Snowflake’s Slowing Sales

On the surface, it would seem that Snowflake’s sales has been doing just dandily, but a deeper look shows some trouble brewing.

On a side-note, I have been asked off-board what I do in times like these - meaning the current massive tech sell-off. Here it is. I put every company I own under a microscope to see if there was something I may have missed and I concentrate my portfolio in only those that I have no or very little doubt about. And I try to ignore the red ink in my portfolio.

But now, on to Snowflake. I think I may have found something that I have previously missed.

The red herrings

I recently read a report that highlighted two very impressive metrics as proof of accelerating sales, with more to come. The first was that NRR is exceptional and accelerated to 173%. And the second was that the number of enterprise customers spending more than $1m per year increased by 128% which is higher than sales growth. And because this number is higher than sales, it points to further growth ahead. They then go on to say that, in addition to this remarkable achievement, enterprise customers also account for a bigger part revenue - to 53%, up from 46% a year ago.

First, NRR.

Management has told us numerous times that it takes upwards of 6 months, and with big customers closer to 9 to 12 months, to ramp up spending. So to understand what the 173% is telling us, we really need to understand how they define NRR. This is their definition:

“We first specify a measurement period consisting of the trailing two years from our current period end. Next, we define as our measurement cohort the population of customers under capacity contracts that used our platform at any point in the first month of the first year of the measurement period. […] We then calculate our net revenue retention rate as the quotient obtained by dividing our product revenue from this cohort in the second year of the measurement period by our product revenue from this cohort in the first year of the measurement period.

This is important. They define NRR as the product revenue of customers at the beginning of two years ago, for that year vs product revenue in the next year. So product revenue of n-2 vs n-1. Our other SaaS companies define NRR differently, for example Monday defines NRR as the weighted average of the ARRs from each month of two years ago’s customer cohorts vs the ARR of the same cohorts 12 months later, up to the last quarter’s end. Datadog uses the same ARR methodology but with no weighting - so is even more recent: n-1 ARR vs n ARR.

I believe this is important because we are not looking at the same metric as for our other SaaS companies. ARR is a year forward looking metric based on the current point in time whereas product revenue is a year backward looking from the current point in time. In time these metrics will be similar, of course, but Snowflake’s definition is further backwards looking vs an ARR definition which uses more recent customer counts like Monday and Datadog. I believe that if Snowflake were to define their NRR as forward-looking ARR, their NRR would already be heading south - quite a bit lower than the 173% they published most recently as I outline below.

So what exactly does the 173% mean?

Snowflake is telling us that their customers who consumed product from them in the first month of Q4 2020 - i.e. just more than 1934 customers, which included just more than 106 fortune 500 customers, generated 73% more product revenue with them in Q4 2021-Q3 2022 than in Q4 2020-Q3 2021.

Here’s the thing though. If Snowflake adds lots of customers - especially large ones like F500 customers - in the quarter or so preceding that two years ago starting point, then NRR is bound to explode, because they are ramping up very large customers from almost zero.However, if their customer acquisition starts slowing, their NRR is bound to decrease quite rapidly because of this same dynamic.

And here’s the next, bigger thing. Their year-end is 31 January. Which means the 173% describes the behaviour of customers they had on the books in the first month of Q4 FY2020 - i.e. customers on the books in November 2019, and we are comparing their product revenue for November 2019 to October 2020 to their product revenue for November 2020 to October 2021.

So the first part of their product revenue for this calc - which is variable/usage based as they point out every time (“we are not a SAAS model”) - was generated in the peak COVID months. And we know from Datadog’s results for example that customers curtailed usage-based cloud spending at all the hyperscalers in that period (Datadog’s famous Q2 dip), so it’s not far-fetched to assume the same happened with Snowflake. Of course, thereafter there was a catch-up again.

So in my opinion, Snowflake’s most recently published Q3 2022 NRR tells us about a massive COVID bump, in addition to a big accelerator-effect on NRR from very fast customer acquisitions prior to November 2019

It does not tell us what is likely to happen next.

For that, we need to listen to management, and this is what the CFO had to say on the NRR point:

Citi conference 15 Sept: “Well, clearly, as we get bigger in our installed base of customers grows, I think it will be hard to maintain at the level we’re at, but we think net revenue retention will remain best in class for a very long time.”

What do we need to look at then?

CFO, Sept 14 Piper Sandler conference:

“I do think RPO - and current RPO more importantly - is important, along with revenue. And the reason revenue is the best indicator of the future is because that’s what customers are consuming. And once you consume, you don’t really decrease your consumption. Yes, you could run certain things. But so we give the metric that we expect that 60% or sorry 56% of our total RPO will be recognized over the next 12 months that was up from what was 54%, 55%, I think it was $85 million increase in current RPO quarter-over-quarter. That is a good metric to look at in conjunction with revenue, because you got to remember, if I end up beating my revenue in the current quarter, say by throwing numbers at $10 million that took out of my current, my current RPO that I’m reporting at the end of that quarter as well, so it’s important to look at both, and how those are both growing over time.

And again, CFO, Q3 results:

“I would say the first thing is the guidance we give. The second thing is historical revenue growth patterns coupled with the current portion of RPO to build your models.”

