AYX RPOs

Investment Analysis Clubs Saul’s Investing Discussions
1

The 10-K for 2019 was released. In the 10-Q/10-K the company detailed the RPOs (remaining performance obligations which is “contracted revenue that has not yet been recognized, which includes deferred revenue on our consolidated balance sheets and unbilled amounts that will be recognized as revenue in future periods”. I have been tracking this for the past few quarters as it gives us a view into how much revenue will be coming over the next 2 years from contracts already signed by customers.


**Qtr   RPOs      Subportion to be recognized by date in right-most column**
Q418  223.1	196.4	12/31/2020
Q119  214.0	178.1	12/31/2020
Q219  238.8	181.3	12/31/2020
Q319  271.8	236.3	9/30/2021
Q419  407.0	340.1	12/31/2021

You will notice that the RPOs made a BIG jump in the past quarter. $340.1m in revenue will be recognized in the next 2 fiscal years (FY20 and FY21). And so far $66.9m in revenue in contract to be recognized in FY22 or later (probably all FY22 since AYX has 3 year contracts and presumably no longer duration contracts than 3 years).

66 Likes
2

It must be nice to wake up in the morning on the first day of your fiscal cycle and have 60% of your revenue for the year already in the bag. What an incredible advantage the SaaS model has over traditional business models…

Rob

4 Likes
3

It must be nice to wake up in the morning on the first day of your fiscal cycle and have 60% of your revenue for the year already in the bag.

Yes, it’s nice to have so much future revenue already in the bag. But if I were to be a bit skeptical, I would question why they need to do longer term deals and what kind of additional terms they are giving to get these deals. True SaaS companies theoretically shouldn’t need to do such deals.

How much pricing is lost now and on following deals as a result of doing these longer term deals now?

Also, one wonders how the net retention rate is calculated given such deals.

I’m not too concerned about it, because gross margin is great and even improving.

Just thinking out loud.

DJ

2 Likes
4

Yes, it’s nice to have so much future revenue already in the bag. But if I were to be a bit skeptical, I would question why they need to do longer term deals and what kind of additional terms they are giving to get these deals. True SaaS companies theoretically shouldn’t need to do such deals.

From my understanding it’s the large companies who are asking for three year deals. they don’t want to have to bother with a new contract every year. And since they are taking larger and larger contracts for a large part of their workforce, they know they’ll be in for at least three years.
Saul

13 Likes
5

" But if I were to be a bit skeptical, I would question why they need to do longer term deals and what kind of additional terms they are giving to get these deals. True SaaS companies theoretically shouldn’t need to do such deals. How much pricing is lost now and on following deals as a result of doing these longer term deals now? "

I can’t speak to their business. But I am in a high-tech product development business. This seems very reasonable to me. We routinely sign multi-year agreements with vendors that both sides are happy with. They know they have a reliable income steam from my company for 3 years, and we know that our supply chain and customer support is ripple free. I get regular cost-free updates of software compilers and other tools, documentation fixes, and quick-turn customer support. I send an email: “I have isolated what appears to be a bug that I don’t think is mine”, and someone is back at me in a day or less to help me. I like this model. I had the money to buy the 3-year ride, and now I have 3 years with no surprises and free updates.

17 Likes
6

Yes, it’s nice to have so much future revenue already in the bag. But if I were to be a bit skeptical, I would question why they need to do longer term deals and what kind of additional terms they are giving to get these deals. True SaaS companies theoretically shouldn’t need to do such deals.

I believe AYX is interestingly unique… its SW is licensed to run on-prem as well as cloud… so its not serviced only through cloud where even monthly subscription is feasible…

Since AYX is so critical for large organizations, have flexible deployment options, customers want consistent updates etc.

So my read is that AYX is able to force 3 years license with terms that help them recognize 30% to 40% revenue upfront and rest over 3 years.

This looks like a hybrid of upfront license + a locked in subscription… its best of both worlds… one can do that only if one has really unique and high value offering.

4 Likes
7

GauchoChris and others are much better with the detailed analysis, but I do notice the details that matter to me. One such detail is how the CEO is straight-forward and does not speak with hyperbole or generalities.

In at least the last two earnings calls (I’m not all the way through this current one, but have gone far enough for this point) - maybe more than 2 - the question has been asked about if Server, or Promote or Connect (their more expensive products - with Designer being the core product that is licensed to each desk) are essentially gaining traction, relatively speaking.

This is important for many reasons, but one reason is that Designer contracts tend to be 1 year in duration (that is tend, so not exclusively, but tend that way) and Server and Promote and Connect tend more to 3 year licenses.

The answer each time is that most of their business is still Designer. Designer is what it in demand and what is making the company. Server is starting to pick up more (you get 15-20 Designer licenses in a company, you get a Server subscription), and Server is starting to pick up (but it does not sound like materially). Meanwhile Promote and Connect are still out there, but not really moving the needle yet.

The CEO does not hide this fact, or obfuscate this fact, where like with Zscaler they will say their secondary product is “fastest growing product in our history” and leave it at that, or how Pure did the same with their AI product (that ended up disappointing by quite a bit despite language to the contrary).

So the purposes for this discussion. (1) gives more credibility to the CEO and his communications with us.

(2) The more Server, and Promote, and Connect they sell, the more contract lengths will tend towards 3 years instead of 2. There has been some measurable (but not earth shattering movement) over the last quarter or two or so of longer contracts (which is probably indicative of the Server comment I spoke about above).

(3) Automation is the future, Server and Promote and Connect become more important when SIs (System Integrators) start building products to automate analytics. Promote and Connect are complicated (unless Designer, which we all know creates its value by providing so much power in such a simple package). Thus to move the needle on Promote and Connect you have to overcome much more complexity. SIs make their money making the complex manageable for their clients.

