Consumer data comes in many forms and from many different sources. There are also a wide variety of data types that aren't about consumers, but that give CP brands insight into consumer behavior. 

Big data sets that could be useful to CP companies in understanding consumer behavior across sectors include:

• E-commerce product reviews
• Social media posts and comments
• Weather data
• Retail transactions
• Social media brand/consumer engagement
• Member acquisition touchpoints
• Third-party consumer research
• Property and real estate transactions

For example, a furniture manufacturer could anticipate demand in zip codes with hot real estate markets using real estate transaction data. A makeup brand could anticipate new trends through an influx of customer product reviews on Amazon. The possibilities are endless, and now with generative AI, they’re a lot more accessible.

Why is consumer data strategy important for consumer products brands?

Consumer data has always been an underutilized resource for CP firms, because of how difficult it is to scrape for practical insights at pace and at scale. Thus far, many CP firms have a fractured consumer data strategy, due to a lack of first-party consumer data. 

Very few CP firms utilize large consumer data sets to inform business decisions—most advanced analytics use cases, like predictive analytics, for example, use smaller, structured data sets which don’t cross business functions. But if CP firms did have data sets as large as the entirety of Instagram, or Google search results as their playground, they could drastically increase their profits.

Use case: Beauty brands on TikTok

According to Google, 40 percent of young people conduct internet searches on Instagram and TikTok—using influencer videos to decide what beauty products to buy. TikTok data, like comments and videos, is unstructured and difficult to analyze. But using a large language model, artificial intelligence can easily predict trends, preferences and patterns from this large and unstructured data set, telling a brand which influencer to work with, what their ads should look like, and which products are about to go viral.

How does generative AI change consumer data strategy?

In order to extract insights automatically and at scale from big data, CP firms need to invest in artificial intelligence, and specifically, large language models across key data sets.

Right now, some CP firms democratize consumer data, customer data, supply chain data and more in a centralized, cross-functional “hub.” But these data hubs are often only usable by internal analytics experts that surface insights manually and on a lag, to the point where it’s not able to keep up with the pace of consumer demand.

There are two main new use cases for CPG companies when it comes to big data, with the advancement of LLMs: demand forecasting and trend forecasting.

Traditionally, demand forecasting models rely on historical data, are prone to manual errors and have such a high margin of error that they can be difficult to trust. At the same time, consumer trend forecasts are so far behind the pace of social media that manufacturing can’t keep up. If CP companies can layer an LLM onto their demand and trend forecasting, they’ll be able to predict at a higher level of accuracy, leading to less waste, higher revenue and more consumer engagement.

In 2024, most CP brands won’t yet be able to unlock the full value of big data for demand and trend forecasting, but that doesn’t mean it’s not time to invest.

In fact, the opposite is true. The more data an LLM has, the more useful it is, which means the faster the better for data-driven CP firms.

And with many nuances dependent on the product types CPG firms cover, Publicis Sapient experts have broken down the key use cases for big data in each of the biggest sectors:

IoT interactivity: Generative AI has the power to connect data across personal electronics devices to form better responses, recommendations and insights. Generative AI combined with natural language processing (NLP) will improve engagement across electronics products.

Personalized Rx: Consumers want tailored and personalized beauty and personal care products and regimens, which are currently impossible for large CP brands to offer. Generative AI can synthesize recommendations from large data sets to create a better recommendation model for consumers.

Decrease waste, increase sustainability: If CP firms can increase the accuracy of demand forecasts, they can decrease retailer waste. Using retailer data and inflation forecasts, generative AI can inform a more accurate and efficient prediction algorithm.