Confluent (CFLT) Result

  • Total revenue of $103 million, up 67% year over year
  • Confluent Cloud revenue of $27 million, up 245% year over year
  • Remaining performance obligations of $385 million, up 75% year over year
  • 664 customers with $100,000 or greater in ARR, up 48% year over year

Confluent Cloud
26% of Q3’21 revenue | 245% y/y growth
Confluent Platform
64% of Q3’21 revenue | 40% y/y growth
10% of Q3’21 revenue | 45% y/y growth

Customers Growth:

Customers          #     YoY
Total Customer   3020    75%
>$100k             64    48%
>$1M               74    90%

Link to presentation:…

Stock went up 7.5% up after earnings.

Nitin (Long CFLT 15%)


Customers Growth:

Customers          #     YoY
Total Customer   3020    75%
>$100k            664    48%
>$1M               74    90%

Sorry for the typo.

1 Like

Another metric to look at is jobs for Kafka on job site have exploded.
Opening for Kafka skill have gone from 4k to about 28k.

Why this this important ? Because this indicates that enterprises are embracing Confluents vision of data in motion. Once this ball is rolling it is hard to stop.

It is becoming a “real time event” future.
An event happens in one part of the world and all systems are simultaneously updated.
Billions of events are happening every second and Confluent is benefiting.

This journey is just started.


Interesting numbers. Thanks for the breakdown.

Looks like accelerating growth which should continue next quarter. That’s not a shock given the ongoing data explosion. The potential downside I see is there doesn’t appear to be any cash flow or profitability leverage yet. I understand the want to grab as much land as possible but am mildly surprised these metrics aren’t narrowing yet given CFLT just passed a $100M quarter.

non-GAAP Operating Income							% Revenues					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019			-$23.37	-$25.45	-$79.39		2019			-60.7%	-52.9%	-53.0%
2020	-$26.86	-$20.42	-$19.73	-$22.31	-$89.31		2020	-52.8%	-38.1%	-32.1%	-31.7%	-37.8%
2021	-$31.53	-$36.80	-$42.63	-$57.00	-$168.00	2021	-40.9%	-41.7%	-41.6%	-47.5%	-44.4%
2022							2022					
2023							2023					
non-GAAP Net Income							% Revenues					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019			-$23.03	-$24.64			2019			-59.8%	-51.2%	0.0%
2020	-$27.10	-$19.66	-$18.86	-$21.98			2020	-53.2%	-36.7%	-30.7%	-31.2%	0.0%
2021	-$31.58	-$37.21	-$43.56				2021	-41.0%	-42.1%	-42.5%	0.0%	0.0%
2022							2022					
2023							2023					
Operating Cash Flow							% Revenues					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019							2019					
2020	-$31.03	-$21.46	-$9.18				2020	-61.0%	-40.0%	-14.9%	0.0%	0.0%
2021	-$19.99	-$43.17	-$18.03				2021	-26.0%	-48.9%	-17.6%	0.0%	0.0%
2022							2022					
2023							2023					
non-GAAP Free Cash Flow							% Revenues					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019							2019					
2020	-$32.37	-$22.70	-$10.25				2020	-63.6%	-42.4%	-16.7%	0.0%	0.0%
2021	-$21.23	-$45.41	-$20.64				2021	-27.6%	-51.4%	-20.1%	0.0%	0.0%
2022							2022					
2023							2023					

Also, does anyone have an insight into what looks like a customer growth slowdown this Q? I don’t think we have enough history yet to say if it’s a seasonal effect or not. Clearly, the market’s not concerned with it. Just thought I’d ask those who might know CFLT better.