Ok, so they guided for 84% yoy at the top for next Q. But for both Q2 and Q3 they guided for 92% and for Q1 they guided 96%. Last year Q4’s guide was 103%. So it went 103% → 96% → 92% → 92% → 84%. This is a pretty big drop in guidance from Q3 (-8%pts) and Q4 last year (-19%pts).

In terms of the other two things he says to watch, Revenue and current RPO, here are both, and how they are trending over time.

Rev	Q1	Q2	Q3	Q4
2020	43.7	60.3	73.0	87.7
2021	108.8	133.1	159.6	190.5
2022	228.9	272.2	334	350

cRPO	Q1	Q2	Q3	Q4
2021				733.2
2022	773.3	841.0	992.2	

From this we can see that cRPO vs revenue growth qoq growth for the last 3 quarters was as follows:
Revenue QoQ: 20% → 19% → 23%
cRPO QoQ: 5% → 9% → 18%

So Revenue growth is outpacing cRPO growth. Not great as an indicator of future growth.

And to get to a trend, let’s divide current RPO by the quarter’s run-rate revenue (most recent quarter last). It gives:

Current RPO as % of run-rate revenue: 96% → 84% → 77% → 74%

Which means that current RPO is becoming smaller and smaller relative to their most recent revenue. The revenue in the bag so to speak, is becoming less and less.

So the key things the CFO tells us to look at - guidance, revenue and current RPO - are not that inspiring.

But what about that 128% growth in >$1m customers?

On to the second of the red herrings above: >$1m customer growth (And here, for full disclosure, I would like to point out that in a previous post I got tricked by this herring myself: Contrary to how our SaaS customers define their >x spending customers, namely by using ARR, Snowflake defines these customers as customers who spent more than $1m in product revenue with them in the preceding 12 months. So let’s unpack this again. These customers went from 65 in Q3 2021 to 148 in Q3 2022. So they had 65 of these in the spending period of 12 months prior to October 2020. And they have 148 of these in the 12 months spending period prior to October 2021. So again - COVID constrained year 1 vs non-COVID constrained year 2 = big growth %.

Another way to look at this. In the same two periods, they had 171 and 223 F500 customers. So the number of >$1m spending customers as a % of the 171 F500 customers on their books 31 October 2020 was 38%. And of the 223 F500 customers on the books on 31 October 2021 that proportion was 66%. Now of course there are non-F500 customers spending >$1m too, but it is safe to assume many, if not most of the F500 should also reach that milestone. So of the large F500 customers they had previously signed up, many more than right after COVID have now reached a significant ramp-up of revenue. Meaning there is less ramp left from existing customers vs in prior periods.

So both the growth in ttm >$1m customers and the sky-high NRR are two ways of telling the same story: COVID plus rapid customer additions two years ago.

Now that we know that their most recent NRR and >$1m customer metrics most likely reflect a big COVID bump, was boosted by very rapid customer growth pre-Nov 2019, and shows old information, let’s turn to the important driver of future growth: recent customer additions.

The slowing sales machine

Let’s look at absolute number of customers added:

Total customer additions

#	Q1	Q2	Q3	Q4
2020				458
2021	328	397	437	585
2022	393	458	426	

YoY	Q1	Q2	Q3	Q4
2021				28%
2022	**20%	15%	-3%**

Looking at the pace of customer additions, this has been slowing in absolute terms over the last several quarters. They signed up fewer customers in Q3 of this year than they did a year ago (426 vs 437; the first absolute yoy decline in my data set), and only marginally more in the first 9 months of this financial year vs the previous one (1277 vs 1162). Not great

Let’s now turn to the addition of customers who have the potential to become very big spenders.
To see what is on the other side of the horizon, we need to look at the number of F500 customers added:

Fortune 500 additions

#	Q1	Q2	Q3	Q4
2020				20
2021	17	15	13	21
2022	4	19	8	

The table above does not need a qoq or yoy analysis to make the point that the number of F500 customers added has decelerated a lot over the last couple of quarters. They added 31 in the last 3 quarters vs 45 in the same three quarters a year ago.

But wait, that’s not all. In the last several quarters, they have been significantly ramping up the sales machine. Below is the number of S&M employees per quarter, and then two tables showing the effectiveness of this machine in landing new customers, in the form of customers added per 100 S&M employees, and F500 customers added per 500 employees.

Number of S&M Employees

# S&M	Q1	Q2	Q3	Q4
2020			907	989
2021	1082	1141	1177	1257
2022	1450	1570	1672	

Customer adds/100 S&M Employees

#	Q1	Q2	Q3	Q4
2020				46.3
2021	30.3	34.8	37.1	46.5
2022	27.1	29.2	25.5	

F500 adds/500 S&M Employees

#	Q1	Q2	Q3	Q4
2020				10.1
2021	7.9	6.6	5.5	8.4
2022	1.4	6.1	2.4	

In FY2021 each 100 S&M employees brought in more than 35 customers in the quarter, on average. This year that number was below 30 for all quarters and reached the lowest yet - 25 - in Q3. And on F500 it’s much worse of course. For every 500 S&M employees, they added a bit more than 2 F500 customers in Q3 of this year vs 10 in Q4 of FY2020. So they have more - many more - sales and marketing professionals bringing in fewer customers overall, and much fewer fortune 500 customers.

So? What did I do?

