Marketing can’t save bad products

This year, I encountered three different presentations on marketing in three different conferences, with a common refrain among them.

Marketing, no matter how well-executed or effective, cannot make up for a deficient product.

A product doesn’t necessarily have to be flawed, in and of itself, to be deficient. A product can be considered problematic if it fundamentally doesn’t have the right market or customer fit. Or the product is a poorly defined solution in search of the perfect problem it could solve. These are just two of many possible factors negatively impacting products.

A relevant corporate story

This principle was clearly on my mind when IBM made the blockbuster announcement that it was going to acquire Red Hat for $34 billion, the third largest ever for a tech company acquisition.

IBM’s decision is seen by some as a measure of urgency to save itself by going back into its computing back-end roots, following muted growth success in the cloud space (it’s well behind Amazon and Microsoft). The acquisition is also interpreted as silent acceptance that fringe technology products, notably Watson, haven’t been panning out as as hoped. Thanks to abundant marketing-generated publicity, Watson was widely touted as the beacon offering that could save IBM’s future.

Watson is of course IBM’s AI and machine learning platform. Created to outsmart "Jeopardy!" contestants, IBM subsequently went on a mission to position Watson for more practical applications, especially healthcare.

A large part of this effort happened by flexing IBM’s enormous marketing muscle. Watson was heavily publicized in mainstream media (remember the TV commercials?), promoted throughout prominent tech and marketing conferences, and paraded around to healthcare and other key vertical industries that could, potentially, benefit from AI and machine learning.

But as it turned out, Watson was adjudged to have been overhyped, over-promised in terms of its ability to revolutionize healthcare. In the end, Watson-relevant products were not yet viable, not yet ready for primetime. A prominent cancer treatment & research center cut its losses with IBM, having invested some $36 million in a collaboration to develop AI-related solutions to vexing cancer treatment challenges.

Marketing was clearly one of the focal points of blame for Watson’s predicament, as duly noted in TechCrunch’s reporting last year (in reaction to an analyst’s highly critical report questioning Watson’s value to shareholders):

The narrative isn’t the product of any single malfunction, but rather the result of overhyped marketing, deficiencies in operating with deep learning and GPUs and intensive data preparation demands.

Avoid hype and be realistic

Particularly striking was that IBM’s marketing was replete with promises that were promoted aggressively to customers but not yet proven viable or realistic.

It’s a cautionary tale for all marketers: always promote your product as something genuinely real and with capabilities your company can confidently back up today. Anything beyond that and you risk getting yourself into an undesirable cycle of excessive hype, fluff, and worst of all, deception.