4 Ways to Turn Your GenAI Experiments Into Business Results
In an astonishingly short period of time, generative AI has entered the zeitgeist. The biggest companies in the world are now all-in on artificial intelligence, and terms like “prompt engineer,” “deep fakes,” and “large language models” (LLMs) have entered our vocabularies. When even your grandparents are talking about ChatGPT, it's clear this technology has gone mainstream.
The same can be said for the consumer goods and retail industry. As the 2024 CGT Sales & Marketing Study shows, GenAI is changing the rules of the game. Two-thirds of executives believe GenAI will have the single greatest impact on the consumer goods industry over the next 12 to 18 months, especially in areas like social media marketing and e-commerce.
Yet nearly a third of these companies have yet to embark on their GenAI journeys. Most of the rest are in the early stages of experimentation and limited deployments. Here, as elsewhere, sales and marketing teams are operating under tight budget constraints and struggling to measure ROI.
We see this every day in the consumer goods industry. Some are unsure how to get started. Others are burdened with the expectations of stakeholders unfamiliar with GenAI’s nuances. Many focus on the technology without considering adoption, scalability, or the business outcomes a broader perspective can help them achieve.
The reality is that every company's journey will be different, but these four principles apply across virtually all enterprises.
1. Adjust your expectations
Many executives with limited exposure to GenAI expect this technology to transform their companies overnight. That's not going to happen. Aiming too high too soon is a recipe for disappointment.
For example, you won’t be able to create your next great marketing campaign or ground-breaking product simply by throwing a few prompts at ChatGPT or DALL-E. These tools have not yet reached the levels of consistency and structure most enterprises require. But they're an excellent way to brainstorm ideas, iterate on them, and arrive at final concepts faster.
That's one of the ways companies like Coca-Cola and Mattel are using GenAI today. These tools can reduce product development life cycles and help companies better understand areas of consumer interest. For example, Mars has applied GenAI to improve ad relevance, which has doubled click-through rates and improved sales lift by nearly 70%.
2. Focus on productivity
GenAI has the potential to dramatically cut costs and generate new revenue streams. However, the greatest immediate benefit will come from boosting productivity by automating manual processes.
Earlier this year, Walmart used GenAI tools to combine more than 850 million pieces of data into its product catalog, using 1/100th of the headcount previously required for the job. While most companies may not see 100x productivity gains to start, automating time-consuming tasks like production information management and assortment planning is a good way to earn quick wins and prove the technology’s value.
3. Put your best data forward
One of the biggest impediments to successful implementation is that many companies don't know if the data they have is clean and accurate enough for use with AI. Forty-two percent of executives surveyed by Informatica cite data quality as one of the top barriers to broader AI adoption.
It’s a good idea to start with use cases that rely on data you have the most confidence in. For example, using GenAI to extract content and generate metadata from your existing documents and images can save countless hours for teams that currently catalog this content by hand. By uploading their proprietary data to a locally managed and deployed LLM, companies can enjoy the benefits of GenAI while keeping their data secure.
4. Remember: Proving real ROI takes time
Be mindful that these are still early days. The tooling that brings the flexibility and control enterprises require is still a work in progress, and employees will need time to become familiar with the nuances of GenAI. Teams will want to collaborate on the best ways to integrate this new technology into their daily workflows.
Demonstrating real return on investment will take time and likely require more than one isolated proof of concept. But getting small wins now can pave the path toward larger victories later. And the sooner you can start, the better. Organizations that choose to ignore this transformative technology for too long are likely to be left behind.
Vanessa Fiola, Qvest EVP & Applied AI Practice Co-Lead, Qvest