SNOW will have many more quarters of very strong growth, driven by significant new customer additions brought on a year or more ago, but that pace is going to decelerate pretty fast in my view. NRR is probably already at its peak. However, for me the clincher was the massive decrease in sales efficiency as measured by customers added per S&M employee. It’s still a great company and there are many other fantastic things about it that I did not analyse in this post - it’s tech, leadership, momentum, etc. etc. However in this environment I’m looking for companies with damn nigh no flaw, and Snowflake did not make that cut on the above analysis.

I therefore sold my position in Snowflake yesterday and used the proceeds to buy ZI, ZS, and S (MNDY is already my #1 position otherwise I would have added there).

Happy to hear dissenting views!



WSM, how do u come at 84% next Q yoy guide? My understanding is that u have compared wrong numbers.

Here is the data from the the transcript:

“For the fourth quarter of fiscal 2022, we expect product revenues between $345 million and $350 million dollars, representing year-over-year growth between 94% and 96% .”

Ur mistake is comparing product revenues with total revenues. For full picture of expected revenues in Q4 u need to add to 350m also typical 7% of that for professional services. U’d be coming at 374.5m guide for total revenues in Q4.

In fact, next Q guidance is an improvement/acceleration from 92% to 96% and not deceleration which u put in the analysis.

Secondly, NRR is indeed a backward looking indicator. This says that existing customers of those cohorts spent the same amount as in previous period +73% (173%). I like that. If u are worried of Covid bump - then please compare with NRR in previous periods which should cover pre-Covid cohorts.


FY19 FY20 FY21

180% 169% 168%

I like all these numbers. Typically each cohort continued to generate the same rate of revenues + 70% each year. This trend is being confirmed by the last Q ER.



Ok, so they guided for 84% yoy at the top for next Q. But for both Q2 and Q3 they guided for 92% and for Q1 they guided 96%. Last year Q4’s guide was 103%. So it went 103% → 96% → 92% → 92% → 84%

Am I missing something here? Their Q4 guide was for $350m product rev (+96%). Where is 84% coming from. This in NOT a deceleration.


1 Like

Sorry my post got crossed up with LearningInvest0r’s who articulated it much better :slight_smile:


1 Like

first, thank you for all your posts.
Second, regarding Snowflake.
On page 9 of their investor presentation they state: “WE ARE NOT A SAAS MODEL. 93% of revenue is consumption based. Revenue is recognized only when consumption occurs”.

So they don’t have ARR (annual RECURRING revenue) - their revenue is consumption based, just like Twilio which charges depending on how much text messages were sent by a customer.

This is a good thing because NRR of companies with consumption based pricing is typically bigger than SAAS companies (check out BVP Stat of the cloud deck for examples).
The other side of this is that the consumption is not guaranteed by a contract. But in fact, they numbers say that when a customer starts using their platform, she is using it more and more.

Also, you state that “However, if their customer acquisition starts slowing, their NRR is bound to decrease quite rapidly because of this same dynamic.”
But NRR is calculated with the same cohort of customers. New customer additions do not impact NRR in any way.
Best, Krokotot



Thanks for the insightful post. I need to think more about the customer add numbers but here is what I found on RPO/cRPO and the net retention rate.

(1) On RPO and cRPO:

For consumption based businesses, RPO is not a very good gauge of revenue because it contains only the contracted minimum and doesn’t capture the actual consumption aspects which really is the major portion of revenue.

Nevertheless I think the CFO addressed the RPO issue and said that Jan 2022 Q will see a very strong RPO (and probably that means cRPO growth also) so that should inflect in the coming quarter.

Here are his words in the UBS TMT conference on 6 Dec 2021:

Analyst: Maybe just flipping to another metric that got a lot of
attention after your print was the RPO growth. So maybe you could just frame how relevant this metric is,
because on the one hand, you’re putting up unbelievable product revenue and usage. On the other hand, your RPO growth is decelerating. So, how do we square that, Mike, and get comfort that the RPO growth has sort of troughed or will trough near-term?

CFO: So what I will say is, first of all, we only really started selling multi-year contracts at the beginning of this past fiscal year. So there is some lumpiness in that. And we’re more focused on current RPO, and that’s why we give that metric of what we think is going to be recognized. Yes, I think this quarter is going to be a big quarter of RPO growth. A lot of that though is driven by, I know we have a number of significant customers that have to renew and buy more this quarter.

(2) On Net Retention Rate

I agree that 173% is too high and they admitted it as such, again in the UBS conference. But let’s take a look at the prior quarters where it stayed consistently at 168% - 169%

Jan 2021 Q: 168%
Apr 2021 Q: 168%
Jul 2021 Q: 169%
Oct 2021 Q: 173%

for Jan 2021 Q: this will reflect the customers 2 year prior (i.e. Jan beginning of 2019 Q) and the 168% is comparing the Covid constrained year (year 2) vs. the prior pre-Covid year. And they still managed 168% despite the Covid constraint.

So yes, 173% is too high but I think it goes to the high 160s.

Here’s what the CFO said again in the same conference:

"At the beginning of the year, we were just saying it was going to be above 160%. I thought it’d be in the high- 160s. I wasn’t expecting 173% … there were some of our top 10 customers that over-consumed, but we did see overconsumption on average across the board from our customers. "


To follow up on the RPO and cRPO point above, first of all, the CFO said in Dec that the Dec quarter will see strong RPO (and hence cRPO growth) so I am not as worried about it.

