DOCU - When is deferred revenue growth bad?

Docusign is a company I’ve owned in some capacity since they were ~$40/share, and currently I own 6.7% position with plans to grow that position this month.

I’ve been a huge believer in their “System of Agreement” platform since before it was a formal thing and just a talking point. The pandemic was a boon to Docusign and, just like Zoom, I’m a believer that growth will continue to happen because (1) it was already happening before the pandemic and (2) I don’t think we’re ever returning to the “normal” that used to be specifically around business travel and work-from-anywhere vs. work-from-office.

With that said, I’ve been really focused on Docusign’s numbers of late as a means to consider exactly how much of a position I want to grow into.
Just to highlight the numbers I love right now before getting into my concerns, check out these growth numbers on total revenue, subscriptions, and billings since mid-2019. These are pretty stellar numbers if you ask me, with YoY growth numbers accelerating across their business in a very meaningful way(note Q1 is seasonally lower for billings, and still Q120 they outperformed the previous year)


                                  Q119     Q219     Q319     Q419     **Q120**     **Q220**     **Q320**    
  Rev Growth (YoY)(Total)          37.36%   41.08%   39.85%   37.66%   **38.79%**   **45.25%**   **53.47%**  
  Rev Growth (QoQ)(Total)          7%       10%      6%       10%      8%       15%      12%     
  Rev Growth (YoY)(Subscription)   35.96%   39.31%   40.55%   37.58%   **39.40%**   **46.56%**   **53.97%**  
  Rev Growth (QoQ)(Subscription)   7%       10%      8%       8%       9%       15%      13%     
  Growth (YoY)(Billings)           27.29%   46.57%   36.06%   39.82%   **59.12%**   **60.74%**   **63.47%**  
  Growth (QoQ)(Billings)           -18%     17%      7%       36%      -7%      19%      9%

At the same time, they’ve been maintaining low to mid 80%!! Subscription Gross Margins and Total Gross Margins of right around 80%. They’ve been consistent with those numbers since 2018.

Then look at the customer growth numbers, in particular the Enterprise growth accelerating to 69%!! YoY last quarter


                                      Q119      Q219      Q319      Q419      Q120      Q220      Q320     
  Customers (Total) (thousands)        500,000   535,000   560,000   585,000   660,000   750,000   820,000  
  Customers (Enterprise) (thousands)   60,000    60,000    65,000    70,000    85,000    95,000    110,000  
  Customer Growth (YoY)(Total)         25%       26%       24%       23%       32%       40%       46%      
  Customer Growth (QoQ)(Total)         5%        7%        5%        4%        13%       14%       9%       
  Customer Growth (YoY)(Enterprise)    50%       33%       30%       27%       42%       58%       **69%**      
  Customer Growth (QoQ)(Enterprise)    9%        0%        8%        8%        21%       12%       16%

Then look at the trajectory of their operating margins (and you can see this going all the way back to Q118 if you dig up those numbers). That’s a rock solid trajectory over several years.


                    Q119   Q219   Q319   Q419   Q120   Q220   Q320  
  Operating Margin   -20%   -27%   -18%   -15%   -14%   -17%   -13%

Lastly, Docusign has been really consistent with their guidance and their beats, averaging around 4-5% beats on top line revenue growth, and their guidance for Q4 shows them continuing the acceleration we see above.

Well….one other thing I LOVE about Docusign is whoever does their PR for Quarterly/Annual reporting. Their metrics are broken out so nicely and easy to read/understand. I really wish other companies did this as well as Docusign.

Ok, now for the critical lens……

Starting with the Balance Sheet….I’ve been tracking the big line items….Docusign uses the terminology “Contract Liabilities” but from what I understand this simply equates to “Deferred Revenue”. I have it notated as “short term” but that’s just an assumption that it’s “current”.


