I only care about the price-to-sales ratio where the past record proves sales turn into cash profits. If a stock has a low price-to-sales ratio and no retained earnings, no free cash flow, etc., I ignore it. Same with net/nets that have no retained earnings – unless they’re liquidating.
I look at the free cash flow margin (FCF/sales). I go back at least 10 years. If I’m real interested, I go back as far as EDGAR goes – usually 13 to 17 years.
I figure the historical free cash flow margin’s mean, standard deviation, and variation coefficient (standard deviation/mean). I love a super low variation coefficient like Pepsi (PEP) or United Technologies (UTX) – both around 0.10. But I’ll take Barnes & Noble around 0.75. BKS’s price-to-sales ratio is so low, I don’t need the free cash flow margin to be what it used to be to make money in the stock.
I won’t buy something with lots of free cash flow margin variation unless it’s a net/net. Even then: I focus on net/nets that once had lots of free cash flow with little variation.
So I don’t look at price-to-sales ratios alone. I look at something like:
(Sales * FCF Margin) * (1 – FCF Margin Variation)
I’ve never written that down. And I don’t mean it literally. I don’t mean you can value stocks that way. But those are the moving parts I see in my mind’s eye: sales, FCF margin, and FCF margin variation.
That’s how I think about earning power.