Customers							% YoY							% QoQ					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019				820			2019							2019					
2020		1,390	1,730	2,100			2020				156.1%			2020	-100.0%		24.5%	21.4%	
2021	2,540	2,830	3,020				2021		103.6%	74.6%	-100.0%			2021	21.0%	11.4%	6.7%	-100.0%	
2022							2022	-100.0%	-100.0%	-100.0%				2022					
2023							2023							2023					
Customers >$100K							% YoY							% QoQ					
	Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR			Q1	Q2	Q3	Q4	YR
2019				337			2019							2019					
2020	374	408	449	513			2020				52.2%			2020	11.0%	9.1%	10.0%	14.3%	
2021	561	617	664				2021	50.0%	51.2%	47.9%	-100.0%			2021	9.4%	10.0%	7.6%	-100.0%	
2022							2022	-100.0%	-100.0%	-100.0%				2022					
2023							2023							2023					


@Stocknovice, IMO the growth rate from Confluent Cloud is one of the key things the market focused which accelerated from 135% a few quarters ago to 200% growth last quarter to 245% growth this quarter. Now 26% of total revenue.

This portion will soon become dominant and we may see quite a few quarters down the road with revenue acceleration.



@ stocknovice
Customers growth rate for all accounts start with insane growth rate e.g. 70% sequentially QoQ in Q2, 2020. I don’t expect that kind of growth is sustainable.

But if we look at customer growth rate for accounts > $100k, the QoQ growth was fluctuating around 10% QoQ. This growth metric looks fine to me. Cloudflare is growing large account at a rate of 15% QoQ so they are pretty close.

In terms of loss.
The absolute cashflow dollar from operating loss increases every quarter. It seems scary at first.

But if we look at the loss as a percentage of total revenue, it’s decreasing. So the growth is sustainable in the long term.

Year	       Cash flows from op	Total Revenue    Op loss/Revenue
TTM	                -92,725	           297,187		-31.20%
2020-12-31	        -82,057	            236,577		-34.69%
2019-12-31	        -68,834              149,805		-45.95%

I think this loss trend is similar to MongoDB:

Year	       Total Revenue	       Cash flows from op		Op loss/Revenue
TTM	        702,165	                  -36,306		          -5.17%
2021-01-31	590,380           	-42,673		                  -7.23%
2020-01-31	421,720	                 -29,540		         -7.00%
2019-01-31	267,016	                -41,989		                -15.73%
2018-01-31	154,519	                -44,881		                -29.05%


Thanks. I get the percentages, but the raw adds are even more concerning. For example, here are some sequential customer adds for the most recent quarter:

CRWD Q1 → Q2

2020: 969
2021: 1660

DDOG Q2 → Q3

2020: 1000
2021: 1100

2020: 98
2021: 230 !!!

NET Q2 → Q3

2020: 4790
2021: 5655

2020: 99
2021: 192

SNOW Q1 → Q2 (*even here I’m wondering if this is enough to maintain SNOW’s growth)

2020: 397
2021: 458

2020: 8
2021: 12

CFLT Q2 → Q3

2020: 340
2021: 190

2020: 41
2021: 47

2020: 6
2021: 4

CRWD/DDOG/NET/SNOW are not only increasing raw adds but showing a ton of operating leverage as they do. CFLT shows neither, and it’s certainly not giving off a viral adoption vibe.

As good as this board is at identifying accelerating growth, I believe it is even better at piecing together the supporting evidence that suggests acceleration can hold or even increase. While Confluent’s RPO and NRR are showing decent strength, I’m not quite seeing the customer growth to match. I’m especially surprised at the lack of total customer growth.

As for the MDB example up thread, Atlas was indeed great off a small base and as a small percentage of total revenue until it hit a wall because it simply couldn’t carry the load anymore. If CFLT’s growing cloud portion is going to be for real, it will need a lot more customers joining the party than it has now.

ethan1234 mentioned to me off board that he understood Kafka is a go-to for big projects, but there’s much more competition for small and medium projects (I hope I’m paraphrasing that correctly). Will that limit Confluent at all? Will it be able to land the little guy and then grow along with him? That’s what the best of the best platforms seem to be able to do.

I like a lot of what CFLT is doing on the top line. However, I can’t help but notice it is spending A LOT of money for what looks like middling customer growth. I really hate to pick this scab, but this trend reminds me more of FSLY than any of the four non-CFLT examples listed above. For the old schoolers, what was that other company that relied so heavily on a small customer base? Was it Pivotal? Them too if I’ve got it right.

