RevRecGals
Insight into the operational side of revenue recognition accounting. A casual conversation between two revenue consultants with 30+ years of combined experience. Join us as we discuss the practical application of ASC 606 and hear stories of how revenue recognition is implemented in various companies and industries.
RevRecGals
E7 Variable Consideration
This episode is all about variable consideration: how it impacts the transaction price and what factors play into its valuation.
EP07 Variable Consideration
Published: May 25, 2023
Welcome to the RevRec Gals Podcast, where two consultants with over 30 years combined experience share stories about the implementation and challenges of revenue recognition accounting.
I'm Susan. And I'm Natasha. And we are the RevRec Gals.
<Natasha> Welcome to another episode of RevRecGals. Today, we are going to be talking about variable consideration, which is a big, meaty topic. I have a love-hate relationship with variable consideration. It's my favorite and least favorite all at the same time. Conceptually, at a high level, it's this idea that the transaction price will be defined in a contract, but it may not be a flat, fixed fee. It could vary based on different elements in the contract. And I think there are many things that could cause those variations. The refund rate, which we alluded to when we were talking about termination clauses, there's SLA credits. Susan, what is your relationship with variable consideration?
<Susan> Well, there's just so much to it, like you said. I mean, when you're talking about your contract value, there are some high-level things that I think every company has to deal with. That goes to the guidance of saying that you don't necessarily have to adjust each contract if you want to do it in a portfolio. Typically, what a company would do is look at their trends. So instead of adjusting the contract value for every single contract, you're going to look at your trends and then do a high-level adjustment to your revenue line. And there could be other things, refund rights, most-favored nation clauses, if by chance you do have a liability there, calculating that and doing an adjustment to your revenue. Have you seen other types of reserves that people do?
<Natasha> When I've done reserves, typically it's been on refunds, and then figuring out, I think it gets a little nuanced when you're trying to figure out what goes into your bad debt assessment, which is part of your write-offs, your AR reserve. And that's really part of the collection side of things. And then what's part of your variable consideration, which is really the refund. So, what's bad debt versus refunds? And that can be a little nuanced when you're going through all those historical credit memos, historical patterns. What really qualifies as a refund? And so, I'd say that's the most common one is SLA credits, I see a lot in contracts, but typically most companies don't issue very many SLA credits.
The one time I've seen SLA credits is when a company actually had an outage, and they knew that they were possibly going to have clients exercise that clause, and so they put a reserve in anticipation of that potential.
One of the companies I worked for, they had a huge outage. Most of their, the nature of their business, most of their clients were retail clients, and they had an outage on Cyber Monday, and it was a total disaster. There was a war room.
<Susan> Oh my gosh.
<Natasha> It was very intense. All hands were on deck, and it was this process of, okay, who has an SLA credit that we automatically owe something to? And then who doesn't have an SLA credit, but we're going to have to do something to keep these people happy. Even for the people who had SLA credits, are they going to require more than what they had written in their contract? So that was a very intense time.
From an accounting perspective, at what point does this become a business practice? Because all of a sudden, we're giving all these concessions, and at what point do we have a practice of issuing concessions? In this case, we worked with the auditors to say, this was a one-time thing. We don't expect this to happen. It's never happened in the last eight years. We don't expect it to happen again. In fact, we're going to do everything we can to prevent it from happening again. We were able to say that it wasn't a practice of issuing concessions. That being said, I do have customers that do have a practice of issuing concessions. And under ASC 606, there's a way to deal with that through variable consideration.
<Susan> Well, I've also seen a lot of that during COVID, because it's a one-time event, and people know that either they're gonna have more inventory sent back, or they're gonna have more issues. And so, they'll do a separate reserve, a one-time reserve for that.
<Natasha> That's a great point. So with COVID, that's kind of like this one-time event. Well, hopefully, hopefully we don't go back to have another global pandemic. It was sort of this one-time assessment that happened, but you did have to reevaluate your transaction price, because I think a lot of contracts were being renegotiated at that time, particularly in certain industries. Travel industries, for example, some of my clients were saying, their usage just went down, way down. Their ability to pay, we need to reevaluate potentially for some of these companies. And so, a lot of times they found themselves back at the table renegotiating contracts and potentially just letting customers out of certain payments.
