Hacker News Reader

Comment by firebones | Hacker News Reader

Can anyone with a finance/CFO background give a primer in how a) to detect such aggressive accounting in a cloud company's earnings statements and/or b) what aggressive means in this context?

It would be nice to have a field guide to how to spot those without bathing suits when the tide goes out.

View on HN

a) All IFRS (International Financial Reporting Standards) compliant financial statements will have a accounting policies statement preceding the 'Notes to the accounts'. This accounting policy section will outline how revenue is recognised and how this is recorded in the financial statements. There is a trend for financial market regulatories to require auditors to provide a summary of 'key audit matters' in their opinion to the accounts - which details the key risk areas in the financial statements for an entity.

b) Aggressive in this context likely means that revenue (and earnings) were being improperly accounted for in the current period. My guess in this context relates to revenue recognition. For instance if I sell SAAS product for $1000 setup and $100 per month and I expect the customer to stay on average 10 years - how do I recognise this revenue?

The aggressive accounting would recognise revenue of $1000 (setup) and $1200 (subscription) in the first year. But it may be fairer (and potentially more appropriate) to recognise $100 (10% of the setup costs as customer expected to stay 10 years - apportion over this period) and $1200 (subscription) in the first year.

On this simple alteration in treatment revenue could differ from $2200 to $1300. NB: This is an oversimplification and the actual recognition criteria depends on scenario, nature and company policies.

Cloud accounting (read: SAAS) is still relatively new and very different to traditional licensing and the accounting/finance community is still grappling with the recognition and treatment.

I'm seeing, in the field, Oracle reps offering bundles of Oracle software licenses + cloud for less than Oracle software license on it's own. Is that something that could be tricky to account for properly?

That's very curious. So by adding cloud capabilities, you are effectively getting a discount.

Generally speaking, when I see hard-to-explain sales behaviours from enterprise vendors my experience is that:

1/ The sales incentives are leading to sales teams optimising their incentives in ways that seem bad for the organisation over all, and/or

2/ The contract has nasty fishhooks that will make all that money back, and more, in year two or three (classic example from my experience with a different vendor: selling servers at $50k, discounted from $250k, but then assessing maintenance on the original price, leading to a $50k/year opex).

Reduced transparency (if one does not realize what comes in year two or three) makes it really hard to make a fair comparison in year one.

As I understands it, some companies may also recognize the contract signing as the revenue and capitalize $13,000 for the year the contract was signed ($1000 setup and $1200 subscription pr year * 10 years).

If the customers cancel the contract they may then book the lost revenue as a loss, so an annual statement with high losses may be an indicator that they booked revenue before the revenue was truly secured.

It's actually wrong to book revenues until one has satisfied the performance obligations to the customers, i.e., until the services have been provided. It means that I can't book revenues for 10 years ahead under no circumstances.

It is possible though to have something called "deferred revenues" in case the customer has pre-paid for those years (i.e., transferred the cash for the 10 years), but those are not revenues (not on the P&L), but a liability to the customer to satisfy the performance obligations (on the balance sheet), and this will be gradually released (apportioned) to actual revenues over the course of the remaining years.

Good points. The new IFRS 15 tries to deal with those issues, and defines principles of performance obligations, un-bundling etc., which should be helpful in this case.

I'm sure the Big 4s have issued Q&As and position papers to their audit clients already on the adoption of IFRS 15, but this is going to be a nightmare to implement...

The gray area is how one counts revenue as cloud or non-cloud. Some software can be used in both modes, and the question is how one categorizes them. For example, Microsoft Office 365 can be used in cloud (in style of Google docs) but it can be installed locally. The gray area affects many firms, and is not isolated to Oracle.

Here is an article that may help:

http://www.theregister.co.uk/2016/01/18/cloud_sparks_boom_in...

b. Just from the article I think the aggressiveness in this case was probably doing something like assuming that current subscriptions continue for longer than they've received payment, and then counting that entire time as revenue.

From what's listed in the news story, we don't know that, but the point of the article is not to single out Oracle, but to realize that financial accounting for cloud software products is a grey area that affects many firms -- including those that sell it, and even the buyers, and having clearer standards will benefit everyone.

Reply on news.ycombinator.com