Can someone please explain what I’m missing? I have a lot of LSPD cash to reallocate (all of it actually), and CFLT is a legitimate candidate. However, I have no real interest in throwing good money after bad.

Thanks, and I hope everyone has a good weekend.


There is an interesting tidbit about seasonality in general on the call:…

“From a seasonality standpoint, we expect Q1 '22 sequential revenue growth to be roughly flat.”

“…I think most people on the line are used to hearing that from the likes of Workday and Salesforce, much larger organizations. So I’m curious why you would be seeing it and whether the fact that you’re hiring a lot of sales reps today and maybe they’re hitting full productivity in the second half might be contributing to that dynamic? Or maybe it’s the fact that you’re signing a lot more multiyear deals and that sort of pushes you into a fourth quarter cadence. Anyway, I’d love to hear a little bit more color on that.”

“…depending on the deals structure and the deals that we have in the pipeline, Q4 tends to be a seasonally strong quarter for us, and that’s applicable to both Confluent Platform and Confluent Cloud. And when you look at other companies in the enterprise space, even with a similar model of ours, like a MongoDB as an example, they see similar dynamics between a Q4 to Q1 bridge.”

I don’t have comparable experience but I also did not like the numbers enough.

Muji spoke at some length and depth, of course–thanks Muji!–and that talk did not make me interested in CFLT either.

My takeaway was that CFLT is an enhanced, full-service, turn-key version of Kafka. I think the parallel was DDOG vs ESTC or SNOW vs Databricks.

He has a deep dive on his website and an audio file.

“…Confluent has a very similar structure to MongoDB and Elastic, being a company that supports the core open-source Kafka platform with enterprise enhancements & support, as well as having a SaaS service for managed hosting of their enterprise platform across the major cloud providers.”

And so it should in principle appeal precisely to organizations that lack the teams required to get the job done on their own.


Based investment thesis on customer number alone is not going to see the big picture.

How can you compare FSLY to CFLT? FSLY was never in fast growth mode. FSLY was growing at 40% per year and the stock was being bid up based on HOPE. I didn’t understand the board’s enthusiasm for FSLY back then. I never bought FSLY. FSLY revenue growth rate never exeeded 50% per year but at the peak the stock was pricing as if it’s growing at 100% per year. and FSLY is NOT Cloudflare. BTW; Cloudflare current situation is different than FSLY back then. Currently, I have 14% position in Cloudflare.

CFLT’s revenue growth was around 60% YoY in 2020.
and recent revenue growth rate is accelerating despite slowing customer add.

Quarter	Revenu(M)	QoQ sequential change(%)
Q1-2020	$50.90	
Q2-2020	$53.90	5.89%
Q3-2020	$61.50	14.10%
Q4-2020	$70.30	14.31%
Q1-2021	$77.03	9.57%
Q2-2021	$88.00	14.24%
Q3-2021	$103.00	17.05%

Q3-2021 sequential growth is 17% or 87% annualized! No, Confluent is not Fastly.

Why is that?
The money is in the cloud.

Large customer number growth rate from last two quarters: 9.98% and 7.62% respectively sequentially.
Cloud revenue growth rate from last two quarters: 43.88% and 35% sequentially. I see there’s a 4.5 multiplier affect here. That means for every 10% increase of large customer, the cloud revenue increases 45%!

The cloud revenue is growing faster than customer adds…
I think there’s some kind of usage fee in the cloud revenue. And this usage is explosive. That’s how Amazon makes loads of money from AWS.

Another example customer count is not the thesis for investment is BILL customer number was growing at around 26% to 30% per year but its transaction fee revenue was growing at hyper growth mode organically because payment volume is growing at much faster rate. and how did stock performed when merely 30% per year customer number growth? It’s up 700% in 2 years or 200% per year compounded return. This performance beats CRWD and DDOG despite the latter two add customer faster.


