further, there is no express provision in the Contract that permits the Appellant’s customers to receive only part of the Appellant’s supply which generates the text for the book. We accept that there is a variance in contractual price in respect of the ghost-writers’ time and/or experience (depending on which package is chosen) but the variance in price also encompasses the number of pages and photographs. Therefore, this variance is not sufficient to lead us to a different conclusion about the purpose of the Contract.

It is true Clause 18 references a process and that the Appellant’s books are the outcome of that process but we do not consider that that imbues Clause 18 with the meaning that the Appellant asserts. In our view Clauses 15 and 21 do not advance the Appellant’s case because such an approach essentially dissects the Appellant’s supply in the impermissible way discussed above.

The test in Město and the Court of Appeal’s comments on it in Gray & Farrar clearly reference the quantitative importance of elements of a supply. However, it does not appear totally straightforward to us how that is to be reconciled with the warnings about dissection of supplies discussed above. In any event, in terms of any quantitative analysis in this appeal we have rejected HMRC’s arguments that the majority of the Appellant’s direct costs are associated with the ghost-writers and so we do not need to consider it further. HMRC’s skeleton briefly observed that the printing costs were lower than the costs of the ghost-writers but that point was not developed further.

As set out above we were asked to consider a number of things that do not form (although some describe) part of the Contract including: the publicly available videos entitled ‘StoryTerrace Swift Webinar’ and ‘Talk: Capture your life story in a book’; the three quotes from and numerous references to ghost-writers on the Appellant’s website; the call script; the additional services (insofar as they included ghost-writing); and Mr Bruining’s uncontested evidence about various communications with its customers. We are satisfied that nothing in these vitiate our analysis of the Contract.

We do not consider that this is a case where the bespoke nature/aspect of the Appellant’s supplies dominates such that we can conclude that the Appellant is making standard-rated supplies. That is because our view is the Contract’s purpose reflects the typical consumer’s and that is ensuring that a book is produced to be shared and enjoyed in its physical form: in other words that is the economic and commercial reality of the transaction, and its economic purpose.

In light of the above we consider that the typical consumer would regard the provision of the book as qualitatively the most important element of the Appellant’s supply. Therefore, in light of the Contract the provision of the book is the predominant element of the Appellant’s supply.

HMRC made the alternative and mutually exclusive argument that the Appellant’s provision of books to its customers was ancillary to the provision of ghost-writing services pursuant to Card Protection Plan Ltd v Customs and Excise Commrs (Case C- 349/96). In light of our conclusions above we do not need to deal with that argument and so do not do so.

An important part of the Appellant’s case was that HMRC’s treatment of it breached fiscal neutrality because, in broad terms, one of its competitors’ supplies were not being treated in the same way as the Appellant’s. It is unfortunate that, and it was not clear to us why, it was not until Ms Seymour’s examination in chief that HMRC revealed for the first time that it was treating the Appellant’s competitor’s supplies in the same way as the Appellant’s. Consequently, the Appellant withdrew its fiscal neutrality ground and we do not deal with it.

For the reasons set out above we uphold the appeal.

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.