Also, I find RPO and cRPO can be quite misleading in general as it is very skewed by the renewal dates: suppose your customers renew in the first quarter of the year. Your cRPO will: (1) swell for that quarter and then (2) it will run down for the successive quarters giving the appearance of decline; and then (2) swell again in the first Q of next year.

I believe for consumption based businesses, the best metric is (1) revenue growth; (2) net retention rate; and (3) customer growth.


NRR is calculated with the same cohort of customers. New customer additions do not impact NRR in any way.

I think wsm’s point was that adding a bunch of new customers just prior to the 2-year window results in customers who were technically in the cohort but had barely begun using the platform. After a year when they’ve ramped up, they appear to have had massive consumption growth, but it was really the one-time bump of getting started. As fewer customers are coming on board and the NRR cohorts are mostly long-time customers, it will necessarily decrease. I agree with this, though it’s hard to know how significant of a factor it is. This is consistent with customers ramping up faster as SNOW improves their onboarding experience, reflected in the downward trend in RPO growth versus steady revenue growth. (Faster consumption means less lingering RPO.)

I also agree that the way SNOW calculates NRR makes it a bit further backward-looking than some others, so decreases there will presage revenue growth decreases further in the future. However, their NRR is not yet decreasing! Others have posted the numbers – there is no trend yet, and even when it starts it will still be absurdly high. Why not wait until this actually manifests?

Lastly, on the topic of COVID: I see two conflicting ideas here. One is that NRR is artificially high because the first-year spend was depressed as companies pulled back on usage-based spending during COVID. The second is that they got a “COVID bump” pushing the second-year spend higher. I don’t see evidence for either of these. Here is their QoQ revenue growth:

Rev	Q1	Q2	Q3	Q4
2020		37.99%	21.06%	20.14%
2021	24.06%	22.33%	19.91%	19.36%
2022	20.16%	18.92%	22.70%	

Other than Q2 2020 (which was May - July 2019 if I counted correctly, long before COVID) their revenue growth has been consistently around 20% QoQ and I see no discernible shift with respect to the pandemic.

So while I agree with the cautions around interpreting NRR, I don’t yet see evidence that it’s hiding weakness in revenue generation, nor that it is obscured by COVID factors.

The slowdown in F500 customers is interesting, but also inevitable. They can only ever have 500 of them, after all. The long-term thesis for SNOW depends on existing customers continuing to grow their spend over time as they continue to generate data and derive more value out of the platform. That, to me, still seems to be the case.


“The slowdown in F500 customers is interesting, but also inevitable. They can only ever have 500 of them, after all.”

Yes, in fact out of their top 10 customers, only 4 are Fortune 500 companies. Another 4 are start-ups founded less than 10 years ago.

The biggest spenders on Snowflake are not necessarily the largest companies, but the most digitally and data driven companies.


Great post WSM! I have also been worried about RPO and cRPO. As with most RPO numbers, it can be lumpy but what is worrying is that the CFO keeps saying it’s important to look at. He’s not dumb either – probably one of the best pedigrees of any software CEO in the world. He led ServiceNow for years with Slootman and they absolutely crushed it.

I would, however, be more worried if we see that show up in the actual revenue numbers. I realize that’s your point – customer growth/RPO/cRPO/NRR are all pointing towards that the future could be less bright.

But so far, revenue looks rock solid. I think a few other posters pointed this out but I believe your YoY guidance numbers are a little off for product revenue.

This is what I have for the past 5 quarters of YoY guides:

96% (the latest for Q4)

And I also look at QoQ guidance versus the actual numbers. That has been rock solid at 12% for each of the last 5 quarters. That is incredibly steady!

NRR will bounce around and that’s probably why we saw so many $1 million customers added. I think customers are growing into that rather than landing at $1 million.

The other worrying thing as you point out is that the net new customers for Q3 was actually lower than the Q3 add in the previous year (426 vs. 437). The only thing I can think of is that Snowflake is incentivizing its salesforce to really grow their existing accounts.

So I think focusing on the NRR makes a lot of sense. If customer growth can stay above 40% and NRR can stay above 60%, they should be in the rough ballpark of 100% growth. Once the NRR goes below 160%, then we may see revenue start to decelerate.



WSM, how do u come at 84% next Q yoy guide? My understanding is that u have compared wrong numbers.

Here is the data from the the transcript:

"For the fourth quarter of fiscal 2022, we expect product revenues between $345 million and $350 million dollars, representing year-over-year growth between 94% and 96% .”

Thanks LeaningInves0r and Fish. Yes, you’re right, of course. I incorrectly entered the product revenue in my own spreadsheet and didn’t double-check it. That is primarily because I added the guidance part as an afterthought; it is not the main thrust of my argument Still, that part of my story should read:

Ok, so they guided for 96% yoy at the top for next Q. For both Q2 and Q3 they guided for 92% and for Q1 they guided 96%. Last year Q4’s guide was 103%. So it went 103% → 96% → 92% → 92% → 96%. This is a bit of a drop from Q4 last year (-7%pts).

The guide is accordingly not a red flag.