<Susan> Also, I've seen where companies that sell commodities, small items like, you know, a mouse and laptops and things where they sell high volumes, the reseller may have the right to return up to, say, 2% of the prior quarter's sales or something like that because of damage. You send a whole box of product, that whole box may get damaged. They would do a general reserve for approximately that amount just to adjust for variable consideration. Or they may look at trends and just say, you know, it's trending at 1%, so we're going to reserve 1%.
<Natasha> And that's what I've seen too. I have one customer or one client that has a refund reserve. For them, the numbers are relatively low, but we put that reserve there just in case in order to have something on the books, 1.5%, I think is where it hovers around. And it's not huge, but it's important to have to make sure we're accurately recording that over the appropriate time period.
<Susan> Yeah, I've seen a lot of people, well, they'll look at history to base their reserves. I've seen another customer where they sell hardware, and periodically the customer just doesn't install it in a timely manner and asks for an extension to the support. And so, if it's a large customer, they may grant it, and they have a reserve on the books specifically for these contracts.
<Natasha> As we're talking about this, we are tiptoeing on the line of modifications too, because even with those COVID instances, it could be as simple as, we're not gonna go after you and send you to collections for this unpaid invoice, or it could be, let's renegotiate the contract, and now we're in modification land. And same thing with these correctional situations where are we modifying the contract, or are we just letting them out? At what point does concession become a modification? I think it's all facts and circumstances, but it's good to pay attention to it.
<Susan> So let's talk about identifying the stand ready versus the commitment.
<Natasha> We've talked about a few ways that variable consideration can arise, but one of the big ones that I see a lot, and I do think this is a growing trend, is this concept of usage-based fees. In the SaaS world anyway, and even the software world, this idea of usage-based fees allows companies to extract more value for what they're providing to their customers. So, they come up with a way to measure a customer's success in using their product, and they price based on that, so that they make sure that the price they're charging the customer reflects the value that customer is receiving. As more and more companies transition to that sort of model, we have all these usage-based contracts, which gets my head spinning. Because what happens is you have a specifically defined price, but you may or may not have a flat or fixed transaction price. And I think the key here is figuring out what are you really providing? Are you providing a stand-ready performance obligation for a period of time, let's say one year, where you will provide the customer as much usage as they require during that time? So, it's up to infinity. And you will charge them on a per-unit fee for whatever that usage is. In contrast, are you providing X number of units over a period of time? And when they run out of those units, they will either get cut off or need to purchase more. And so, I think that's the difference here is, are you providing however much they want? It's a defined but undetermined fee. It's this variability that you don't know until the usage has actually happened. Or have they committed to a very specific number at a per unit price? And then any additional purchases are considered just that, an additional optional purchase that represents its own contract at that time. I kind of gave very clear definitions of what that might look like, but there's many instances where it could look a little bit gray. Maybe it looks like both of those, and it's unclear which way you should go.
<Susan> I've seen where they have a minimum commitment and whether they meet that minimum or not, they have to pay a certain amount. But then if it goes above it, it's a per unit pricing. You always pay a minimum, and then you basically pay an additional fee if you go above it. So, it's written in a different way to what you were talking about, where you have a bandwidth, but it's the same concept.