                                                       Q119     Q219     Q319     Q419     **Q120     Q220     Q320**    
  Cash and Cash Equivalents (in millions)               236.45   224.29   197.70   241.20   442.24   404.26   374.98  
  Lease (Long Term Debt) (in millions)                  154.78   150.49   150.36   162.43   173.75   117.86   169.08  
  Convertible Senior Notes (in millions)                445.39   451.93   458.58   465.32   472.16   479.11   486.15  
  Contract Liabilities (Short-term Debt)(in millions)   385.46   402.73   423.74   507.56   **552.35   624.03   686.19**  
  Growth Contract Liabilities (YoY)                     36%      39%      34%      33%      **43%      55%      62%**     
  Growth Contract Liabilities (QoQ)                     1%       4%       5%       20%      9%       13%      10%

Here’s the thing……Deferred Revenue has been growing at a significant rate, hitting 62%!! YoY growth this past quarter, and though I think Deferred Revenue is generally a positive sign, when it starts to balloon and never come down, I think we have some precedent for that being unhealthy (which I’ll get into in a moment), but while I might normally look at this and say “well, duh, it’s a pandemic and they’ve got a ton of new business commitments!”, the trend extends well back beyond the pandemic when you look at the numbers.

So, what other company have we seen with consistent, ballooning deferred revenue??? Remember Nutanix? Before that sounds insane that I’d compare the 2 companies (trust me I remember the shenanigans and there are many obvious ways they are different), just hear me out on the numbers.

Let’s look at the 2 (calendar) years of Nutanix’s Balance sheet from Jan 2017 - Dec 2018. The Deferred Revenue growth was getting borderline out of control, both in terms of short-term and long-term debt. There’s perhaps some explaining away, as they tried to do, with the move from a hardware to software business, but the numbers are what they are. Nutanix saw double digit sequential growth and a massive jump to 72% YoY deferred revenue growth in calendar Q3 of 2018. Now, I think the focus on selling out was more on the decelerating billings and the clear and present shenanigans with how they reported their earnings (Bear once called it “Pivotal-eque”), but Q318’s report was really when the tide seemed to turn for Nutanix.


                                                   Q117     Q217     Q317     Q417     Q118     Q218     Q318     Q418    
  Cash and Cash Equivalents (in millions)           200.74   138.36   132.45   610.45   376.79   305.98   399.79   466.01 
 
  Deferred Revenue (Short-term Debt)(in millions)   207.02   233.5    190.59   231.73   243.77   275.65   307.2    337.42  
  Deferred Revenue (Long-term Debt)(in millions)    255.98   292.57   218.25   246.27   296.12   355.56   394.61   442.44  
                                                                                                                           
  Total Deferred Revenue Growth (YoY)                                 9%       14%      17%      20%      72%      63%     
  Total Deferred Revenue Growth (QoQ)               10%      14%      -22%     17%      13%      17%      11%      11% 

So what’s happened since then you say? Well, take a look….they’ve NEVER gotten their deferred revenue under control…


                                                   Q119     Q219     Q319     Q419     Q120     Q220     Q320    
  Cash and Cash Equivalents (in millions)           445.12   396.68   233.82   211.69   262.33   318.74   504.48  
                                                                                                                  
  Deferred Revenue (Short-term Debt)(in millions)   361.43   396.67   429.43   472.68   503.31   534.57   549.94  
  Deferred Revenue (Long-term Debt)(in millions)    476.83   513.38   545.85   583.9    618.8    648.89   656.54  
                                                                                                                  
  Total Deferred Revenue Growth (YoY)               55%      44%      39%      35%      34%      30%      24%     
  Total Deferred Revenue Growth (QoQ)               7%       9%       7%       8%       6%       5%       2%

Nutanix, admittedly, had and still has many, many other problems that have been well chronicled, but what starts to happen when you can’t deliver the products and services in a timely fashion to your customers? What is the tangible consequence to this?

I searched Docusign’s past 3 earnings call transcripts and couldn’t find a single mention of the word “deferred” or “contract liabilities”, so maybe no one’s asking because it’s easily explained, but it seems to me like a number we shouldn’t just ignore, so if someone can shed some light on this I’d very much appreciate it.

Why is it that Docusign, a pure software company, isn’t able to close the loop on their business at a faster pace than their billings growth? Someone can tell me I’m blowing this way out of proportion and I’m happy to be educated. I don’t know that this will become a problem, but at this point its more a question of whether this is something worthy of keeping an eye on or not.

Much less significant perhaps, but worth noting - take a look at Docusign’s stock-based compensation as a % of revenue. Not crazy, but they’ve had a couple previous quarters pushing 30%, and numbers above 20% consistently are, comparatively high from our other stocks. In Q3, Docusign posted GAAP EPS of -$0.31 vs Adjusted EPS of $0.22, a difference of $0.53. As a comparison, CRWD reported GAAP EPS of -$0.11 and Adjusted EPS of $0.08, a difference of $0.19.


