They really choose to start somewhat from the bottom then, doesn't Whole Foods have something like ~3%-5% of the grocery market?
Compare that to companies with comparatively massive market shares like Kroger and Walmart. I can see an argument being made that groceries is a good way to go because people have to eat, but the topology of the industry doesn't really have a medium sized company that you could pour gasoline onto to gain a lot of market share. Something in a different industry might have been a better play to pilot before moving into other retail spaces.
They're at the top in brand reputation and at the bottom in market share. People who care a lot about food and are willing to spend on it accordingly shop at Whole Foods. There are relatively few of these people, hence their small market share. (A similar dynamic happens in most consumer industries: Apple famously has only 18% of the smartphone market but makes 87% of the profits, because they market to people who are willing to spend a lot on a cell phone.)
This is a pretty rational choice for Amazon, because Amazon's core competency is making things cheaper and more widely available but they have little experience in making things better - and indeed, their whole value proposition is largely in making things commoditized, helping you buy a "good enough" product more conveniently for less money. With Whole Foods, they're buying the brand and the expertise in quality, and then they can apply their existing strengths in logistics and supplier management to wrangle down the price and expand market share.
People that like to spend a lot of money for the same product shop at Whole Foods. Smart people, and price conscious people, buy the same product for a fraction of the price at their local franchise grocer, whether that is Hannafords, Market Basket, Price Chopper, Publix, or what have you, wherever you are.
Walmart does have typically inferior grocery items, still; it's been 15 or so years since the super centers were built in this area with groceries, but the meat and produce is still well below par.
Many grocery chains are regional. The national ones (Kroger, Albertsons) are so huge they were probably out of Amazon's price range. Walmart and Target aren't for sale.
If the article is right, all Amazon may need to kick things into profitability is enough economies of sale (built in customers) that their perishables can be fresh.
From there they can sustain a viable business, and they can increase its size as needed.
In other words, one theory is it's not so much about buying market share, it's about buying a roughly viable business they can piggyback on while they build market share of something new.
Compare that to companies with comparatively massive market shares like Kroger and Walmart. I can see an argument being made that groceries is a good way to go because people have to eat, but the topology of the industry doesn't really have a medium sized company that you could pour gasoline onto to gain a lot of market share. Something in a different industry might have been a better play to pilot before moving into other retail spaces.