The main thrust of my argument, however is about the slowing customer growth and the impact that is bound to have in a couple of quarters on revenue.

Krokotot, this is the point which I think is potentially misunderstood, and kinda the point that I’m trying to make:

Also, you state that “However, if their customer acquisition starts slowing, their NRR is bound to decrease quite rapidly because of this same dynamic.”
But NRR is calculated with the same cohort of customers. New customer additions do not impact NRR in any way.

Yes that’s certainly what it would look like on the surface, but I disagree. Their customers take a long time to ramp up from a very low base and they have added large numbers of F500 customers in prior years. So the dynamic I’m suggesting will play out goes something like this. Suppose year 1, 2, 3 revenue from two types of customers.

Customer A (Small):
Yr1 $10.00
Yr 2 $20.00 (NRR 200%)
Yr 3 $25.00 (NRR 125%)

Customer B (Large):
Yr 1 $100
Yr 2 $200 (NRR 200%)
Yr 3 $250 (NRR 125%)

So in this example if one were to add lots of new customers - either big or small - just prior to your Yr 1 first month measuring point, you will have a much higher NRR than if you didn’t, because of the massive ramp-up in the beginning. If you, on top of that, start adding less larger customer B’s, your NRR will tank, because of the disproportionally large impact that these ramp-ups have on NRR.

BUT because of the long lag in both their ramp-up period and their definition of NRR this will only show up more than a year after the customer acquisition slow-down

catsunited, on your cRPO point

Nevertheless I think the CFO addressed the RPO issue and said that Jan 2022 Q will see a very strong RPO (and probably that means cRPO growth also) so that should inflect in the coming quarter.


Your cRPO will: (1) swell for that quarter and then (2) it will run down for the successive quarters giving the appearance of decline; and then (2) swell again in the first Q of next year.

The main issue I have here is that the cRPO as a proportion of the annualised revenue of the reported quarter has been decelerating. And I see no swelling and then running down as you suggest, perhaps you’ve calculated those and can show that happening? This is what I see for the last quarters for that metric (most recent Q last):

Current RPO as % of run-rate revenue: 96% → 84% → 77% → 74%

If you look at their key numbers (NRR, revenue) as lagging customer acquisitions by a couple of quarters, then the problems of absolute customer adds and the declining efficiency of the sales force come to the fore and it becomes clear that it will lead to a deceleration of NRR and revenue in future. And clearly customer growth has stalled as has large (f500) customer growth as well as the sales machine efficiency. So the revenue and NRR deceleration is coming - its just a question of how many quarters to go. Or at least that’s how I see it.



Customer number growth

Total custome number is not siginificant factor.

The more important number is customer with >$1m revenue. Asuming they have average revenue of just $1m, $1m x 148 is $148m. Divide by product revenue of 312m, they contribute around 50% to total product revenue ! In reality, it should be more than that because the average revenue per customer of this cohort should be more than $1m per customer. and they are growing at around 100% per year with a still small numbe of just 148 or 1.8% of total customer number. So 100%X50% = 50% growth contrubition to revenue from >$1m customers plus 70% NRR, then we arrive at 100% to 120% per year revenue growth. And total customer growth should bottom at around 7% QoQ or 30% per year. This growth in total customer is enough to keep $1m customer growing at highspeed for many years, because the customers with >$1m revenue is just 1.8% of total customers right now, the total customer base will continue to feed into customers with >$1m revenue for years to come. With current growth rate,. In 2 more years, customers with >$1m revenue will increase to just 6% of total customers.

Revenue vs RPO

Since Snow is a consumption based model business, to predict the future growth, revenue growth is much meaningful than RPO for growth trend. What good is RPO if customers never use it? LIkewise, it doesn’t matter if RPO is slowing down in the short term. When customers used up the credit and need to buy more credit, they will do so. So we may see seasonality in RPO growth. But consumption (product revenue) has been very consistently hovering around 20% per quarter. This is good sign. This is an indication of secular growth in data. Once customer moved data to the cloud, they are not going backward to on premise.

COVID affects on Snowflake business

I agreed with above posters that there’s no COVID affects on SNowflake business as we can tell the consistency of product revenue for the past 2 years. e.g. no big upswing and downswing of revenue. The revenue hovers around 19% to 24% QoQ.(22.85%,18.92%,20.16%,19.35%,19.88%,22.37%,24.07%,20.12% )

Long snow with a small 3% position.


Re the arguments above about the supposed lack of Covid impact, let’s have a look at a longer time series of NRR.

I think the COVID impact is clearly visible. Bear in mind that what I have as “2021” is actually FY 2021, i.e. the year ended 31 Jan 2021, so as close as dammit to the actual calendar year 2020.

I can see the COVID quarter there in the NRR in Q2 and Q3…

NRR%	Q1	Q2	Q3	Q4
2020		223%	189%	169%
2021	171%	**158%	162%**	168%
2022	168%	169%	173%	

So if there was a slowdown in Q2 and Q3 of FY2021/calendar 2020 as would seem to be the case judging by the numbers above, then it would be reasonable to also assume that there was a catch-up in the quarters after that which would have boosted the NRR. That is my argument above.


1 Like


in re: COVID slowdown… Snowflake is more about process and accessing and leveraging data… it’s analytics based… and that trend was impacted more negatively by COVID (aka, slowdown in spend) than positively.

