I doubt I know a single person under the age of 75 who hasn’t purchased at least one item of merchandise from Amazon over the years. And I know quite a few people whose only shopping experience for the holidays is a date with the Amazon website.

Still, some of the breathless stories and statistics that are put forward about Amazon and its business model seem almost too impressive to be true.

I’m not just talking about news reports of drone deliveries (a whole lot of “hat” and far less “cattle” there) or the idea that fully-robotic warehouses are just around the corner – although these stories do make for attention-grabbing headlines.  (Despite the continued need for human involvement, the way that robots are being used inside Amazon warehouses is still quite impressive.)

Moreover, a study published recently by BloomReach based on a survey of ~2,200 U.S. online consumers finds that Amazon is involved in most online shopping excursions, with nine out of ten online shoppers reporting that they check Amazon’s site even if they end up finding the product they want via another e-commerce resource.

More than half of the BloomReach survey respondents reports that they check on the Amazon site first — which is a new high for the company.

But are all of the reports about Amazon as credible?

Recently Doug Garnett, CEO of advertising agency Atomic Direct, penned a piece that was published in the December 2016 edition of Response Magazine. In it, he threw a dose of cold-water reality on some of the narratives surrounding Amazon and its business accomplishments.

Here are several of them that seem to contradict some of the commonly held perceptions:

“Amazon is a $100 billion retailer.”

Garnett notes that once subtracting Amazon’s non-retail revenue for 2015 (the last year for which financial data is available), the worldwide figure is more like half of that.

In the United States, Amazon’s retail sales are closer to $25 billion, which means it makes up approximately 6% of total retail sales.

That’s still very significant, but it isn’t the dominating presence as it might seem from all of the press hype.

“Amazon is profitable now.”

Yes, it is – and that’s after many years when the company wasn’t. However, approximately three-fourths of Amazon’s profits are due to selling cloud-based services, and the vast majority of the remaining profit dollars come from content delivery such as e-books plus music and video downloads.  So traditional retail hard-goods still aren’t generating profits for Amazon.

It turns out, just as retailers like Wal-Mart, Target and K-Mart have discovered, that replicating a retail store online is almost always a money-losing proposition.

To underscore this point, Garnett references this example of a merchandising campaign in 2016 as typical:

“When one unit was sold on Amazon, eight were sold at the retailer’s website and 80 were sold in the brick-and-mortar stores. The profit is in the store. 

For mass-market products, brick-and-mortar still dominates. Amazon is a nice incremental revenue stream, [but] not a valid alternative when you’re playing in the big game.”

It also means that companies that are looking to Amazon as a way to push their products into the marketplace should probably think twice.

At the very least, they should keep their expectations realistically modest.