The 50% YoY Revenue Rule

Over the past few years, tech stocks have ruled the world with amazing returns and many here, including myself, have tried to develop criteria or patterns that might be applied to other potential tech candidates.

We have been through many many permutations from valuation considerations (P/S, etc.), TAM/SAM considerations (Tinker Rules), Revenue acceleration/deceleration rates (Saul’s criteria), competitive advantage rules (Michael Porter), Founders rules (Rule Breaker), Industry insider technical folks, Technical analysis, etc.

Many (perhaps most) of these stocks have been rule breakers from all/most previously accepted rules of norms for stock investments…these previous “well accepted” criteria were essentially irrelevant.

I have tracked many a portfolio criteria from buy and hold, the valuation portfolios (published here by Saul), etc.

There appears to have been ONE main criteria for investment success these past few years…all others essentially largely irrelevant to the stock performance. One reason to consider patterns is that they can save enormous time analyzing stocks and avoid distractions of largely irrelevant information that might cause an investor to buy or sell when the data element was largely poorly predictive of stock performance.

Let’s be honest that most trades today are performed by quants and algorithms by which massive data has been entered into trading programs that trigger buy/sell decisions. These sophisticated programs are gaining more popularity such that many previous investment gurus like Mauldin, etc. have suggested that software based trading should be part of anyone’s portfolio.

But unless we have some sophisticated quants on this board, we are beholden to using somewhat archaic criteria for buy/sell decisions…archaic criteria which actually appear to have been quite superior to the general stock market indices unless under the market conditions/macroeconomic conditions we have chafed these past few years. Saul has been his own quant and like most of you, I have followed his buys and sells which quite clearly emphasize revenue growth.

So what was the one criteria that seemed to rule the world for decisions? I see many of you post your portfolio results which have been pretty solid for sure and without a doubt the fact that you have generated these returns certainly suggests an element of courage to have invested in these stock types.

It turns out that the single criteria that predicted stock returns was…highest YoY revenue growth rate. Margins???..nope, the SaaS argument about margins didn’t change the fact that ROKU with its lower margins still had the highest return YTD at 190%.

So if we essentially bought the highest revenue growth companies at beginning of year in equal amounts what would be the returns:

ROKU 190%
SHOP 104%
ZS 89%
SMAR 76%
TTD 74%
OKTA 70%
MDB 69%
AYX 50%
TWLO 49%
ESTC 17%

Portfolio return YTD was 79%!

How does that compare with your portfolio? This was a mindless portfolio…simply stocks whose YoY revenue growth rates were > 50%…no other criteria. No concerns about revenue deceleration, margins, TAM/SAM, Founders, etc.

Now thinking back over the past few years, I have seen Saul jump off stocks with decelerating revenue growth (SHOP being the most recent in memory) and he has been very facile to anything that might affect the revenue growth rates. IMO, he has essentially stress revenue or all else.

But what I am suggesting here is that this mindless approach would have yielded portfolio returns that exceeded most of your individual returns with far less time to manage and therefore far more time to enjoy your life outside of investments…little to no analysis but for revenue growth.

We have also argued that these Uber high P/S stocks could only be justified because they were largely SaaS with high gross margins and therefore they could ultimately turn over to profitability with profit margins of up to 40%. But perhaps the somewhat unexpected finding this year is that the two highest return stocks in this high revenue growth index, had some of the worst gross margins (ROKU and SHOP)…at least as compared to an AYX for example.

So what this data suggests to me, is that revenue growth rate reigns supreme OVER ALL other criteria. And then the natural corollary is that additional parsing of data was unnecessary.

Hope this analysis was somewhat provocative and of interest to you, your portfolio and the time allotment you spend maintaining it.

Best:
Duma

124 Likes

Hope this analysis was somewhat provocative and of interest to you, your portfolio and the time allotment you spend maintaining it.

Great analysis! It shows that the “S” curve does track revenue growth but in addition it tells you where to take profits, when revenue growth decelerates. It will happen when the market is getting saturated - no one left to sell to, but it will also happen if the company missteps or if a disruptor appears on the scene. If you don’t have that data then just a reduction in growth rate should be enough, if not to sell, to put the stock on an alert list. You can always buy it back if the signal proves to be a false deceleration - which is what Saul does.

Denny Schlesinger

7 Likes

So what this data suggests to me, is that revenue growth rate reigns supreme OVER ALL other criteria.


If it is that simple, invest in nothing but pot stocks growing 400% y/y. Or stocks like HUYU and other china ADRs.

Or companies that have massive growth rates due mainly to acquisitions.

Or stocks that have underperformed miserably and are now crushing y/y compares. (This will likely be nvda over the next year, as an example).

Finally…i think statistics need a larger sample size than 5 months.

What would your results be for the 5 months of Aug 2018 thru end of Dec 2018?