Cybersecurity and connectivity capital RACED to the front as endpoints exploded. Therefore, taking priority in the IT space on spends.

Analytics, data processing, etc., is more business strategy that COVID response.

Taking that into consideration, I believe what folks are saying is that as Snowflake gets embedded into systems, what you are seeing is growth in consumption of data, which makes sense, as reports, intelligence, processing, and unlocking of data and analytics explodes due to the capability to access and tie together datasets.

Databricks would be a similar story. Databricks, based on my limited usage, work excellent, but has a high floor for business adoption as you do need to know Panda and some others programming languages to leverage the technology.

I TRULY APPRECIATE the challenge to the Snowflake numbers. What we all watch for, and ask about, as revenues grow and grow, is what is a “steady and predictable decline in growth” based on increasing revenue figures and “is there a plateau they can sustain” or “are they approaching a cliff”.

Such is the FUN of growth investing

Just a Fool


Customer B (Large):
Yr 1 $100
Yr 2 $200 (NRR 200%)
Yr 3 $250 (NRR 125%)

What a concise way to show that NRR is by nature declining over time for each cohort. Even if it’s something more like this, the point still holds:

Yr 1 $100
Yr 2 $200 (NRR 200%)
Yr 3 $300 (NRR 150%)

The reason NRR has been crushing this year is because of the huge customer adds from last year. But for NRR to stay this high in the future, they would have needed to add (this year):

  1. a lot MORE customers, or
  2. a lot LARGER customers

Last year they added 1747 customers, and this year they have added 1277 so far. So they have to add 470 in Q4 just to tie last year. More importantly, customer count would be up 42%, where last year it was up 73%.

So they won’t add a lot more (#1), so the hope is the ones they’ve added are a lot larger (#2). But we can’t point to numbers that back up the hope (e.g. RPO doesn’t). So I’m in agreement with WSM here: NRR will be coming down in the near future. That’s what keeps me out of SNOW. Best case scenario would be a gradual slowdown, but then I see it as similar to CRWD. I.e., nothing looks awful now, but things won’t be accelerating; they’ll be decelerating.



The reason NRR has been crushing this year is because of the huge customer adds from last year.

I don’t agree. The NRR is not affected by new customers entering after the 24-months-ago starting point. The cohort is defined from active customers in the 24th previous month, and then revenue is calculated from that cohort for the subsequent two years. Huge customer adds will drive the revenue and RPO growth. That’s where a slowdown in customer adds will be most evident.

What affects the NRR related to customer adds is twofold:

  1. Change in the mix of new customers to established customers when the cohort is defined – early on, there will be relatively more new customers who are only starting their ramp. Later, as the business is mostly established customers the overall NRR will decrease as more of the customers are further in their journey.
  2. Speed to ramp up, where new customers get to a more steady-state rate of consumption and consumption growth within the first few months instead of, say, taking a full year or more. (This may be indicated by the decrease in RPO relative to revenue, as contracts reach target consumption sooner.)

I’m curious what those who are staying out of SNOW would want to see on NRR. It’s just impossible to stay at 170% for years. A slowdown is inevitable. That doesn’t mean they’re not executing or the market will re-rate them soon. I can’t see anyone having a thesis that SNOW in 2024 will still have 170% NRR. The big question is whether it will slow to 130% or 105%. All of the other factors surrounding the company lead me to believe it will be the former rather than the latter.



One way of looking at such a high NRR is that customers (and Snowflake!) didn’t know how good the service was. This means new customers were dipping their toes, only to later realize how much more they need. It could also mean that Snowflake was too shy during the sales process.

They won’t let that happen with new customers. The initial sale will start from a higher base, and so NRR for the newer cohorts won’t be as dramatic. If that NRR tanks for new cohorts it doesn’t necessarily mean SNOW is struggling; it could just be they learned how to coax their customers into bigger initial buys.


As to some other points raised by folks my thoughts are as follows.

  1. Revenue growth outpacing cRPO growth

It’s not SaaS, it’s consumption business. They mentioned many times that actual revenues and revenue guidance are the most important metrics to focus on. RPO/cRPO are important but not as much as actual revenues and guidance.

Scarpelli said in the fall:

“And the reason revenue is the best indicator of the future because that’s what customers are consuming. And once you consume, you don’t really decrease your consumption.”

So, I personally mostly focus on revenues and revenues accelerated QoQ.

  1. Customers acquisition slows

This is natural for total number of customers. Please check their long-term model. The way to grow for them is to (1)acquire large customers - basically they have to go from current 223 F500 to sign up the second half of F500 companies - and (2) steadily increase consumption - that’s why they are creating verticals of different industries, increasing data exchange possibilities among the customers, increasing types of data etc. By the time they reach 10b in revs they will have around 1400 clients generating more than 1m in revenues (from 148 in Q3) with average spend of 5.5m - from around 3.5m in summer last year. And this will represent around 80% of their revenues - from around 50% in summer last year.

Clearly - they are focusing on consumption. Yes, they should be onboarding of new clients but with time there will be less total customers added - based on their business model plan.

  1. Comparison to CRWD

I don’t think it makes a lot of sense. CRWD now goes into mid market segment and look at their new customer addition - it does not look that bad. But we know that they are acquiring smaller clients which will be bringing less revenues on average.

