How to Predict Sales Growth
There are a number of reasons why you might be interested in the future sales/revenue growth of a company. For instance, in order to perform a discounted-cash-flow analysis of a stock, it’s one of the essential inputs. Or perhaps you believe that the higher a company’s projected sales growth, the more attractive an investment it is, regardless of the price. Or perhaps you’re like me: it’s just one of many factors you take into account when you’re deciding whether or not to buy a stock.
If you’re an analyst, future sales growth is a prime concern, and your job is to come up with reasonable guesses through dogged investigation and thorough research. But for those of us who aren’t analysts, what numbers should we be looking at? This article attempts to solve this problem.
Here is my approach (you can skip this paragraph if you’re not interested in methodology): I took a likely collection of about 35 different factors and downloaded, via Portfolio123, the value of those factors for about 5,000 stocks eighteen months ago. I then looked up how much the sales of each of those stocks has grown over the past year (there was a six-month interval between the earlier data and the growth period to minimize look-ahead bias). I eliminated stocks for which I had no sales data as well as stocks that ranked in the top and bottom 5% in terms of sales growth numbers. I then repeated this experiment going back five years and ten years. Finally I measured the correlation between each factor and the stock’s future sales growth. I came up with seven factors whose correlation was consistently high in all time periods and was significantly higher than a random group of numbers might have been.
So in order to forecast future sales growth, I would take the following into consideration, in order of the highest correlation:
- Price momentum. The 52-week price change is the best indicator available of how much a stock’s sales will grow. Why? Because investors put their money into stocks that they think will grow, and investors are, on the whole, smart—very smart indeed.
- Past and estimated sales growth. Take the average of the following three numbers: the growth of the most recent annual sales over the sales three years ago, annualized; the growth of the current fiscal years’ sales estimate over the sales two years ago, annualized; and the growth of the next fiscal years’ sales estimate over the sales one year ago, annualized. This formula allows you to approximate sales growth pretty well. If analyst numbers are not available, a simple three-year sales CAGR figure works well too. It’s important, however, to look at three-year growth rates, not one-year growth rates. Trailing twelve-month sales growth does not appear to correlate with future sales growth.
- Average recommendation. This is the mean analyst recommendation. Stocks that are rated strong buys are more likely to increase in sales than stocks rated strong sells.
- Asset growth. Not surprisingly, there does seem to be a causal relationship between asset growth and sales growth. I used the average assets over the trailing twelve months compared to the average assets over the previous twelve months.
- Net operating assets. I’ve written an article on this measurement. A company with a low proportion of NOA to total assets has a lot of cash and non-debt liabilities to invest in increasing sales.
- Sales trend. This is the correlation coefficient between the sales of the last seven years and the years themselves. If sales over the last seven years, from earliest to latest, are $1B, $2B, $3B, $4B, $5B, $6B, and $7B then the correlation coefficient will be 1; if sales are $7B, $6B, $5B, $4B, $3B, $2B, and $1B then the correlation coefficient will be –1. Anything less smooth will give you middling figures.
- R and D expenditures. Take these and divide by total assets. (Companies in certain industries are N/A and are ranked in the middle.) R&D investments don’t always pay off in sales growth, but sometimes they do.
Using Portfolio123, I created a ranking system that includes formulas for all these factors. I weighted the factors according to the correlation they had with future sales growth minus a fraction of the correlation they had with the other factors.
Out of the universe of stocks from which I choose my investments (this includes everything from nanocaps to megacaps), the top stocks according to this ranking system are China Online Education (COE), Cryoport (CYRX), Cue Biopharma (CUE), DarioHealth (DRIO), eXp World Holdings (EXPI), Hyrecar (HYRE), TG Therapeutics (TGTX), Vapotherm (VAPO), and Vaxart (VXRT). I would venture that these stocks have a pretty high chance of strong sales growth over the next year.
The stocks with the worst rankings—First Eagle Alternative Capital (FCRD), Forum Energy Technologies (FET), Invesco Mortgage Capital (IVR), Martin Midstream (MMLP), Oxford Square Capital (OXSQ), and Pennsylvania REIT (PEI)—are, with one exception, MLPs, BDCs, and REITs. This makes some sense since these types of companies are required to distribute the large majority of their earnings to shareholders and therefore cannot reinvest profits into the business, stymieing future growth. These may be excellent investments, but you wouldn’t look for high revenue growth in this space.