                                                  Q119     Q219     Q319     Q419     Q120     Q220     Q320    
  Employee Stock Based Comp (Shares in millions)   $42.27   $55.79   $52.74   $33.41   $53.55   $68.77   $80.92  
  Employee Stock Based Comp (% of Rev)             19.75%   23.68%   21.14%   12.15%   18.03%   20.10%   21.13%

All things considered, I’m a big believer in Docusign’s near and long term thesis and I noted this in my portfolio review, but I’m seriously considering pushing my position up close to 20% in the very near term, so this is why I’m being so hyper-critical before I make such a move.

Would love to hear your critical thoughts on this (good or bad). Thanks!

-Chris

42 Likes

I searched Docusign’s past 3 earnings call transcripts and couldn’t find a single mention of the word “deferred” or “contract liabilities”, so maybe no one’s asking because it’s easily explained, but it seems to me like a number we shouldn’t just ignore, so if someone can shed some light on this I’d very much appreciate it.

Just as a thought, maybe their customers pay for the year in advance at the level the expect to pay, and then pay extra if they overshoot. That would account for what you are worried about. (I admit that I didn’t look it up to see if that is actually what they are referring to, just giving you a possible solution).

Saul

7 Likes

Read under “Contract Liabilities” both “current” (within one year) and “non-current” (more than one year out)

Note 7: https://www.sec.gov/ix?doc=/Archives/edgar/data/1261333/0001…

The link opens to the correct spot for Note 7.

7 Likes

Chris, thanks for a great post. I think the main issue keeping Docu relatively flat in past few months (compared with the market and our other names) is the notion which was confirmed by the management that growth will slightly go down in post-vaccine world. CEO said that growth should be between pre-Covid levels (~30s) and pandemics levels (~50s). So, the market implies growth around 40s for 2021. With 40b+ valuation we need more signs on future business growth rates. International has a big potential but it won’t be rapid conquer here as every local market has own legal framework etc.

I have mid-sized position in DOCU (~8%) but recently reduced it somewhat by adding to NET as NET is my higher conviction position and its valuation became more attractive for me to add. Could be wrong with this move though if Docu will surprise and Net disappoint.

Best,
V

3 Likes

Deferred revenue is essentially prepayments for services not yet received, so it is a good thing. I would expect deferred revenue to increase as revenue increases.

For example, a customer may sign a two year contract for $100k per year and pay the full $200k up front. At the outset, on the balance sheet Cash will be $200k and Deferred Revenue $200k, because the customer is yet to receive any services, therefore no revenue has been earned by Docusign.

Supposing the customer receives services worth $100k in year 1, at the end of year 1, $100k will be recognised as revenue by Docusign and $100k will remain in Deferred Revenue on the balance sheet.

(For simplification, I’ve assumed the contract is signed at the start of the company’s FY!)

I hope this helps.

Alex

19 Likes

cfee,

I don’t see any reason to be concerned about growth in deferred revenue in DOCU’s case. All it means is that their billings are increasing (which you noted in your analysis) and the contractual lock in length is getting longer (management has said contracts are typically one year but can be up to three). In my mind if contract lengths are lengthening it means that customers are happy with the product and are no longer in a wait and see mindset.

It does mean, I think, that management is probably not sandbagging too much when they warn of lower growth going forward, since they have visibility into what customers they have locked in and can’t raise prices on. I don’t think DOCU is terribly undervalued at the moment given their size and forward growth rates, so returns on the company are likely to match closely to growth rates for the foreseeable future. I know we’re spoiled around these parts but 40-50% growth (in revenue and in stock price) is a pretty good year, imho.

11 Likes

So what’s happened since then you say? Well, take a look….they’ve (NTNX) NEVER gotten their deferred revenue under control……

Well that’s a first, I never thought I’d hear rapidly consistently rising deferred revenue interpreted as a bad thing.

Nutanix’s growing deferred revenue over the periods presented above is a fantastic summary of their successful transitions from, first, hardware to software sales (which caused rising deferred revenues) and then from a seller of software, to a subscription model, (with that beautiful recurring revenue) which leads to even more cash received upfront but, under accounting rules, prevents them from recording as much revenue up front, hence even more deferred revenue.

