The internet has been abuzz the last couple of months about the prospects of AI in business in the wake of OpenAI’s public launch of ChatGPT. Reactions have included everything from thinking we’re on the imminent verge of a techno-futurist scenario up through complete dismissal. The reality is somewhere in the middle of the vast chasm between those two extremes.
AI has some amazing potential to enhance business, if you understand its current state, limitations, and capabilities. A few years ago, I co-wrote an article with data scientist and business AI expert Doug Pestana on how SME’s can leverage AI using open source tools on a limited budget. The state of AI has certainly advanced since then (with ChatGPT being a watershed example), but is still a long way off from any wild sci-fi scenarios. Given that understanding, what are some ways your business could use AI in 2023? Here’s a few ideas worth exploring:
1. Customer Personalization. This is along the lines of what Doug and I wrote about in our 2019 article. For example, machine learning technologies can take your customer data and find statistical correlations that help you maximize revenue through up-sell/cross-sell initiatives by generating product recommendations. This is likely already familiar to you if you do any online shopping or use any streaming services. Example: you watched this show, so you’d probably also like this one.
2. Customer Service. Again, something you’ve likely already experienced yourself with service chat bots. A bot trained to respond to common service questions or lookup information on customer orders for service requests can save you a lot of time and overhead. Just be mindful of not being overly-reliant on this solution for everything; companies that implement deep cuts to human-assisted service and make it really hard to contact an agent when you truly need one tend to do serious damage to their NPS and customer satisfaction ratings. Keep the bots focused on answering the low-hanging fruit: easy questions that you get a lot of. Let your human agents deal with the complex service cases.
3. Forecasting and Inventory. Time series model forecasts can take your historical sales and inventory data and extrapolate seasonality, inventory turn, sales projections and more so that you can more accurately make merchandise purchasing decisions and plan your sales cycle. This is one of those cases where human intervention in the forecasting tends to produce more harm than good (not to mention a lot of corporate politics trying to explain away why such and such a business unit missed their targets), while relying on the statistics tends to be more accurate.
4. Store Planning. This will likely require the deployment of adjacent technologies to get the right data for the AI to leverage and so is definitely a more advanced approach, but AI can be used for things like optimizing your store layout and floor plan (i.e., what path do you want a customer to walk to maximize ticket? Where are they spending the most time? Where are the bottlenecks?) and analyzing traffic patterns to optimize store hours and staffing.
5. Automatic Dynamic Pricing. Again, a more complex strategy that will likely require a good deal of external data feeds, but training a pricing model to learn and adapt to customer spend habits while also considering competitor pricing is an incredible way to stay nimble in your pricing optimization. This is certainly easier to achieve in e-commerce than brick & mortar, but both are possible. Just make sure you implement proper parameters in your model so that your prices don’t pass certain thresholds that you would find undesirable.
These are just a few of the ways that retailers and other businesses can use AI in 2023. Technology continues to change the way we do business, and while AI may not yet be anywhere near a techno-futurist scenario, it does have powerful capabilities to augment and strengthen your business today.