CRM GPT

Before the AI revolution, CRM had a well-defined, if not often discussed, role that tied it to the product commoditization cycle.

For instance, when a category is new, businesses employ many people to sell and service the new, new thing. The high labor component makes new products relatively expensive compared to other, older solutions. Yet customers willingly spend extra because the solution is that much better than what came before. For a time, it might even be unique.

Over time, competition drives vendors to reduce costs often by eliminating labor, but this requires product simplification and developing some level of customer familiarity with how the product is used.

The trouble is that if I don’t need instruction on using some gizmo, chances are good that it operates in ways that are close to how the competition’s thing works too. In other words, the cost of change is nil. For example, if you have an iPhone, it works very much like an Android phone. That’s a problem because without a person selling the advantages of your product, customers go with what feels right to them.

Ideally, if a vendor could fire most of its employees and still provide top-notch sales and service while maintaining existing prices, the labor cost would fall to the bottom line.

However, as a practical matter, the commoditization cycle has dictated that as products get simpler or at least easier to use, vendors come to rely on the retail channel rather than direct selling, and margins get squeezed so that only a small handful of vendors serve the whole market. More importantly, profit only comes with a massive market share.

Gen-AI and CRM

Generative AI’s appeal stems from the potential to lower costs (labor) while maintaining close customer relationships. Hence, CRM vendors are all over it, like junkyard dogs.

If a machine can do the work of a low-level employee, theoretically, the vendor could pay once for the technology, release the human, and still maintain the customer relationship. This approach would pay dividends later during the inevitable cross-sell and upsell cycles.

By the way, this is not about reducing labor for the sake of it. It’s about escaping the retail vice in which many vendors submit to selling their wares on so-called platforms but give up a sizeable portion of their margins in the process.

Generative AI has the potential to stop the commoditization cycle in its tracks. Instead of going to retail, vendors could send their old products to places on their websites where generative AI runs everything. Staying out of the retail channel has many advantages for the vendor, among which are greater control of the product, its merchandising, and, most importantly, price.

Logically, competition won’t let up, so the new margin space afforded by generative AI will erode. But maybe not. If airlines can be used as an example, more markets may become oligopolies where a small handful of vendors control things. Unlike a monopoly in which a single vendor controls product and distribution, an oligopoly maintains enough of the fiction of a free market to avoid censure.

Airlines using the hub and spoke model don’t collude to fix prices, for instance. They simply publish their fares and schedules. That’s precisely what you’d expect in a free market or a price-fixing scheme. But it’s legal.

My Two Bits

So, there’s a lot for vendors to gain with generative AI. Platforms like Amazon have used algorithms to upsell and cross-sell for many years, and generally, many vendors have benefitted from this form of active selling. But if every vendor can now have the power of generative AI in-house and eliminate the possibility of some platform selling customers away from their products, why not use it?

There is a small risk in this scenario that markets could simply freeze up as customers discover it increasingly difficult to find what they want. But generative AI can help there too.

We’ve already seen search engines optimized with generative AI, and this could replace the infinite shelf space of a site like Amazon, moving it to the commons. Such platforms might continue to exist simply because they have massive logistics behind them. But in a big turnabout, generative AI might become the thing that commoditizes those platforms.

Plus ça change, plus que c’est la même chose.