But just wait until that first round of subscription renewals starts kicking in (they haven’t yet) at very little cost to Nutanix!

This post really tells the story better than I could of why I’ve continued to hold a significant amount of my portfolio in NTNX, despite, what I believe, is the market not yet catching on to how much their business has grown these past two years, after a stumble in early 2019, despite it not (yet!) being reflected in the revenue line of their income statement.

When we look back at this thread a year from now, I bet that no one will interpret Nutanix’ deferred revenue growth as somehow out of control.

Also, it’s strange to refer to deferred revenue as “short term debt” and “long term debt” above. Deferred revenue is not debt. Not at all. Yes, it is a liability on the balance sheet, and comes with an obligation to provide the service that was paid for. But especially when it comes to a subscription purchase, once the initial sale is made and agreement signed, the company’s costs are very low to provide access to the software and make sure it is up and running over the course of the subscription term. Assuming there is not a high risk of the company being unable to perform the required future obligations and provide access to the software (which is not much of a risk for NTNX’s subscriptions), then huge, and growing, deferred revenue would be hard to interpret as anything other than a very good thing!

But I know nobody wants to hear me talk about NTNX, so back to the regularly scheduled Docusign programming…

…which by the way, I did buy some more DOCU this week. I don’t expect they are going to keep growing at pandemic-propelled 50%+ through the second half of 2021, but even as just a 30-35% grower, I think today’s stock price will look cheap in retrospect…plus I believe the Agreement cloud will really kick in, in about two years, right when the e-signature momentum slows further, probably preventing the total overall revenue growth from decelerating below the 30’s% for the next few years, and maybe even accelerating again in a couple years (although I’m not relying on that to feel the stock will perform really well going forward, regardless).

I know that’s further out, and less of a growth rate, than many on here care to focus their attention on, but that’s how I was looking at the situation when making my additional purchases

-mekong

28 Likes

Hey mekong,

Thanks for the reply. You said…
“But especially when it comes to a subscription purchase, once the initial sale is made and agreement signed, the company’s costs are very low to provide access to the software and make sure it is up and running over the course of the subscription term. Assuming there is not a high risk of the company being unable to perform the required future obligations and provide access to the software (which is not much of a risk for NTNX’s subscriptions), then huge, and growing, deferred revenue would be hard to interpret as anything other than a very good thing!”

Yea so the risk of not performing (or having the customer cancel) is I guess where my head was at w/ the Nutanix reference. If you had $1000 of deferred revenue and risked having $500 cancelled that would be a smaller risk than if you had $1,000,00 and risked having $500k cancelled…but, even with a big number, if the risk is super low based on what you’re saying, i guess it’s still mostly irrelevant.

By my saying “NEVER gotten their deferred revenue under control” I guess that’s probably presumptuous to a certain degree, in that I don’t have direct evidence that the risk’s of the mere increased liability itself have seeped into other parts of their business. That said, Nutanix deferred revenue has continued to climb at well above 30% YoY while their revenue and billings have flatlined (at best) over the past 2 years. They also have had wider operating losses during that time. But…you’ve certainly provided an plausible explanation for some of that with - "transitions from, first, hardware to software sales (which caused rising deferred revenues) and then from a seller of software, to a subscription model, (with that beautiful recurring revenue) which leads to even more cash received upfront but, under accounting rules, prevents them from recording as much revenue up front, hence even more deferred revenue."

From what I understand, the Nutanix subscription term lengths are 1-5 years. I guess maybe they’ve just been locking in a large majority of their subscriptions at longer end of that? Otherwise you’d think that revenue recognition would have started showing up (unless bogged down by that transition process you described) and leveling the deferred revenue, and therefore taking that liability and thus the risk off the table of cancelations or service delivery issues.

But to use your “back to regularly scheduled programming” reference - the Nutanix comparison was probably erroneous (I should have checked to see if there was any evidence of cancellations before posting). Docusign’s subscription plans are I’m pretty sure just annual or monthly, so I guess that makes it apples to oranges, but it should also imply (for Docusign) that revenue recognition should have a stronger correlation to that deferred revenue on a quarterly and annual basis, and thus we shouldn’t see spikes in deferred revenue that don’t also find it’s way to revenue recognition in a timely manner.

