Here's a conversation that happens in a lot of publishing businesses. Sales say it's been a cracking month: renewals are in, a big corporate deal has landed, cash is up. Finance say revenue is flat. Both are right.
The gap between them is deferred revenue. If you run subscriptions, it's worth understanding, even if you never go near a ledger.
I'm not an accountant, and this isn't accounting advice. Your accountants and auditors will have the final word on your policies. But the basic ideas are straightforward, and most of the pain comes from the data, not the theory.
Cash isn't revenue (yet)
When a subscriber pays £120 upfront for twelve issues, you have £120 in the bank. But you haven't earned it. You owe that subscriber twelve magazines. Until you deliver them, the money is effectively held on their behalf.
So in the accounts, the £120 goes in as deferred revenue (sometimes called deferred income), which sits on the balance sheet as a liability. As each issue goes out, a slice of it moves from the balance sheet to the profit and loss account as revenue.
That's the principle behind the UK and international accounting standards (FRS 102 and IFRS 15 both recognise revenue as you deliver what was promised). How your business applies them is a question for your accountants.
Releasing revenue per issue
For print subscriptions, the natural approach is to release revenue as each issue is delivered. In the simple case, £120 for twelve issues means £10 is released with each issue.
Real life is messier:
- Irregular schedules. If you publish ten issues a year with a double issue in summer, does the double issue release twice as much? Your policy needs an answer.
- Starts mid-cycle. A subscription that starts with issue seven releases its first slice at issue seven, not in January.
- Extensions and goodwill issues. If you add two free issues to say sorry, the same money now covers fourteen issues, and the value per issue changes.
- Suspensions. If a subscription is paused, nothing should be released while it's paused.
- Cancellations and refunds. Whatever hasn't been released yet is what's potentially refundable.
Releasing revenue over time
Digital subscriptions don't always have discrete issues. If someone pays for twelve months of website access, it often makes more sense to release the revenue evenly over the term, a twelfth each month.
Some publishers smooth print revenue the same way for simplicity. Again, that's a policy decision. The important thing is to choose one, apply it consistently and make sure the system follows it.
Bundles: splitting the money
A print and digital bundle, or a subscription covering three titles, raises a fair question: which product earned the money?
The usual answer is to split the price between the products in the bundle by a set proportion, then release each part according to that product's own schedule. Get this right and each title's profit and loss account reflects what it actually earned. Get it wrong and one title looks brilliant while another looks like it's losing money.
It also matters for VAT. Print and digital publications are both zero-rated in the UK at the moment, but bundles that include other things (events, reports, access to data) may not be, and VAT rules differ for overseas customers. That's another conversation to have with your accountants, and another reason to set the split per price rather than working it out by hand.
Corporate and multi-currency subscriptions
Corporate licences are often invoiced annually, sometimes in advance, sometimes in arrears, and often in a currency other than sterling. The same principle applies: release the revenue over the period the licence covers. The complication is keeping track of the exchange rate used, so your month-end figures can be explained later.
Why it goes wrong
In my experience, the theory isn't the problem. The data is. Deferred revenue goes wrong when:
- the finance team calculates it on a spreadsheet from a report circulation ran last week, which is already out of date;
- issue schedules in the subscription system don't match what was actually published;
- extensions, suspensions and refunds happen in one system and nobody tells the other;
- bundle splits are worked out by hand, differently each month.
The result is a month-end where finance and circulation produce two different numbers and spend a week reconciling them.
What a good month-end looks like
At month-end, finance should be able to see, from one run:
- deferred revenue by publication, in summary and in detail;
- what's been released this issue and the value per issue;
- debtors, cash, refunds and unallocated cash;
- VAT.
And all of it should tie back to the same subscription records customer service is looking at.
That's how we've built Avio's finance tools. A full month-end pack runs in one go, including deferred revenue summary and detail, value per issue and revenue released this issue. Revenue can be smoothed evenly over the term where that's your policy. Bundle revenue is split by percentage across the products inside it. Runs are saved for audit and can be downloaded as a single zip. The VAT rules engine applies VAT from the customer's VAT number, invoice and delivery countries, EU status and print or digital, with a built-in test tool to prove each rule.
For dashboards on top of that, Avio data can be connected to Metabase, set up by our team. See reporting and BI.
Questions to ask your current system
If you're not sure your current set-up is doing this properly, try these:
- Can you produce deferred revenue by publication without a spreadsheet?
- Does an extension or suspension change the release automatically?
- Are bundle splits set once and applied consistently?
- Can you reproduce last March's month-end figures today?
- Do finance and circulation agree on the number of live subscriptions?
If the answer to any of those is "sort of", it's worth a look. Our guide to choosing subscription software for publishers has more questions in the same vein.
If you'd like to see Avio's month-end on realistic data, book an Avio demo. Bring your finance director. They tend to ask the best questions.