Snowflake goes upmarket and not downmarket as Crowdstrike. Every 1m client or F500 client of Snowflake is probably worth much more than Crowdstrike’s addition of 10 or more smaller clients.

Secondly, how does it work to create new revenues for SaaS with subscription model compared to data consumption model business? If Crowdstrike wants to grow it needs to add new clients or add new modules to existing clients. If Snowflake wants to grow - it only needs to increase consumption, so addition of new clients is less critical - in principle. In real life, obviously, new clients addition should be taking place as there always will be some type of attrition of existing clients.

Now, the best part - consumption increase is practically guaranteed! Why? Because it’s data. Data is every day growing around all of us. Every second and minute another terabyte of data is generated. This is particularly true for corporate world. And guess what - there is not upper limit on that. It’s EXPONENTIAL. So, if someone is sitting on exponential revenue stream and charging very small fee for each piece of constantly flowing and growing data - what’s the potential? What’s the durability of such business model? I think it’s huge and likely under-appreciated as we still cannot understand properly exponentiality as such. Of course, there are a lot of conditions related to execution, competition etc but overall, this business is much more potent compared to Crowdstrike’s model.

  1. Comparison to hyperscalers

Hyperscalers aka BIG 3 cloud players were growing around 40-50% YoY in Q3 with annual run-rates of many dozens of billions (AWS run-rate was in mid 60b). To my knowledge there is no public data about how many new customers they have been adding but it should be very well below 40-50% of revenue growth - AWS had more than 1m customers in 2015, probably addition of new clients now is in single digits. What I’m saying that revenue growth is mainly based on CONSUMPTION and not addition of new clients for hyperscalers. I think that Snowflake is more like hyperscalers from business model perspective than our typical subscription based SaaS (Datadog is an exception as well).

To sum-up - one can think that risk/reward with SNOW at 90b valuation is lower compared to ZS, ZI and others. This is perfectly fine thinking (I have positions in ZS, ZI, S and MNDY as well). Nevertheless, I am keeping my 13% allocation with SNOW as I am betting on continuing strength in data cloud as all the companies are in early stages of moving into the cloud and Snowflake’s products are critical. I think that consumption will only be growing and Slootman should be able to push Snowflake for “triplish” digits growth for some quarters to come. And I think that slowdown from triple digits to double digits for Snowflake will be considerably slower than Crowdtrike’s path. And, most importantly, current numbers which we have been discussing here support investment thesis and not contradict it.


Thanks for diving into some of SNOW’s numbers. I agree that SNOW’s growth will eventually slow. This always happens with hyper growers. The questions are when and how fast the slowing will happen. I’m not sure that the we have enough information in the number that you presented to determine that growth will slow in short order. Your post got me to look back at the Q3 2022 (Oct 2021 quarter) earnings transcript.

Customers >$1M Spend

First, I think that looking to customers over $1M can lead to some misinterpretation particularly if we assume that $1M is a target for SNOW or we assume that the growing slows down once a SNOW customers hits $1M in annual spending.
CFO: In Q3, five of our Top 10 customers grew at or above the company’s product revenue growth rate of 110 percent year on year. So customer can grow well beyond $1M in annual spending. How much more? Take a look at this:
CFO: In the quarter, we signed a three-year $100 million deal to an existing customer, as well as five additional eight-figure multi-year deals. So, one customer contracted to spend $100M over 3 years is $33M per year (on average), but we know that the contract is the minimum spend and most end up spending more. $33M from a customer makes the $1M KPI look ridiculous as a KPI. The five other customers contracted to spend at least $3.33M per year, but eight figures goes all the way up to $99M so these customer could be contracted to spend much more than $3.33M annually. The main point here being that their existing customers can still grow their consumption much more.

Data Sharing Clouds

Data sharing is just getting started and it will make Snowflake even more sticky. Here’s Slootman in response to a question on the growth of data sharing on Snowflake:

But we are in the beginning. You know, what happens a lot in our field and in our business is that people look at modernizing legacy workloads. Those kinds of things often have priority, you know, over getting, you know, to data sharing, because we cannot even consider data sharing unless we get our data to the cloud. We start moving that workloads, we migrate our databases, and so on.

So, yeah, we are in the very early stages. But, you know, as you see from the metrics that we report on, there is a very, very steady aggressive growth happening quarter-on-quarter. But we sort of haven’t reached that tipping point yet where sort of the floodgates are open and things are just expanding at a meteoric rate. But we’re anticipating that that will happen at some point.

It’s very nonlinear in the way the adoptions is going to develop.

My investment in SNOW is a bet that data sharing is going to be big and it’s just getting started. SNOW is targeting eight vertical markets and they so far only have data clouds for two of these. The others will roll out in the future. It makes sense to establish data sharing clouds once a critical mass of customers in a vertical are customers. Some verticals (like the Public Sector) will be slower in moving their data to the cloud. And these data sharing clouds will create network effects that are just beginning now:

Slootman in response to a question on whether SNOW is seeing network effects: [network effects] by industry because the industry and sub-industry, they really induce and invoked a network effect because the entities have relationships and do business together. So, you know, some of the data clouds that we announced during the quarter like the financial service data cloud, and media and advertising, there’s a huge amount of network effect in that area. And advertising, for a lot of reasons that people know, is under enormous pressures.