-Chris

1 Like

Okay, I’ll make a case. Rapidly rising deferred revenue, by itself, is certainly a good thing as it indicates accelerated Billings and provides insight into the revenue pipeline, and deferred revenue are more assured than are bookings (aka remaining performance obligations).

However, if a business is playing fast and loose with the cash offsetting the deferred revenue liability (like making a non-accretive acquisition with those customer advances) then one could end up with a highly-levered balance sheet, which could lead to a precarious situation should billings decline in the future. If taken to the extreme, it’s possible that the business may lack the cash to actually perform on the contract. For SaaS companies, though, such a scenario is unthinkable as the cost of sales is minimal, 10-20% maybe; this is a much bigger concern for companies with high costs. Imagine Boeing using its cash from deferred revenue to buy a new factory, the factory takes a year to retrofit, and after six months, Boeing is cash poor and needs to hit the capital markets just to run its current assembly lines.

In short, as long as the cash balance keeps pace with the deferred revenue balance, then we want as much rising deferred revenue as possible, as revenue—-by definition—will eventually be recognized for that amount.

Eric P.,CPA

38 Likes

Well that’s a first, I never thought I’d hear rapidly consistently rising deferred revenue interpreted as a bad thing.

It depends, look at it this way, it is carried as a liability on the balance sheet. So it is important to follow the cash flow. A company could have $1B of deferred revenue on the books as they collect the payments. But if they spend like crazy and they hen orders slow down current cash flow will as well. If orders stopped, there still is 1B of revenue coming as long as it can be delivered. But not cash coming in to support it. For companies growing deferred rev quickly, hopefully they have a good amount of positive cash flow to reflect that it can support future deliveries of services

If you had $1000 of deferred revenue and risked having $500 cancelled that would be a smaller risk than if you had $1,000,00 and risked having $500k cancelled…but, even with a big number, if the risk is super low based on what you’re saying, i guess it’s still mostly irrelevant.
</>

I am not sure in what way it would get cancelled unless you had a large invoice that they then refused to pay so you had to reverse the entry. But the majority of it would have had the cash collected already so cancelling it is difficult unless some serious issues which would requiring to return the payments

1 Like

However, if a business is playing fast and loose with the cash offsetting the deferred revenue liability (like making a non-accretive acquisition with those customer advances) then one could end up with a highly-levered balance sheet, which could lead to a precarious situation should billings decline in the future.

A company could have $1B of deferred revenue on the books as they collect the payments. But if they spend like crazy and they hen orders slow down current cash flow will as well.

In the two quotes above, both Eric and jdc make the very valid point that, if a company is spending/wasting their cash in an undisciplined way, then yes, certainly that is a very bad sign and good reason to avoid the company as an investment.

I would simply say that possible overspending is not correlated to whether a company has a lot of deferred revenue, or rising deferred revenue.

It doesn’t matter where the cash came from, whether it be from an IPO, or bank debt borrowings, or deferred revenue, or other prepayments such as the float an insurance company gets from their premiums before they have to pay out subsequent claims…if the company blows through cash without building value in the business, then sure, that would be bad.

Most tech companies we watch (not all, but most) are still more focused on land and expand, verses maximizing profits or cash flows. Of course we don’t want to invest in companies that are unsustainable and going to run out of money, it’s just that short term profit maximization is not top of mind for them.

So, at least in my opinion, it feels like a stretch to say that companies that are getting paid more and more money in advance for their goods or services, resulting in more deferred revenue, (a good thing), are more prone to spending their money in a manner that isn’t good for the business.

It’s a very real risk that a company could outspend beyond what they can afford to, in any business, I just don’t see the connection to deferred revenue.

-mekong

10 Likes

It is hard to imagine that any company will go back to manual signatures fax etc after trying Docusign. So in that way it is very sticky. And still plenty of TAM left, because the TAM is almost all companies that require document signatures. The success of Agreement is TBD, but it sounds very useful to me.
OTOH it will not be very hard for other software companies to to copy DOCU products. The more Docusign expands their offerings (Agreement ? ) the harder it is to copy

re theater of the absurd - several years ago I bought a house. The price and terms required a lot of dickering, fax after fax. By the time it was finally purchased the copies of the copies of the faxes were illegible.