Blindly selecting stocks based solely on rev growth rate in a vacuum is probably not a smart idea.

Dreamer

62 Likes

To Dreamer’s point, these were the stocks discussed on these forums with the highest growth rates. Does this thesis hold up if you blindly chose stocks with the highest growth rates over the past say, year or two, across the entire market, regardless of industry?

Blindly selecting stocks based solely on rev growth rate in a vacuum is probably not a smart idea.

Dreamer

This was clearly stated as “tech” stocks…not pot stocks…and most certainly NOT your China stocks.

NVDA would have been sold at any revenue less than 50% YoY growth which would have been Oct 2017 at the then price of $205.

And yes, the results hold for trailing 12 months as well. I have not back tested this prior to that time frame.

It’s OK if you doubt it, even if it bests your portfolio results…could be coincidence though I doubt it.

Please feel free to back test this to prove it is not a smart idea.

12 Likes

There is a simple mathematical reason for this. 5 years of 50% growth is 7.59x, 5 years of 40% growth is 5.37x, 5 years of 30% growth is 3.71x, and 5 years of 20% growth is 2.48x.

There are massive and material differences in the size a company becomes based upon not that great difference of growth rates. And remember, thereafter, the faster growing companies tend to still grow faster than the slower ones, and they will be doing so from a larger growth base to begin with.

The math there does not lie. Actually obtaining the pro forma numbers is the question. This uncertainty of obtaining the pro forma numbers perpetually undervalues the higher growth companies as they actually meet the numbers (even exceed them) until such time as they do not. This is of course combined with CAP and relative cash flow printing capabilities.

All of this perceived of course. I do not actually run the numbers, it is just what Saul and I have intuitively known. Also happens that many of the companies that can achieve this are disruptive, in huge TAMs, and like so many disruptive companies have little real competition with huge first mover advantages creating large CAPS.

Lets put that one in the textbook - at some point they mature but lets not thrown out the Tom Brady before his time has really come.

Tinker

23 Likes

Duma,

What criteria are you using?

For example, ROKU is on the list when they grew at 39% and 45% the 2 quarters before the start of the year. They were not a 50% grower.

I agree that growth rate is the most important factor. I think its because the stocks you mentioned are beating expectations.

If you look at analyst estimates, they continue to forecast slowing growth for these companies. Then the companies report flat or accelerating growth, beating expectations, causing analyst revisions upward, and the stocks rise with higher expectations.

We need to watch out when the expectations get to high. Higher risk of disappointment.

Jim

2 Likes

I don’t think Duma said this is the ONLY way, but that it is A way, to filter the universe of tech stocks. A shortcut for those that are time challenged. Or, as I have learned, a significant factor among several that suggest future gains (or not).

He also didn’t say he was applying this theory (yet) to the universe of all stocks.

I believe Denny is right when he says that this is an illustration of the S-curve in action.

Also, as Saul says, it’s different this time (until it’s not, and we all know there will come a time when it’s not).

2 Likes

It used to be that “Growth Investing” was a combination of revenue/earnings growth and the PEG ratio was used to determine how fairly valued the stock was compared to its growth rate. The internet tech bubble back in 1999 changed that because the companies didn’t have any earnings. So everyone relied upon how fast the revenue was growing only. And worried about profit only when the company matured. So that’s pretty much how I judged growth companies since 1999. Concentrate on revenue growth, earnings should sort itself out.

But to me there’s still a rough estimate of how one could value, and that is basing on what it’s earnings power could be if the company focused on it. For example many software companies out there such as ORCL have a net profit margin of 25%. So I use that as a guideline. It means if a company is trading at 40x sales, you multiply that by 4 to get a rough estimate of what the PE would be at 25% net profit margin. If a company is trading at 40X sales, it would have a PE of roughly 160. If a company such as OKTA is growing revenues at 50% and P/S Ratio of 35, it has a rough PE ratio of 150 and a PEG ratio of 3. A PEG ratio of 1 or less is what is ideal.

So OKTA will have its work cut out for it and need to sustain a long trend of growing revenues at a high rate to grow into earnings. But the PEG ratio still has a flaw: Would you buy a stock growing at 3% a year with a PEG of 1 or a stock growing at 50% a year with a PEG of 1? One would lead you to a 3% annual return if all remains constant while the other will lead you to a 50% annual return if all remains constant. So it does make sense that very high growth companies would have a higher PEG ratio than 1.

