What goes on inside a customer's mind long before a purchase is made is fascinating. Decisions are shaped by many factors, such as price, brand reputation, emotion, convenience and more. 

First comes the moment of recognition, realising that you need a product. This might be a true need, i.e., a product essential for your survival e.g. food, or a want (not necessary for survival but something that improves the quality of life). This recognition can be triggered either by external stimuli e.g. advertisement or by internal feelings/motivations. Once that need is identified, the person then starts searching for ways to tackle it as well as evaluates alternatives between different options based on things like price, quality, features, value. A purchase decision is then made. After the purchase, the consumer reflects on whether the purchase decision taken met their expectation which often leads to repeat purchases... or vice versa in case of dissatisfaction. 

Inside our minds, long before the purchase is made, a subtle dance of trade-offs takes place. We weigh different combinations, until the "right" combination is found based on our needs, budget, priorities, urgency and even what we've heard about the product (online reviews for instance can make or break a purchase). The factors that affect the purchase decision are personal to us. 

Let's take the example of choosing a mobile phone. I might prioritise a better camera because I enjoy taking photos wherever I go. Given my budget and my preference for a specific brand, I narrow the choice between two models. One offers a better camera and better battery life, but it is last year's model. The newer model has a shorter battery life. Because I would rather have the latest version, I am willing to accept the shorter battery life, even if that means carrying a power bank, rather than buy the previous model. Meanwhile, someone else with the same budget and same brand preference might value battery life more than having the newest release and therefore chooses the other option. 

So decision-making is sort of a black box (or as that phrase from Forrest Gump "like a box of chocolates, you never know what you're gonna get"). However, once you start working with preference-driven models, that box opens and starts "showing colour", and that box of chocolate, is not so random after all. 

Starting with a simple question when it comes to designing a successful product: "Which features do people TRULY care about?" Be careful not to assume the features they value, but the ones that they actually choose when no one else is watching. Once you start quantifying these patterns a story emerges. 

Customer Preference Modelling becomes a window into how people weigh different attributes, how they comprise between these and hence how they prioritise. It reveals how one product feels intuitive to one person and completely irrelevant to the other. However,  understanding these is not enough. It is what you will do with that knowledge that is important. 

That is where Product Portfolio Optimisation comes in. It is a balancing act between customer appeal and business performance, a bit like assembling a puzzle in which every piece matters, as well as those very human preferences that underpin them all. With multi-objective optimisation you can actually see where the sweet spots lie and then there are the "what-ifs" that businesses ask, for example: "What if the trend shifts?", "What if material costs increase?", "What if we redesign the product entirely?", "What features can we remove without losing market share?"

Scenario simulation becomes a bit of a crystal ball! Not to predict the future but to illuminate how different futures might unfold: a small change in one variable can cause huge ripples.

Along the way, segmentation plays a part in it too, it turns out there isn't one customer, there are many! Segments with shared motivations, shared frustrations, shared budgets, shared lifestyles.... and the list goes on. Once you understand these groups you can tailor your product, pricing and communication in a way which feels more meaningful. 

So yes data, models (that’s whether they’re built on historical data or are dynamic and looking to predict a future state), simulations… they’re the backbone into understanding which features work, what to do if the scenarios/variables change and how to adapt quickly not to lose market share, as well as how to design the product to make it appealing and interesting to your customer base. Because the story is always about people in the end: how they choose, what they value, and how little shifts in design can make a world of difference.