<Natasha> And I think that's a great example of where things can get gray, and you really have to look at what specifically is written in the contract. What is the process by which this happens? What are the facts and circumstances? One thing that I've assessed in these situations is, is it a substantive minimum? Do we expect them to hit that minimum basically all the time, and they're always going to go over? Or do we think that this is like, they might hit it during the contract term, but really, they probably won't. And they've just come up with a committed amount. And so, assessing what is that minimum really, which could be a company-wide strategy of how they set those, or it could be different from customer to customer. And then the other thing that I've assessed when trying to figure out, are they getting unlimited, or are they making a separate purchase transaction when they go to purchase additional units? That's part of the assessment is, is this really a separate purchasing decision? And one of the things that I saw a client factor in is who actually makes that decision? Is it someone in marketing using the product, and they just keep using, and they're not really thinking about, right now, when I add another user, or when I publish another email, or when I add another contact, this actually increases our usage, increases the price of how we use this product. That decision making isn't really part of how they use. Versus someone in IT who's very aware of, we have 100 licenses. If we go above that, we need to check in with procurement and say, hey, we need to add 10 more licenses because we're hiring more people. And they say, okay, great, we're going to issue a PO, and then, okay, I'm going to provision 10 more users to this particular application. And so, there's a very different process for that. Is it really a purchasing decision, or is it they're just using more of it, and there's not really this optional purchase that's happening each additional or incremental unit they're consuming.
<Susan> And how would you account for those differently when you're looking at your total contract value?
<Natasha> In the instance where you're just consuming more and you're not really making an additional purchase decision, that's where you most likely are triggering variable consideration, where you are, as a company, you're providing access to a platform or access to software, and the people consuming it or using it are just going for it. And so, you have this consumption-based model where they're using your product, and the price will be determined based on whatever the consumption is, and you're probably in the world of variable consideration. In contrast, then there's this idea of there's a committed amount, and each time an incremental amount is being added, so in the case of users is a typical one where you have a certain number of committed users, and you want to make a purchasing decision to add more users. In that case, you really have a committed amount and each additional purchase, even if it's predefined by optional pricing in the contract, would be considered a separate purchasing decision. I'm going to put a little caveat star there saying, of course that optional pricing has to be evaluated for material right. That's kind of another topic. So, let's just assume for a second it's not a material right. But each one of those exercises of, hey, I want to add on 10 more users, that's considered an add-on purchase in a separate contract. It doesn't actually represent variable transaction price to that original contract. That original contract still has that fixed price.
<Susan> So at the time that you sign the contract, say you have this usage-based, typically, I don't see where a customer can estimate. Unless there's a minimum, I don't see an estimate of the total contract value because it's an unknown. What I do tend to see is as you're going on and you're seeing this usage, when you come to the end of a quarter, if you have a pattern and you have a reasonable assessment of how much you are to receive or an expectation, they'll do an accrual to gross up their revenue for what they reasonably can expect to receive from that customer. And this assessment gets done typically on a quarterly basis. So instead of this whole concept we have of at the time the contract signed, figuring out what it is, this one is more of a rolling assessment.
<Natasha> That's a great point, Susan, because I think that's why this topic becomes so big. It's not a one-time assessment. You have to keep doing it. I have some clients that do it on a contract-by-contract basis. I have some clients that do it on a portfolio basis. And then I have some that do both, where they do some on a contract-by-contract basis, and they do others on a portfolio basis. I also see where coming up with that estimate to your point at the beginning of the contract, you might really not have any idea, but you may have historical transactions that can provide information. So, what they might do is say, I'm going to look at a portfolio of my historical transactions to determine what I think would be a good estimate on a contract-by-contract basis. So, they're still doing the accounting on a contract-by-contract basis, but that estimate is based on a historical portfolio of similar contracts.
<Susan> Well, that brings us into the ways to estimate that value. There's typically the most likely amount, which would be looking at your history and thinking about what are we likely to get from this customer. Some customers are really steady. There's a small band in which their usage sits. And then there's the expected value method.
<Natasha> Quickly, with the most likely amount, sometimes it's based on a contract. It'll be really obvious that the most likely amount is more appropriate, because maybe there's only a couple outcomes. Maybe there's variable consideration, like in a construction contract, for example, you get a $200,000 additional in the total transaction price if you complete the project by X date. And then you look into, okay, so it's either $800,000 or a million dollar contract. What's our history on being able to hit our timelines? Do we think we're really gonna get it? So, what's the most likely amount? So, some contracts, it'll be really obvious that that's the appropriate way of doing it. With the expected value method, that's where you really look at a weighted average of historical performance. And this is really more appropriate when you have lots of similar contracts with similar characteristics, and you can use historical outcomes to help inform your future outcomes. And so, this is where I often will use the expected value method with my clients to come up with an estimate based on a portfolio of historical transactions and use that to apply to their new transactions.