When yo listen to the Alteryx CEO he has had macro themes moving his business. He is now on Analytics 2.0 where they are moving from digital transformations to (I forget the exact verbiage - but automations, and socialization, of analytics).

So long and short, if the greater complexity of explaining, selling, and utilizing Promote and Connect can be handled by the SIs that are jumping on board (one analyst said it was all across the board, not just PwC), and the trend of automation and socialization of analytics takes hold as the new paradigm being pushed for analytics, then we might finally start to see Promote and Connect start to become more material products. These products sell for 3 year contracts (again, mostly), are higher cost products, and under 606 would really move the needle because RPOs would increase because (1) higher contract price than Designer, and (2) average contract duration longer.

I put that out there as a trend to follow over the next year or two to see if it happens or not.

Finally, in regards to competition we don’t necessarily have to take the CEOs word for it, we can see what they did in the real world. They INCREASED the price of their Server product (back in Q1 of last year). Alteryx is not discounting their software, but rather the opposite, and even at this growth is accelerating.

Some things to watch play out. It could be Promote and Connect turn out to be failed products (to date they are not what we might have hoped). AYX believes that are the corner stones of their PLATFORM from end to end that no one can match. That Analytics 2.0 is just starting and Connect and Promote should become more relevant (of course with their AI open source tech they purchased from MIT as well).

If this plays out, Alteryx will be one of those companies killing it with Designer, and then able to upsell and expand with Server (which is already an established trend with 15-20 Designer seats) and with Promote and Connect, and start selling platforms instead of just Designer seats.

I am not being facetious when I say “we can hope”. That is the company plan, but there are not facts out there right now to support that Promote or Connect will start to take off more than they have already. Management, however, is not forgetting those products and has its philosophical road map and sees those products becoming more important down the road. So we can hope, but its not something to bank on at this time.

Tinker

55 Likes
8

The CEO does not hide this fact, or obfuscate this fact, where like with Zscaler they will say their secondary product is “fastest growing product in our history” and leave it at that, or how Pure did the same with their AI product (that ended up disappointing by quite a bit despite language to the contrary).

Hi Tinker,
Or like Twilio with Flex, which was promoted like it was going to take over the world within a quarter or two. Yes, it’s nice to have a CEO who you can believe and take at face value.
Saul

7 Likes
9

SIs make their money making the complex manageable for their clients.

Back in the days when I still had a job, dealing with complexity was pretty much a daily activity. After gaining quite a lot of experience with complex problems in a business context I observed that there was a level of commonality among all these disparate problems.

I came to understand that there is no such thing as a simple solution to an inherently complex problem. While one should always strive for simple solutions, they are not always possible. That’s the bad news. Here’s the good news: complexity can be relocated. If you’re clever and careful, you can often move complexity from one area of the problem to a different area. In fact, this pretty much sums up the entire march of IT development. When I was an analyst I always strove to abstract complexity from the business process. The objective was to move complexity away from the manual aspects of the business process and relocate it into the program code.

I encapsulated this understanding in a “law” which I labelled The Conservation of Complexity: Inherent complexity may be relocated under certain circumstances, but it cannot be destroyed.

Let me emphasize that one needs to distinguish between inherent and artificial complexity. Inherent complexity is that complexity that arises from the problem space itself. Most often, inherent complexity in a business context arises from a process which demands the analysis of multiple interacting variables. An example would be solving the order point problem in a manufacturing enterprise. MRP/ERP systems abstract the complexity from manual aspects of the business process and relocates it to a computer application. Foregoing a detailed explanation of the order point logic I’ll simply state that the demand/supply balance point for a given inventory item needs to consider information from every bill of material in which the item appears, manufacturing rates, supply on hand, supply on order, lot size, quantity discounts, shipping schedule, capacity constraints (both man and machine) and a number of other variables. This problem was insoluble until computing powers became available. In its stead a bunch of inferior order point calculations amenable to human solutions were used (EOQ, 2-bag buffer, etc).

On the other hand artificial complexity is that complexity which is imposed on a problem by ineffective and/or inefficient process design. I’m sure every one reading this can identify numerous examples from daily life even if you are retired as I am. I could be wrong, but I’m pretty confident with the assertion that artificial complexity most often arises from organizational politics.

A little OT, but an observation about the value of solving widespread commonly occurring complex problems. If you look at it objectively, you will realize this is central to the success of Amazon. Bezos looked at one of the most frequently executed business process, the retail buy/sell transaction. What factors interfere with this process? Where’s the friction? The first bit of friction arises with having to physically go to where the product is on offer. He eliminated that via automation. I won’t go through the entire process, you can do that exercise for yourself if so motivated to break down the entire process from warehouse delivery to end customer receipt of the product (including possible returns). But this is the heart of what Amazon does. The genius of Bezos was to scale up the solution and make it available (for a fee) to anyone who wants to take advantage of it. This is exactly the same tactic he used to build AWS. He solved the generic data center problem for Amazon and then made the solution available to every business that wished to take advantage of it. Quite suddenly, businesses that were too small to build their own data center were able to have one. Amazon is again engaging the same strategy for POS retail. The goal is to allow the customer to walk into the store, pick up what they want and walk out (if the customer is unknown they must register before entering the store). The customer’s CC is automatically charged, the receipt is emailed to the customer. Or, if it’s an unwieldy item, punch in the desired delivery time selected from those available on a calendar. No clerks, no checkout lines, none of that. And once he’s got the bugs worked out, you can bet he will sell it to other businesses.

33 Likes
10

I’ve been in IT, building implementing and managing complex systems for 30+ years, and that is one of the greatest phrases I have seen up to now. “Inherent complexity may be relocated under certain circumstances, but it cannot be destroyed.” Brilliant!

3 Likes