The uncomfortable truth about AI: what if it's not for your business?
Are you thinking about adding AI to your business?
The moment has finally arrived: you finished rolling out your management systems, your product is resonating with your target audience, and everything is up and running. The business that once lived only in your imagination is now a living thing. Small and growing, but alive.
Then a colleague tells you they started “incorporating artificial intelligence” into their company and are already seeing extremely positive results. You notice your CRM now has “AI features” that promise to make every task easier. And your competitors claim they “use AI to improve customer experience” with excellent outcomes.
You start feeling like you’re missing something. Maybe having a data management system was just the beginning? There’s some truth to that. But you also need to stay sharp about when technology actually helps your company grow, and when adopting it is just throwing money away.
It starts with the data
AI is not going to magically solve all your problems, and not every problem can (or should) be tackled with AI. It’s a tool, not a solution.
Many early-stage companies believe AI is the key to standing out from competitors and boosting their metrics. While that can be true, it’s incomplete: the reality is that AI is only as good as the data it runs on.
Before thinking about applying AI, ask yourself:
- Am I collecting quality data?
- Are my processes mature enough to automate?
- Am I clear on which business decision I want to improve?
To start with, what most people today call “artificial intelligence” is actually a very specific subset of a larger group of technologies. And what used to be called “AI” has been around for over 70 years. Chatbots like ChatGPT and the advanced applications you see in various SaaS products are technically called large language models (LLMs) and are part of generative models: algorithms that can create content (audio, text, video). These capabilities are so impressive that they feel more like science fiction than statistics.
However, your business probably doesn’t need any of that. What it almost certainly does need is a concrete, actionable way to exploit its data and get fresh perspectives from it. A generative AI chatbot probably won’t grow your business, but having the ability to predict when a customer is about to churn, to personalize product recommendations in a shopping cart, to segment your audience and understand it deeply. That will.
What really matters is making your data work for you, not adopting a technology because it impresses you.
Don’t run without knowing where you’re going.
Human intelligence
Does all this mean you can’t use AI? Of course not. But be careful: not everything that glitters is gold, and it’s critical to understand the “why” before adopting any new technology.
The reality is that in the vast majority of cases, an organization gets more value from non-generative models: “classic” machine learning. These algorithms and statistical techniques may not be as “flashy and magical,” but they are far more useful for businesses that want to do things right and grow, rather than jumping on a bandwagon out of fear of missing out.
Traditional machine learning (predictive models, segmentation, churn analysis, etc.) typically delivers much more real value than flashy generative AI applications.
With simpler but robust models you can:
- Predict customer churn
- Personalize product recommendations
- Segment audiences for commercial purposes
- Optimize your marketing campaigns
- Forecast sales or stock rotation
AI doesn’t work alone: it requires strategy, culture, and follow-through
To put your mind at ease and cut through the FOMO, it’s worth noting that building data-driven solutions is still an unsolved challenge for the vast majority of organizations.
According to a BCG study1, only 26% of companies currently manage to move past the PoC stage with AI and generate sustainable business value and impact.
And here’s the interesting part: according to that same study, many companies don’t track the KPIs of their AI initiatives, but those that do are the ones that actually deliver on AI’s promise: cost savings, revenue growth, broader business improvement, and so on.
AI (and traditional machine learning) can be a valuable ally, but only if it’s aligned with your business strategy. If AI, or any technology, isn’t helping you grow and improve in a concrete, tangible way, you’re wasting your money.
We recommend asking yourself these questions before moving forward:
- What’s the specific problem I want to solve?
- Do I already have the data needed to solve it?
- Can I measure the impact an AI solution would have?
- Am I deciding out of need or out of hype?
It’s important to have a plan where you can clearly outline where your data strategy is headed and how aligned it is with your business metrics. If you’re evaluating how to get more out of your data, or you want to understand whether AI makes sense in your context, we might be able to help.
Getting value from data requires a strategy. It takes technology, cultural shifts, leadership commitment, time and resource investment, and yes, a lot of patience.
At deployr we work with companies that want to apply technology with purpose and strategy. If you think the time has come to explore how AI can contribute to your business, without magic formulas or inflated promises, we’re just a message away.
See you in the next one!