3 Likes

I am not sure in what way it would get cancelled unless you had a large invoice that they then refused to pay so you had to reverse the entry. But the majority of it would have had the cash collected already so cancelling it is difficult unless some serious issues which would requiring to return the payments

While the deferred revenue can generally be considered to the reliable revenue (thus the premise of Saul’s SaaS philosophies) I think we all agree it is usually not guaranteed revenue.

Usually a SaaS customer (depending on the contract language) can cancel if the seller fails to deliver as agreed upon. Maybe it is a new capability that was promised, fixes to the underlying SW, promised expansion of user support (think RingCentral Meeting limit on number of attendees), etc. that the company fails to deliver in the timeframe promised.

Also, many SaaS and Software customers to not pay up front. That was a big selling point of moving from perpetual (high up front cost) licenses to term or subscription models. Payment terms for large organizations are usually quarterly, or in some cases annually, electronically paid net 45 from receipt of the invoice over the term of the multi-year agreement. Others can be monthly (think of Adobe’s Creative Cloud annual subscription for personal use). For large enterprise contracts, payment terms and cancellation rights are just 2 of the negotiation terms in the deal.

So if failure to deliver occurs and a customer cancels, it generally isn’t an issue of refusing to pay an invoice and asking for money back. It will likely be a 60 day cancellation notice delivered prior to a specific date stated in the contract and thus the next invoice won’t even be sent. Then the entry is not reversed (reversing entries in accounting is bad), the lost future revenue is removed from the balance sheet.

My understating is that Docusign has a mix of term lengths and payment terms. I do not know their cancellation policies, though have reached out to someone at a large corporation who negotiated their last agreement to see if he will share what their terms are.

So to answer the OPs question of “When is deferred revenue growth bad?” – one example is when a company begins to fail to deliver and runs the risk of cancellation of existing contracts.
Is this likely to happen, probably not to a significant level. But it is a possibility that needs to be accounted for – thus the term “Liabilities”.

And to mauser96’s “not be very hard for other software companies to to copy” comment and house buying example:

  • I too am concerned about the small moat around DOCU’s business
  • I just refinanced and had to spend an hour with a traveling notary (wearing masks and sitting 5 ft apart) signing all the documents. I can’t wait until Docusign can be used for the mortgage financing closure activities (it’s already used for the application side) and they start collecting those fees too.

I am long DOCU - though it is a small % of my portfolio and am slowly increasing it.

1 Like

While the deferred revenue can generally be considered to the reliable revenue (thus the premise of Saul’s SaaS philosophies) I think we all agree it is usually not guaranteed revenue.

Usually a SaaS customer (depending on the contract language) can cancel if the seller fails to deliver as agreed upon. Maybe it is a new capability that was promised, fixes to the underlying SW, promised expansion of user support (think RingCentral Meeting limit on number of attendees), etc. that the company fails to deliver in the timeframe promised.

Also, many SaaS and Software customers to not pay up front. That was a big selling point of moving from perpetual (high up front cost) licenses to term or subscription models. Payment terms for large organizations are usually quarterly, or in some cases annually, electronically paid net 45 from receipt of the invoice over the term of the multi-year agreement. Others can be monthly (think of Adobe’s Creative Cloud annual subscription for personal use). For large enterprise contracts, payment terms and cancellation rights are just 2 of the negotiation terms in the deal.

So if failure to deliver occurs and a customer cancels, it generally isn’t an issue of refusing to pay an invoice and asking for money back. It will likely be a 60 day cancellation notice delivered prior to a specific date stated in the contract and thus the next invoice won’t even be sent.

The transaction that will normally create the deferred revenue entry on the balance sheet is the invoice. The cancellation of the contract in your example is like cancelling a PO, it would be a debook or a drop in the backlog. A debook is neither going to reverse revenue or reverse deferred revenue though as impact on the future of the company as they will likely miss expectations. But backlog does not sit on the balance sheet.

If no invoice, then no revenue or deferred revenue entry

A broader point re deferred revenue. In the large company I work, the vendor management team usually pushes back on prepayments covering multi-year contracts. And Saas companies in my experience usually ask for annual pre-payments. But try to sign up companies like ours for multi year contracts that should feed into higher ARR. If this is normal, rational behavior, typical deferred revenue should be highly correlated with next year actual revenue. Am I missing something in this reasoning?