Also, “don’t worry about valuations” is not a new concept. I have a copy of Philip Fisher’s “Common Stocks and Uncommon Profits” written in 1958 and he talked about it there. I also have a copy of “Money Masters” by John Train and he talked about the late sixties, which they called the “Go-Go Era” where “Performance Fund” managers would rack up 50% or more annual returns chasing the hottest stocks like Teleprompter, Xerox and Polaroid completely disregarding valuation. In fact, some saw high valuations as a positive because it meant high institutional sponsorship. The "Go-Go Years’ apparently ended poorly in the late 1960’s as the Performance Funds such as the Manhattan Fund managed by Gerry Tsai, the poster boy for the strategy, suffered large losses. But not all though. Since we have been through all this before, I tried researching the methodology of these performance fund managers of the 1960’s but there’s not much information on them out there. I do know that Ned Johnson, who had the Fidelity Magellan and had a 29% annualized return during his tenure, had a concentrated investment philosophy and focused on growth companies, but held Chrysler for some time which was not nearly in the same league as Polaroid and Motorola during that era. But his performance during the late sixties, while he had losses, was more than made up for by his performance in the early sixties and seventies. It seems Gerry Tsai was the one who really blew up in the late sixties, his records are more available.

10 Likes

Duma, this is just utterly wrong. As Jim points out:

For example, ROKU is on the list when they grew at 39% and 45% the 2 quarters before the start of the year. They were not a 50% grower.

Also, you need to add SQ, who had grown at 68% the last time they reported before 2019 started. They’re only up 13% YTD.

With just those two changes:

SHOP 104%
ZS 89%
SMAR 76%
TTD 74%
OKTA 70%
MDB 69%
AYX 50%
TWLO 49%
ESTC 17%
SQ 13%

So we’re down to 61% already. Not 79%. What other companies might you have left out? It doesn’t really matter, because a 5 month sample size doesn’t prove much, does it?

Sorry, but it’s just not that easy.

Bear

27 Likes

Duma, this is just utterly wrong. As Jim points out:

So we’re down to 61% already. Not 79%. What other companies might you have left out? It doesn’t really matter, because a 5 month sample size doesn’t prove much, does it?

Do you disagree that revenue growth is a major driver in stock returns? I’m not sure what you’re disagreeing with? It seems to have been going this way for years? How else did ILMN do so well over the years other than the company itself growing over time? Of course YTD is an extreme example, and we haven’t seen this many fast growers doubling their share price in a long, long time, so it’s not like one could expect 61% YTD returns on a regular basis, but this seems pretty consistent with history.

Other than when they individually get taken out to the woodshed for not meeting their growth rates.

2 Likes

So we’re down to 61% already. Not 79%. What other companies might you have left out? It doesn’t really matter, because a 5 month sample size doesn’t prove much, does it?

Sorry, but it’s just not that easy.

Thanks Bear.

61% is not that good YTD? Your portfolio beat those returns? More backtesting is in order for sure.

I meant that original post to be provocative and I do appreciate you taking the time to consider it…perhaps some revisions will be added in the future and it also doesn’t consider the actual macroeconomic conditions we have been in at the time (bull vs bear markets, etc.).

But again, this data suggests that revenue growth rate was the single most important metric for stock appreciation.

Now that does fly in the face of many who think their detailed analysis was able to parse out the winners and losers within this category…but it just doesn’t pan out that way. Even within this category of stocks, the returns did not seem to correlate with better margins.

There was also the prospective portfolio analysis I did over the preceding year with the tech companies commonly discussed here and previously at the NPI…the highest growth companies also won…happened to also have the higher P/S.

Essentially this theory goes, don’t look for hidden value (PVTL, NTNX, etc.)…just stay with high revenue grow companies…and from a time savings perspective, much less handwringing.

And again Bear, you do not strike me as a Uber dogmatic person so hopefully you will file this away in the hmm category…your call of course.

Best:
Duma

13 Likes

High revenue Growth (with high gross margins) is clearly a key factor as Tinker has demonstrated by the math - this is also covered here

https://tomtunguz.com/why-growth-matters-so-much/

I think that in the case of ROKU, the “hidden growth” is in the platform revs, (which are the truly relevant metric) regularly exceed 70% with gross margins of circa 70%

I think that you have given a reason why hidden growth shouldn’t be immediately discounted by including ROKU as an example, duma.

1 Like

I think that you have given a reason why hidden growth shouldn’t be immediately discounted by including ROKU as an example, duma.

Exactly. NTNX does not prove or disprove a strategy regarding hidden growth. Years ago I made like 150% off Blue Coat Systems in a single year (when such returns weren’t as common as now) because they launched a WAN Acceleration product line I could see was growing fast and becoming a bigger part of their number, while the rest of the company stagnated. That must have been around 2007.

Hidden Growth is for sure a good reason to buy, but not when it’s some mysterious huge percentage growth with unknown actual numbers or relation to the rest of the business, or in the case of these cloud companies, converting fixed license to subscriptions which is just offsetting one way of recognizing revenue for another, doesn’t mean the company is about to start growing revenue.