If we look only at S&P 500 stocks, those with the highest projected sales growth are Amazon (AMZN), DexCom (DXCM), NVIDIA (NVDA), salesforce (CRM), ServiceNow (NOW), and Vertex Pharmaceuticals (VRTX). The stocks with the worst projected sales growth are Apache (APA), Coty (COTY), Marathon Oil (MRO), Noble Energy (NBL), and SL Green Realty (SLG).
Of course, this ranking system tells me nothing about whether these stocks are worth investing in. That depends in good measure on a lot of other things, including their price.
Based on past rankings, I’ve done some regression work, and have come up with a good formula for approximating future sales growth. Take the rank of the stock (between 0 and 100) according to the ranking system described above and divide that by 100. Call it x. The projected sales growth will then be approximately 95x3 – 115x2 + 50x – 7. This will give you a range of growth between –7% and +23%.
So, given this, let’s look closely at the projected sales growth of fifteen stocks. I’m going to choose the five stocks I’m most heavily invested in, the five most heavily traded stocks right now, and the five stocks with the biggest short interest as a percentage of their shares outstanding (excluding nanocaps). I’ll discuss them in alphabetical order.
But first, I should make clear my investing biases. Just because a stock is undervalued (using back-of-the-envelope three-year DCF analysis), doesn’t mean it’s a good investment; just because a stock is overvalued, doesn’t mean it’s a bad investment. Similarly, just because a stock is likely to increase its revenues substantially doesn’t mean it’s a good investment, and just because a stock is likely to decrease its revenues doesn’t mean it’s a bad investment. I take a huge number of different factors into account when I judge whether a stock is a good or bad investment, and I’ll illustrate that below in my discussion of these fifteen stocks.
Amazon (AMZN). Sales growth estimate: 19% to 23%. Amazon ranks highly on every single one of my factors. I still think it’s overvalued: it’s trading at a free-cash-flow multiple of about 180. But it’s probably a good investment nonetheless.
Apple (AAPL). Sales growth estimate: 4% to 5%. While Apple has terrific price momentum, its other measures are pretty average, and it has shrunk its assets by 6%. I think it’s quite overvalued, but not a bad investment on the whole.
Bed Bath & Beyond (BBBY). Sales growth estimate: –2% to 0%. Bed Bath & Beyond has terrible estimated sales growth (–7%; my method is a little more conservative), a negative sales trend, and only 3 out of 19 analysts rating it a “buy.” I think it’s fairly valued at about $12/share, but a poor investment.
Dorel (DIIBF). Sales growth estimate: –1% to 0%. Most of Dorel’s numbers are average or below average, and it has shrunk its assets by 7%. On the other hand, it’s hugely undervalued, with a price-to-free-cash-flow ratio of 4.5. I think that despite its negative outlook in terms of sales growth, it’s an excellent investment: it had a great quarter, it has stable financials and sales, it has strong price momentum, a nice PEG ratio, excellent forward earnings yield, low price to sales, nice shareholder yield, and a high ratio of unlevered free cash flow to enterprise value.
Facebook (FB). Sales growth estimate: 19% to 21%. No real weaknesses here. Again, like Amazon and Apple, I think it’s overvalued, but a decent investment nonetheless.
GameStop (GME). Sales growth estimate: –2% to –4%. Except for its strong price momentum and slightly below-average NOA/TA, GameStop’s other numbers are abysmal (analysts are estimating its revenue will shrink by 12%!). With negative free cash flow and negative projected sales growth, it’s either impossible to value or worthless. It does have a few things going for it overall, but I would not recommend it as an investment right now.
Issuer Direct (ISDR). Sales growth estimate: 13% to 15%. On my first three factors, Issuer Direct does very well; on the other four, it’s more or less average. By my calculations, its equity is fairly valued, but it’s an excellent investment for other reasons: high growth last quarter; strong industry momentum; low debt; strong price momentum; strong forward earnings yield; and a high ratio of unlevered free cash flow to enterprise value.
Ligand (LGND). Sales growth estimate: 0% to 1%. While average on most measures, Ligand has poor past and estimated sales growth and has shrunk its assets by 23%. It’s very undervalued, though, and quite a good investment, in my opinion.
Limbach (LMB). Sales growth estimate: 2% to 5%. Limbach’s numbers are pretty average. I think it’s a terrific investment anyway: excellent earnings growth, accelerating sales growth, low accruals, stable sales and financials, high asset turnover, low debt, healthy free cash flow, low capital expenditures, good technicals, high earnings yield, low PEG and price to sales, and an excellent ratio of unlevered free cash flow to enterprise value.
