Thirsty for more expert insights?

Subscribe to our Tea O'Clock newsletter!

Subscribe

From retail to luxury: two innovative Generative AI use cases

fifty-five
Published on
8/8/2024
As generative AI continues its expansion into all kinds of industries, businesses are looking for new ways to apply the technology to improve their processes, drive growth, and stay ahead of their competitors.

Still, beyond its most obvious applications, such as basic text and image generation, many are struggling to find relevant use cases and, equally important, make said use cases work properly and enable development at scale. 

At fifty-five, we work closely with our clients to find the GenAI solution that fits their specific needs, as demonstrated through our work with Chronodrive, an online grocery-shopping app wanting to build a “Shazam for food” or Concentrix + Webhelp, a B2B specialist who partnered with us to perfect their lead-finding processes with prospect summary, personalized ice-breaker and follow-up messages generation. On May 30th, during our “GenAI for Luxury” conference organized with Google in Geneva, we presented two of our newest case studies, with Breitling and Nespresso, which we will explore below. These case studies illustrate the depth of generative AI’s potential to solve common yet persistent problems for enterprises. 

And thanks to fifty-five’s generative AI flow builder, SyMap, each of the prototypes presented here was delivered within record times.

Meet SyMap, fifty-five’s generative AI flow builder

To help both our consultants and our clients explore GenAI’s many capabilities, fifty-five developed SyMAP, a simple, user-friendly tool that lets you:

  • Drag and drop functions from every model available
  • Create new GenAI applications in a few clicks
  • Test your ideas within a secure sandbox
  • Compare model accuracy, quality, and cost to find what best fits your needs
  • Save and share your projects with your team.

Created in the winter of 2023, SyMap has already been used by several of fifty-five’s clients and prospects to successfully turn GenAI ideas into actual use cases, and proved its effectiveness yet again during fifty-five’s latest Hackathon.

NESPRESSO - E-COMMERCE
Improving product visibility online by applying GenAI to product feeds

Selling coffee capsules and machines worldwide, Nespresso wanted to improve its e-commerce visibility with a particular focus on its product feed. Driving online visibility for a brand’s product, a product feed is a file containing all the details about an item’s properties and attributes to display for shoppers; this component is crucial to driving online sales, as search engines promote ads by prioritizing data quality, product relevance, predicted CTR and the bid.

Due to data quality issues and mismatches between product descriptions and user search requests, Nespresso products were frequently not promoted on search engines, causing the brand to periodically lose out to competitors. 

Leaning on SyMap’s flow-building capabilities, Nespresso and fifty-five set out to overcome four specific product feed challenges:

  • Adapting to Google’s platform guidelines
  • Ensure conformity with Nespresso’s guidelines
  • Correct and enrich product descriptions
  • Overall feed data quality

While the improvements obtained would be individually slight, they can lead to substantial gains  when combined.

Focusing on a specifically scoped POC (Proof of Concept), fifty-five combined functionalities from various GenAI models to form a product feed expert custom flow on SyMap. This POC was built to generate a best-in-class product feed catalog, fixing issues that hindered visibility in Google Merchant Center. Pre-prompting input by fifty-five combined best-in-class product feed catalog requirements, guidelines from Google Merchant Center, and guidelines from Nespresso. The SyMap flow was also calibrated to identify a list of issues to fix according to product type, generate format suggestions, and justify and evaluate its own output. The resulting product feed was both corrected and enriched to ensure optimal visibility as well as accuracy. 

The scope’s new product catalog was subsequently pushed to Google Merchant Center. fifty-five ran test campaigns and then measured before and after variations on impressions in CTR. While POC results should always be read with caution, as their scope is limited by design, the results were already promising, with x3 visibility/impressions and x2 traffic. On a broader level, this POC shows that GenAI can enhance efficiency and optimize operational, time-consuming tasks, freeing up time to improve feed quality and test innovations from a media/marketing perspective.

BREITLING - PERFORMANCE REPORTING
Making data available to C-level executives in just a few keyboard strokes

Breitling wished to deploy an internal tool for C-level team members after realizing that they often started their workday with the same questions regarding specific performance metrics, and had to constantly seek answers in Power BI. To simplify their work, Breitling partnered with fifty-five to explore how to accelerate such a tool’s development through GenAI.

In just a few weeks, fifty-five presented their prototype to the brand, having used an anonymized testing Biquery dataset in a Symap flow; this allowed users to ask questions in natural language and obtain clear, data-based responses in visual form if needed, all within seconds. This flow allows Breitling team members to ask questions in natural language and obtain clear, data-based responses in visual form if needed, all within seconds. To build this flow, fifty-five worked with Breitling to recreate the train of thought of a C-level officer sifting through data to find the cause of revenue loss, with questions including: 

“What is the weekly evolution of my revenue?”

“What product collections registered the largest decrease in total revenue from month X to month X?”

“What is the daily evolution of my “Collection Y” revenue in month X, year Z?”

Note: the graph presented here was generated with mock-up data.

The flow successfully provided clear results, interpreting numbers correctly throughout fifty-five’s tests thanks to thorough work on pre-prompting and prompting the models used to guide them in reading the provided data and double-checking their own answers. This use case would require a much more complex architecture to be fully deployable as a tool, notably adding more data sources as input and deploying an application for multiple users, but this first test impressed the Breitling team, who is eager to keep exploring GenAI solutions. 

Both Nespresso and Breilting’s POC were delivered within weeks by fifty-five, proving that with the right expertise and the appropriate tools, tailored-made GenAI solutions are well within the reach of businesses. These projects also relate to 2 of our 6 major Generative AI use cases determined by our experts to be the most likely to benefit from GenAI: Data Analytics and Product Feed Improvement. AI expertise combined with creativity and problem-solving skills is essential to unlocking this technology’s true potential, resulting in tremendous productivity gains.

If you would like to explore how fifty-five can help your organization find the best GenAI solutions for your needs, please contact us!

All articles

Related articles

Data Break Podcast Recap: How is GenAI Shaping the Future?

4 mins
Robin Clayton

Fad or Forever: Is Gen AI here to stay?

8 mins
Dustin Stewart

The Potential of Generative AI for Accelerated Business Transformation

6 mins
Margaux Montagner

Thirsty for more expert insights? Subscribe to our monthly newsletter.

Discover all the latest news, articles, webinar replays and fifty-five events in our monthly newsletter, Tea O'Clock.

First name*
Last name*
Company*
Preferred language*
Email*
Merci !

Votre demande d'abonnement a bien été prise en compte.
Oops! Something went wrong while submitting the form.