I think you’re making a good point here Stocknovice. I dove into this and when I look at the investor presentation on slide 41 they remove all the quarterly customer numbers. They only provide YoY.…

I was trying to track down the customer counts by quarter and they make that very difficult to find. Is this intentional? No idea.

The over revenue growth looks good and the expanding cloud revenue also is nice, but the slow down in overall customer adds would seem to indicate slower growth may be on the horizon?

I don’t think anything is certain at this point. It’s a small company and growth could accelerate. I would just watch this closely.


Hello All,

I have also been tracking CFLT.
The third and final lock-up period for insiders is this Monday.

With some luck we see a prime buying opportunity with insiders selling at these new price levels.

Good luck!
Just took a entry position in PLTR.


Based investment thesis on customer number alone is not going to see the big picture.

Thanks for the tip. I’ll try to keep it in mind. In the meantime, let me try to expand a bit…

How can you compare FSLY to CFLT?

Quite simply, actually. FSLY and Pivotal (or whatever) are examples of firms we’ve discussed that struggled to maintain a short a run of acceleration because it was driven mostly by an increase in current customer spend without adding enough new customers to keep it going. The comparison has nothing to do with the business being cloud or non-cloud. It’s about the levers that drive future revenue growth.

You see, one of things I’ve learned here is the most successful companies tend to nail both the “land” and “expand” part of the business model. I’d consider net retention rate a pretty simple proxy for the expand part. Well, how’s that look?

CRWD: 125% at the end of last year and its usual 120%+ so far this year.
DDOG: 130%+ for all eternity.
NET: 119% at the end of last year and 123%, 124% and 124%.
SNOW: 168%, 168%, 169% (holy crap).

CFLT: FY20 134%, FY21 125%, 1Q22 117%, this Q back to 130%+, a stated baseline on the call of 120%+ the next few quarters, and a long term goal of 130%+.

Good for CFLT on reversing the trend. That looks like a lot of those other companies and could mean the start of big things. Sign me up for that.

Of course, that must be balanced by the land part. The simplest proxy for that (in my opinion, of course) is customer growth. Let’s take a peek.

Large customer number growth rate from last two quarters: 9.98% and 7.62% respectively sequentially.

Hmmm. That sounds OK, doesn’t it? Well…

DDOG: 11.7% and 14.6% at a larger scale.
NET: 15.1% and 15.8% at a larger scale.
SNOW: 9.5% and 10.1% at a larger scale (using total customers since we don’t have $50K or $100K).

Do you notice how those other companies’ rates are increasing while CFLT’s are decreasing in the same general business environment? And Confluent’s total customers trended the same way at 11.4% and 6.7%, meaning a lower number of customers that might one day grow into the >$100K group. Two quarters doesn’t necessarily tell the whole story though, which is why I’m asking for help in understanding this dynamic from others who might know better. Should I be concerned about this or not? I can’t help but notice CFLT isn’t showing the same recent strength as those others. Why is that and will it be a problem going forward?

The cloud revenue is growing faster than customer adds. I think there’s some kind of usage fee in the cloud revenue. And this usage is explosive.

And this is partially my point. Explosive usage is great and exactly what we want to see. The downside is current customers can only explode so much before budgets tap out. That’s a disadvantage of usage-based models vs subscription-based. Of course, that’s not nearly as big a concern if the company keeps adding enough new customers to explode on the front end. You know whose business model benefitted from a usage explosion but couldn’t add enough customers to keep it going? Fastly. (And please let’s not spin to FSLY as a company. This is only about the customer/usage/revenue relationship. I wish I had a better example, but I unfortunately don’t.)

So, does CFLT’s current performance indicate a future CRWD/DDOG/NET/SNOW or is it a FSLY/PVTL head fake? I’m not saying which one because I don’t know yet. But I do know there’s a lot of smart people here, so I figured I’d ask.