<Susan> And what about constraints?
<Natasha> The constraint. You know, I think when the guidance first came out, people were like, yay, a constraint. Accountants don't like uncertainty. We don't, I mean, we like a little bit of judgment because it makes it more interesting. But when things are so subjective, we like the idea of conservatism. What's the conservative approach? And the constraint felt like this. Oh, thank goodness they gave us a constraint. We can be conservative. I think the reality is, is this was not meant as an easy card. It's kind of a CYA in that you don't want to end up reversing a bunch of revenue. You don't want to end up, as you mentioned, accruing revenue and then having to reverse it later. The intention of the guidance is not to bring your revenue up and down and make it unpredictable or unwieldy. And this constraint is just that. It's the idea that you should only recognize revenue up to the amount where it is improbable that you will have a significant reversal of revenue. So, you have to look at both the probability and the materiality here to use that constraint. I've seen it used a little bit here and there to err on the side of being a little bit conservative, but this is in no way a way to be like, well, we're going to constrain down to zero because we really don't know. You really have to, you still have to come up with a solid estimate based on your historical transactions. If your historical transactions say you perform 10% above the committed amount, well, then that's what it is. You can't constrain that down to zero.
<Susan> The only time I've really seen that constraint to zero is in a really small company that's just selling a new product, and you just don't know.
<Natasha> Yeah, you can't prove that probability. I don't have many clients that end up using the most likely amount methodology, but I think certain industries would have more of that. I would see it there, because at some point, you have a very significant amount. 200,000 on a million dollar contract is not nothing. That's significant. And for you to accrue up to a million dollars of revenue, you have to feel pretty confident. And if you're not that confident, well, that's significant if you're wrong and you don't want to have a significant reversal of revenue. I would see potentially it being used there more successfully.
So, the last thing I think that's worth highlighting is we've talked about the challenges associated with estimating variable consideration on a periodic basis, each reporting period, usually quarterly for most companies. Even though the constraint isn't an easy card, there are a couple exceptions in the guidance. And each of them are meaty and painful to consider, but it's important to know that they exist because they can make the operational side of things significantly easier. The first one is the as-invoiced or as-incurred practical expedient, where you really just get to skip all the steps and recognize revenue as you invoice it. And so, that's a really popular one from an operational perspective. There are some pretty specific criteria in order to use that one. Another one of the exceptions is the variable consideration allocation exception, which is you still go through the five-step model for recognizing revenue. But when you get to allocation, you may be able to allocate your usage to specific periods, which would, in essence, kind of get you close to what you invoice you recognize. And so that is another popular one, because again, operationally, it's less estimation and less operational burden to recognize revenue as you incur the usage. And then the last one is the royalty exception for sales of IP or usage-based IP. And so, that's another one where you really get to hone in on that usage in the period in which it is incurred. So again, these all have specific criteria. It's painful to assess them, to be honest. But if you get through that upfront investment in time to assess whether these apply, it's a huge relief on the backend from an operational perspective. Have you seen those in action?
<Susan> I've seen royalties where you don't take the revenue until you get the royalty report. It’s typically either when you're going to issue the invoice or the customer sometimes just sends in the royalty report along with payment.
<Natasha> This is actually something I get hired a lot to do because it's this upfront investment that saves a lot of time on the backend, but you want to get it right.
<Susan> Thank you for joining us for this episode of RevRecGals. We look forward to having you join us again for future episodes.
The examples discussed are based on specific company dynamics. Check in with your auditors before making changes to your current processes.
Specializing in revenue recognition may result in employment for life. Please consult your friends and family before pursuing this career.