2 Likes

True enough on ROKU. It’s YoY revenue growth rates the past 4 quarters from most recent to present were:

51.32% 46.47% 38.95% 57.40%

So one might have bought in after June 2018 quarter and the question would be were you shaken out the next. The most recent 2 quarter were quite substantial of course but their growth story looked pretty solid starting June 2018.

OK, last post on this from me.

Best:
Duma

I wonder what a similar simplistic one-shot revenue growth criteria screen would have turned up in January of 1999.

Oh wait, we know… (from old MI screen posts)
DELL… VISX… BBY… EMC… INTU… ORCL…AOL… YHOO…

Someone from January 15th, 1999:
As I speak, I am plotting and rounding up more funds just to dump into Internet favorites AOL and YHOO. Unless they leave the screens, my resolve will be to sink with them - if that is the direction they go. For if they sink, I will assume that the has market crashed and the economy collapsed. After all, there is a certain honor in being loyal to a screened stock. Isn’t that what being a “fool” is all about?

I know this time is different and these new companies are killing it, and their stocks are all fine on the way up. But you might want to have some 10% stops in place or leave some cash lying around in case they miss revenue growth estimates or get substantially below 50%. At these valuations they’re not set it and forget it holds. If they do miss, the stocks will get crushed just as quickly as they went up, hopefully not as badly as their ancestors DELL, VISX, YHOO, AOL, QCOM, JDSU and others 20 years ago. We made a lot of money on them on the way up (I was up @300% in 1999 alone)… but without some mechanical sell rule in place you can pull the trigger on it’s all at risk. (down 50% in 2000-02).

4 Likes

What stood out for me was this quite frankly extraordinary statement…

“But what I am suggesting here is that this mindless approach would have yielded portfolio returns that exceeded most of your individual returns with far less time to manage and therefore far more time to enjoy your life outside of investments…little to no analysis but for revenue growth. Duma.

Now, unless I am reading this wrongly and if so, apologies, but I will take this “mindless approach” all day long.

1 Like

Now, unless I am reading this wrongly and if so, apologies, but I will take this “mindless approach” all day long.

For me, the problem with the “less time/mindless” approach is that when things happen to the stock price (sudden drop or rise), you don’t know what to make of it, and therefore what action, if any, to take (add, hold, sell). Saul’s approach necessitates keeping relatively close tabs on your companies (more than MF LTBH methods).

3 Likes

And again Bear, you do not strike me as a Uber dogmatic person so hopefully you will file this away in the hmm category…your call of course.

Thanks, Duma. I don’t feel that I am dogmatic. I’m not saying that my way is best. I’m simply saying that we can’t employ a strategy like you’ve outlined. Why? Because as Saul said earlier, things change. Just because this worked in the past 5 months, or even the past 5 years, doesn’t really mean much. To be honest, it’s completely mechanical and arbitrary to say 50% as you did.

If we said companies with revenue growth of at least 60%, we’d be left with only:

SMAR 76%
TWLO 49%
ESTC 17%
SQ 13%

And our average gain would be only 39% My YTD gains actually are better than that.

Or if we said companies with revenue growth of at least 50% but not over 60% we are left with just:

SHOP 104%
ZS 89%
SMAR 76%
TTD 74%
OKTA 70%
MDB 69%
AYX 50%

And our average gain is now up to 76%! So do you think we should look only for growth rates in the 50’s? Of course not.

Obviously, all things equal, more growth is better. But all things aren’t equal. The fact that the 60%+ companies under-performed the 50%+ companies shows that our project is simply more complicated than screening for the highest growth rates.

That’s all I’m saying.

Bear

30 Likes

Obviously, all things equal, more growth is better. But all things aren’t equal. The fact that the 60%+ companies under-performed the 50%+ companies shows that our project is simply more complicated than screening for the highest growth rates.

I’m not sure why this would even be a revelation, let alone an argument. All else being equal, the faster growing company is going to have the highest share price appreciation. The sample size gets too small as you get less and less companies. ESTC has been a low performer as a new IPO with lockup overhang and doubling on the IPO date. SQ I don’t follow enough. that’s 2 of the 4

Here are the solid 25% up but less than 50% growers. Not cherry picked by any means.

NOW: 48%
CRM: 13%
HUBS: 40%
ZEN: 45%
COUP: 77%
VEEV: 60%

Average of 47%.

I don’t think this is a big shocker to anyone. It’s why nobody sees the CRMs or ServiceNows of the world as the firecrackers that SMAR or ZS are, because they are growing slower.

The only problem is the higher price of the faster growers and the blow up one may encounter due to a disappointing quarter… which, seems over the past year they haven’t, we’ve only been left with examples such as TLND or NTNX. NTNX software revenue was good. But it was still relatively cheap and not a high flyer like ZS or OKTA.

1 Like