Don’t get me wrong. I hope everyone who holds CFLT makes obscene amounts of money on it. In fact, it looks like many are already doing quite well. If I can determine CFLT’s business trends are good enough to augment my other ideas, I might join the party. Can anyone send an invite that addresses the above?


I was a cloud-architect who designed and developed cloud native apps during 2015-2020. Data-in-motion is a critical part of cloud architecture. (I explained it here a while back:… ).

I have no doubt about the importance of the product, and how it’s usage will keep on growing.

As far as ‘customer growth’ is concerned, can “hiring freeze during Covid” be a satisfactory reason?
CFO during Q2 call: As a reminder, we hit the pause button on hiring at the front end of the pandemic in FY '20, and we didn’t resume hiring in earnest until late Q3.

And we’ve caught up on hiring and we’re catching up on hiring. And from a – just a headcount standpoint, we’re on track relative to hiring the sales talent that we need. In 2021, we have more ramping reps than ramped reps. And in '22, given the fact that we have a 12-month onboarding process and ramp process.

CFO during Q3 call: As a reminder, our profitability in FY '21 is being impacted by our plan to catch up on hiring given the pause we prudently took in FY '20.


Very informative post. I’m a big fan of your monthly portfolio review. Just want to point out a mistake in Datadog’s numbers.

DDOG: 11.7% and 14.6% at a larger scale.

Datadog’s $100K+ ARR customer numbers from Q3/Q2/Q1 2021 are: 1,800/1,610/1,437. The sequential growth rates from last two quarters should be: 12% and 11.8% respectively, roughly flat.

Thanks, simpleisbetter. I appreciate the compliment.

Datadog’s $100K+ ARR customer numbers from Q3/Q2/Q1 2021 are: 1,800/1,610/1,437. The sequential growth rates from last two quarters should be: 12% and 11.8% respectively, roughly flat.

I believe you might be looking at old numbers. Datadog restated the $100K+ ARR customers in its October 27 Dash Investor Day after updating their definition of MRR and therefore ARR. The new $100K figures can be found on slide 101 here:…. The footnote at the bottom of the slide explains the change.

Using this new info I have the last three customer totals as 1,800/1,570/1,406. That would make the QoQ rates going backward the 14.6% and 11.7% I originally stated. Please let me know if you (or anyone else for that matter) interpret it differently.


Please delete the one above. Not sure how I submitted that one! (oh, I see, tab gets you to ‘submit message’!)

I ran a little comparison MNDY vs AMPL vs CLFT vs BILL. You may want to double-check the numbers for yourself before taking what follows as anything other than food for thought.

REVENUE last quarter YoY per last reported Q:

1 MNDY 96%
2 BILL 164% (78% organic)
3 CFLT 67%
4 AMPL 66%

REVENUE by base, smallest preferred
1 AMPL 39
2 MNDY 71
3 CFLT 103 (Q3)
4 BILL 116 (Q3)

REVENUE by sequential growth:
1 BILL 49% but acquisitions
2 MNDY 21%
3 AMPL 18%
4 CFLT 17%

REVENUE guidance, sequentially
1 AMPL 12%
1 BILL 12%
3 MNDY 6%
3 CFLT 6%

1 MNDY 87%
2 BILL 74%
3 AMPL 71%
4 CFLT 64%

1 BILL -9%
2 AMPL -10%
3 MNDY -14%
4 CLFT -41%

FCF margin
1 MNDY -3.5%
2 AMPL -10% (first half 2021)
3 CFLT -20%
4 BILL -22%

1 CFLT 1B to 33 mln
1 MNDY 875M to nothing
3 AMPL 300M to nothing
4 BILL 1.3B to 1.1B

1 AMPL 8.5
2 MNDY 15
3 CFLT 25
4 BILL 34

Now, summing up all rankings with the lowest score being best:

1 MNDY 16
2 AMPL 20
3 BILL 23
4 CFLT 28

One can add customer growth and other metrics.


Building on StockNovice’s good line of questioning around customer growth.