So industry players that are left out of the network of data sharing will have a competitive disadvantage. I’m willing to wait for more data sharing clouds and to see if Slootman is right about the “meteoric” expansion is coming.

RPO is a Minimum Contract Number

I think this was mentioned before: RPO is the contracted amount of revenue and is therefore a minimum. The minimum can and often is exceeded. Overall, the actual future revenue will be higher than RPO.

Revenue Guidance

On guidance, the CFO was explicit that the companies tries to guide to a 5-7% beat. So call it 6% and $371M in product revenue for Q4. Adding %24M of Other Revenue and we get a $395M Q4 at 107% growth. That’s what I’m expecting and I wouldn’t call this a slowdown especially since Q3 was unexpectedly strong (management was disappointed that their guidance wasn’t better).

Importance of F500

F500 companies aren’t necessarily the biggest spenders today. Some of these stodgy old companies are behind in getting their data digitized.
Slootman: The other thing that I would say is that, you know, we shouldn’t sort of view things in the historical way that, you know, the money is going to come from Fortune 500 companies. This is absolutely not the case. I mean, you’ll be stumped if you look at the number of customers that we – who are not Fortune 500 and how high their revenue contribution is…newer enterprises, they are born in the cloud, and digital, direct-to-consumer-oriented, and they have a very different culture toward data and a very different orientation. They will definitely feature very, very prominently in our business mix. It doesn’t mean that Fortune 500 isn’t important, you know, it obviously is. But, you know, their adoption as traditional enterprises is often not as fast as the newer entities

Sales Force Efficiency

I like to give SNOW a bit of slack and leeway to adjust to the changes that the company made last year. They redirected their sales force at the eight verticals. This doesn’t happen overnight, but it seems to be working effectively as evidenced by the fact they are sticking to it and by how Slootman described it:

That is a tried-and-true part of, you know, of our business. You know, others – and this is a reason, you know, why verticalization is such an important trend is, you know, that has very little to do with legacy workloads. It has everything to do with preparing companies for their digital direct-to-consumer futures, you know, where they’re trying to really mobilize data, monetize data, and make data really the core, you know, of what they’re doing. And that is not just in – those are not just in enterprises as you typically think of.

But even in enterprises like in retail, they are traditionally brick-and-mortar. They’re becoming very, very focused on their data – on the data that they have, the value that it has, and how they monetize that. So, these are very, very big things that are happening. Everybody is realizing they’re sitting on something extraordinarily valuable if they can enrich it and mobilize it, you know, in the right way.

So, you know, demand comes from many different places. You know, the sales organizations are, you know, over the last year, are being redirected to really understand the customers’ context. You know, what are their challenges? What are people doing? What are people not doing? How are they approaching it so that we can bring that value, you know, to our customers so it’s no longer hey, you know, here’s our architecture versus the next guy? You know, let’s benchmark and POC these workloads to see how they do. That was historically what Snowflake did.

Now, we’re completely leaned into the customers’ context, you know, what are their issues, their challenges. And, you know, we are becoming experts at their business, and that is a big evolution. It’s very exciting because it also elevates us inside the enterprise. You know, we’re no longer, you know, talking exclusively to IT types and CIOs and so on, but, you know, we’re now talking to the business side of IT.

This is a very different conversation.

I see the same numbers that WSM presented, but I’ve arrived at a different conclusion. SNOW is just getting started. Phase 1 of the journey for a customer is to get the data digitized on SNOW. Some customers are far along on that while many others are just getting started while others are not yet SNOW customers. It’s my view that SNOW’s best days are still ahead, but I do like what I’m seeing in the numbers, in the strategy, in the execution, and in the discussions by management.



“I’m curious what those who are staying out of SNOW would want to see on NRR. It’s just impossible to stay at 170% for years. A slowdown is inevitable. That doesn’t mean they’re not executing or the market will re-rate them soon. I can’t see anyone having a thesis that SNOW in 2024 will still have 170% NRR. The big question is whether it will slow to 130% or 105%. All of the other factors surrounding the company lead me to believe it will be the former rather than the latter.”

Yes, most companies cannot sustain 100+% revenue growth indefinitely but the key is the slope of the growth slowdown: is it a gentle descent or is it falling off a cliff?

Also, my observation is that the market does not penalize growth falling from 100% a year to 80% a year or even 60% a year (because a slowdown from those stratospheric heights is inevitable). At those levels, the revenue growth is so fast that it more than offset slight multiple compression. BUT it does penalize you if you continue to fall below that as hypergrowth investors start to look elsewhere.

Let’s have a look at Datadog:

Year           2018   2019   2020   2021E
Revenue         $198  $363   $603   $1,014
% YoY           97%    83%    66%    68%

Stock price
returns (%)     n/a    n/a   161%    81%

Look at Datadog - from 83% growth in 2019 to 66% growth in 2020, the market rewarded it with a 161% returns. When Datadog was growing in the 90s it was not even public and had a far lower valuation than when it was growing in the 80s and in the 60s.

In my opinion, a deceleration in snowflake’s revenue from 100+% to the 90s or even 80s is expected and should represent the golden period in its stock price. It should not be a selling event.