Macerich (MAC). Sales growth estimate: –7% (as low as I go, but this agrees with what the analysts estimate). Abysmal numbers all around. The company seems fairly valued at $8.27 per share, but I think it’s an absolutely terrible investment. It has some of the worst numbers, in terms of recent growth, I’ve ever seen, including a declining dividend; its financials are unstable; analyst sentiment is awful; short interest and share turnover are through the roof; and the price momentum has gone off a cliff.
Mallinckrodt (MNK). Sales growth estimate: –4% to –5%. Most of the important numbers are very low, though the sales trend and NOA/TA ratio are average, and their R&D spending is healthy too. This is an absurdly undervalued company, with a price-to-free-cash-flow ratio of 0.2, but that doesn’t mean it’s a good investment; my back-of-the-envelope DCF analysis says it’s worth $85 per share rather than $1.48, but I don’t put much stock in DCF analysis these days. Even at $1.48, it’s a middling investment opportunity at best.
Microsoft (MSFT). Sales growth estimate: 16% to 17%. All the numbers are above average, and price momentum is quite strong. The company seems fairly priced, but I think it’s a very good investment anyway.
Pro-Dex (PDEX). Sales growth estimate: 21% to 22%. All the numbers are very good, except the NOA/TA ratio is average. The company seems overpriced, but I think it’s an excellent investment nonetheless. It’s growing its dividends, its earnings growth is consistent, it just came off an excellent quarter, its sales are accelerating, the industry momentum is great, its return on assets by any measure (free cash flow, operating cash flow, earnings) is very high, it’s buying back shares, analysts love it, and it has a nice forward earnings yield.
Superior Group (SGC). Sales growth estimate: 6% to 14%. Great sales trend here and good price momentum (I’ve gotten a 90% return on my investment so far), though the NOA/TA ratio is a bit high. Although it’s now fairly valued, I think SGC remains an excellent investment, and I’ve only sold a small portion of my holdings as the price has increased.
Tesla (TSLA). Sales growth estimate: 12% to 14%. Nice numbers all around except for the analyst recommendations, which are abysmal. On the other hand, those same analysts are estimating a sales growth of 31%! Either way, the company is absurdly overvalued, with a price-to-free-cash-flow ratio over 250! A back-of-the-envelope DCF analysis says that the stock should be trading at $19 to $29 a share rather than over $400, but that doesn’t make much sense. At any rate, even with such strong estimated sales growth, I would not advise investing in a company that’s this expensive.
Disclosure: I am long DIIBF, ISDR, LMB, PDEX, and SGC.
5 Replies to “How to Predict Sales Growth”
Hi Yuvol – Your article is of great interest to me and very timely as I am starting a project related to predicting sales growth. I was wondering if you could provide greater detail regarding the look ahead bias? If I only use complete data do I need the delay? And what do I do for recent data? Should I still be incorporating a delay?
My desire is to predict sales 52 weeks ahead. Companies will generally be somewhere in the middle of a quarter, so I intend to blend the quarterly sales, based on how far into the quarter the company is. This is an attempt to harmonize stocks that have different year-end dates.
The major flaw in my study is that I’m not dealing with stocks that have disappeared through M&A activity or bankruptcy. So there’s clear survivorship bias, which I simply didn’t know how to avoid.
The chance of look-ahead bias comes in because sales growth is measured as trailing twelve-month earnings divided by previous twelve-month earnings. If I had not incorporated a six-month delay, there would have been a chance that companies with late statements would have already begun their TTM growth by the time I measured the factors. I wanted to err on the safe side. A six-month delay is probably unnecessary, but I think at least a three-month delay would be a good idea. Your suggested approach–blending quarterly sales depending on how far into the quarter the company is–may also avoid look-ahead bias if it’s done carefully.
The biggest thing that I have to get my hear wrapped around is the how to deal with the difference between the company fiscal year and the delay in quarterly reports.
I have a question for you. Are the analysts’ estimates synced with the annual reports? Can I assume that current and next years sales rolls over at the time the annual report is published? Or is this a bit of a gray area? I know that P123 was struggling with this a bit I think before you arrived and with previous data vendor. They were blending the current and next year for some applications.
Estimates are still handled independently of fundamentals. So things won’t always coincide.