I guess the question I have around all this is…will this Product work for smaller customers, too? Perhaps that will answer the question we have around lower customer growth, too. Yes, hiring sales can help, no doubt. But does this product work for smaller customers, too? Near as I can tell, most of their customers are big dog customers. Guessing it can work for smaller customers, but why hasn’t it, to date?

3000 customers at $100M of quarterly revenue. $CRWD was around the same ---- $96M and 3059 customers ---- and at that time was adding around 500-700 customers each Q. $CFLT on average these prior 5 Qs adding around 370 a Q. This Q just 190 - ouch.

SO, can they add smaller customers? Will it work for smaller customers? Will adding salespeople solve the problem (not the question, but the obvious problem)?


I would like to delve a little deeper into the Confluent Cloud success.

The way it seems to me is that they have two very different products. One is a roaring success (Confluent Cloud) and just getting started, and the other a bigger, bit older success (Confluent Platform). Given how quickly Cloud is growing, it is likely to overtake Platform soon.

However we don’t know how much of the Cloud growth is cannibalising the Platform revenue, we don’t know the NRR of different size customers or of Platform only vs Cloud only vs Platform & Cloud; and we don’t have a good split between the two types of customers. That’s a lot of unknowns - which Stocknovice rightly points out.

However whereas it is possible that the Cloud offering could be a Fastly in the making, it is also very possible that it is a Snowflake in the making. Snowflake has some similarities to the Confluent Cloud offering - it spans all 3 cloud providers, it is usage based, it is fully managed, and it gives customers - typically large ones - access to a fully managed next gen database. Or at least that’s my non-techie opinion. Would love to hear more tech-savvy people’s views here. And if it is a Snowflake in the making - fuelled by the Cloud offering - this could be a very good time to get in relatively early.

Anyhow, if we assume that the reason for the relatively slow customer growth can be explained by a bit of a hiring freeze and/or seasonality, then focusing on Confluent Cloud could hold the key to understanding where continued/accelerating hypergrowth could come from.

They had the following to say about Cloud (slide 39):

Key milestones:
? November 2017: General availability of Confluent Cloud
? September 2019: 50% of all customers on Confluent Cloud (includes self-serve, pay-as- you-go customers)
? August 2020: Confluent Cloud available through the marketplaces of the three leading cloud providers

→ So it’s a very recent product. It also sounds to me that they are enabling existing Platform customers on Cloud (50% on in Sept 2019) and then customers are increasingly using Cloud in stead of Platform. It also looks like Aug 2020 is important - and recent - given the availability in all 3 cloud marketplaces.

So what could Q3 2022 look like if we assume that 200%+ growth in Cloud can continue (while, granted most of it will come from existing customers)?

We know the split and the growth of the 3 components of revenue, so let’s assume the following for next year Q3 yoy and see where that gets us (Q3’21 → Q3’22e):

Platform 40% yoy → 35% yoy (est)
Cloud 245% yoy → 200% yoy (est)
Services 45% yoy → 40% yoy (est)

Applying the splits they gave us and the above estimated growth rates will give the following for the last 2 Q3’s and next year Q3e.


**$m	Q3'20	Q3'21	Q3'22e	21yoy%	22(e)yoy%**
Platf	47	66	89	40%	35%
Cloud	8	27	80	245%	200%
Serv	7	10	14	45%	40%
**Total	62	103	183	67%	78%**

So if Cloud can continue at >200% growth and the other components decelerate somewhat, Cloud will be on a par with Platform a year from now, and will contribute sufficiently to accelerate revenue growth from 67% yoy this year Q3 to 78% yoy next year Q3.

Given the strong growth in their very large (>$1m) customers of 90% yoy as well as the examples they give in the deck (slide 22) - all huge companies expanding fast but with relatively modest expenditures still - it seems fair to assume that growing much more within these customers is quite possible and feasible (again comparing to the likes of Snowflake, and not Fastly) and won’t necessarily require a deluge of new customers signing up but rather some larger customers.

So while I agree there are some clear gaps in what we know of the customer dynamics, it seems that the Cloud offering, judging by revenue growth, is a huge success, likely to remain so and also likely to become the main product in time.

Accordingly I took a starter position.



So what could Q3 2022 look like if we assume that 200%+ growth in Cloud can continue (while, granted most of it will come from existing customers)?

I don’t think we can assume that at all. I mentioned in this post (…) that I had taken a position in Confluent, but I sold a few days later. My pro-confluent thesis had centered around NRR exploding again. It drove a huge Q3 and looks to continue in Q4. After talking to some techie friends, I’m not sure the NRR explosion can continue for long, because the cohort of customers that will be able to spend larger and larger amounts on confluent may not be that big. In other words, the value prop Confluent offers just might not be enough for customers to spend as much as they do with Snowflake. Because of that, I think we need to take management’s comments seriously when they talk about a sequentially flat Q1, and a 36% guide for 2022. Obviously they’ll beat 36% handily, but I don’t think it’s reasonable to expect them to more than double it.

But maybe even more importantly, we simply shouldn’t underestimate the extent to which cloud is cannibalizing legacy revenue. So the legacy growing at 40% in your prediction might be a risky assumption as well. We can’t know, but if we look back at what Mongo had with Atlas a few years ago, we can see a peak of 78% YoY revenue growth in the quarter ending April 2019, when Atlas was 35% of Mongo’s revenue and Atlas grew 340% YoY. The very next quarter Atlas grew 240% YoY (to 37% of overall revenue) but Mongo’s total revenue only grew 67% YoY because legacy revenue grew less than 30% YoY. The next quarter legacy grew less than 20% YoY and Atlas growth rate fell to 185%. Mongo was more of a 50% grower after that, I believe (I stopped following it). It’s just far too easy to see the same pattern potentially happening for Confluent within the next few quarters.

Lastly, the CFO specifically called out their “hybrid model” when asked about the flat Q1 guide – essentially he too was calling out the cannibalization of legacy revenue. I think this, plus the customer slow down StockNovice identified, is enough to make me prefer watching from the sidelines for now.



we don’t know how much of the Cloud growth is cannibalising the Platform revenue

I had thought about this question a few weeks ago and gave some analysis.

Let’s take a look:
Legacy platform revenue from past three quarters: 63m → 68m → 76m. QoQ changes: 7.71% → 11.76%
Cloud Revenue from past three quarters: 13.9m → 20m → 27m, QoQ changes: 43.88% → 35%.

We can see both types of revenue grew nicely. In fact, Legacy revenue accelerated while cloud revenue decelerated a little bit. To say Cloud revenue is cannibalising legacy revenue, legacy revenue has to decrease. This is not the case here. My hope is Cloud revenue growth can remain as hyper growth or at least stay with a baseline growth of at least 10% to 20% QoQ.

To answer another concern from others about customer growth slowdown:

For the recent quarter decline of growth of customer with 100k+ ARR to 7.6% QoQ. I think it’s part of seasonality like MongoDB. I don’t think it’s a secular slowdown. Here’s the QoQ growth of customer with 100k+ ARR from the past 7 quarters(most recent to oldest) : 7.62%,9.98%,9.36%,14.25%,10.05%,9.09%,10.98%. and also note that the average QoQ growth was just 10.2% for CFLT so the last quarter of 7.6% was not that far away from 10%. If it drops to 1% to 5% QoQ then I’ll start to concern. But data will only continue to grow so even if that to happen, I’ll continue to hold.

Then let’s look at MongoDB’s growth of customer with 100k+ ARR(most recent to oldest),it went back and forth from 6% to 12%, then to 8% QoQ: 8.21%,8.06%,9.73%,11.88%,9.78%,8.24%,6.92%,6.00%,5.63%,5.97%.

Conclusion: Kafka was created at LinkedIn for a reason. They needed it. I am sure many other companies felt the same. Data in motion is a new category to: data analytics, database. It’s here to say and data will only grow in the future.

Long CFLG 8%. MDB 9%.