‘Revolutionizing Retail’: Content creation and management with AI (webinar recap)

Large potential of GenAI in retail

Let's start with some statistics: Artificial Intelligence in the retail market will cross USD 19.57 billion by 2028. But more importantly, a CAGR (Compound Annual Growth Rate) of 30% is expected to last until 2028 (started in 2022) due to increased adoption of AI to boost customer experience (Source: SkyQuest Technology Consulting Pvt. Ltd.).

So, retailers investing in AI are facing a more than interesting ROI ratio. But what exactly is this ROI based on?

First of all, AI is a driver for cost savings. Most cost savings are predicted to be in marketing and sales, software engineering and customer operations. But despite its promising potential in these areas, the adoption of generative AI tools remains low. Less than 1 out of 5 (18%) of managers in retail and CPC (cost-per-click) regularly use these tools at work. And less than 25% of leading CPG (consumer packaged goods) companies started scaling high-priority Generative AI use cases.

This could have something to do with the concern over the level of quality of AI-generated content. Or maybe it's simply the lack of understanding of and expertise in the GenAI field. Both were listed as the top concerns of the webinar attendees in one of the polls.

Secondly, there’s the GenAI potential for transformation. The biggest impact in the short term is considered to be for knowledge workers: generative AI will help them do their jobs faster and use and analyze available data in order to reach better conclusions.

Architectural campaign landscape: going beyond the weekly flyer

In order to capitalize on the full potential of GenAI, a fundamental change in the retail campaign dynamics and architectural landscape is imperative. Many retail promotions are still based on a campaign process that is centered around a weekly flyer. But online channels are becoming increasingly important to convey your message and the variety of those channels is increasing by the day.

Campaigns are more and more about storytelling. They are centered around different themes, social issues, or specific events that resonate with your customers. Retailers want to be able to create targeted campaigns on the fly for stock they want to get rid of in a certain location. Or to boost products for specific local customer needs.

What they need?
A multi-channel data-driven architecture that allows for content orchestration instead of touchpoint-centric campaigns. A central platform that supports agile and linear approaches will do just that, with AI and machine learning automating activities and taking on an advisory role.

Appliance of GenAI in retail in practical terms

In theory, we can all talk about how AI can revolutionize content management and creation in retail. But instead of hypothesizing, let’s put theory into practice! Take Amazon, this global retailer created its own AWS (Amazon Web Services), a cloud computing offering with a wide range of services from computing power and storage to AI tools. The practical use of AWS as far as AI tools are concerned, is best shown based on real-life examples.

The first example is the GenAI for Amazon Sellers. This makes it easier to write engaging, effective product listings and helps shoppers find what they are looking for. After all, better descriptions lead to better conversion!

But GenAI can also be used for troubleshooting problems with all kinds of equipment. In these cases, AI is coupled with the retailers' own solutions to come up with extensive information that helps find quick answers when things go wrong or in case customers want more information on the use of the equipment.

And finally, AWS AI is also used to check and optimize planogram compliance. Retailers use it to analyze pictures of shelves in the store and use that information to assess the current planogram and to come up with ways to improve.

Proven AI ROI in retail

Experimenting with a wide range of AI applications in retail does pay out. Not all tests are successful, but when they are, it saves time and costs but also generates different benefits such as better search rankings, increased traffic to company websites, and an easier process for consumers to buy products and services.

• AI is used to extract product features from factual descriptions delivered by suppliers in a wide variety of formats, including PDF files and images. AI extracts the information, converts it into the required format and, after human approval, this data is automatically uploaded to the system. All data is available for sorting and filtering and ready to be published on the company website.
• AI can also improve data quality by comparing product information with a wide range of collected data; not only from within a certain company, but also from the internet at large. After defining a number of trustworthy websites upfront for example, information like product measurements can be compared with similar products in order to detect anomalies.
• If you have a large product assortment, checking all product descriptions can be a humongous task. By using templates with their own specific conditions, you can let AI do the checking for you, ensuring 100% correct text. If you require multiple languages AI can step in as translator, with excellent results. Especially when you use your own dictionary of preferred translations and spelling to correct the machine translations.
• For many consumers, ease of use still has a huge impact on actual online sales. You also notice this when AI is used to present initial search results with similar or related products so customers don’t have to go search again. For one do it yourself retailer this application of AI resulted in a 4,9% increase in revenue.

AI for consistent product images at scale

Let's take a look at leveraging AI in the image production flow. We are all too familiar with challenges like product images that are supplied in a mixture of formats, resolutions, and aspect ratios. How can you ensure brand and format consistency, not only for your own brand, but also for suppliers that have your products in their assortment?

By getting AI on your team!

Based on predefined rules like full or partial product shots, the required asset ratio and other requirements coming from your brand guidelines or those of partners you work with, AI can do the bulk of the job. You can use life style shots for your products for example, but for a partner you can drop out the background or take out the logo’s that happen to appear in the image.

If you want to grow as a retailer, you will have to automate your image production. AI offers you choice at scale. Anything you can think of, just craft the rules and push it out to the entire image base – how many times you want to or have to. That way, AI is not about reducing your headcount but about enablement: making the people you have perform better.

Watch the full webinar below

AI in retail: start experimenting now

The benefits of using AI in retail depend on the type of business you are in and your skills to use AI. But however scary it may seem, just start. AI is here to stay, developments go fast and you don’t want to be left behind.

Take a small piece of your business and experiment in a safe environment. Based on the outcome, you can take your next step. Start with those parts of the workflow that can benefit from AI recognizing and filtering images as part of your digital asset management, for example, or automatically setting focus points.

Walking the walk is the only way to gain experience in the use of AI and benefit from it. Let AI help you make your life easier and reduce work, so you can pay attention to those parts that add creativity to your campaigns and make your offering stand out from the rest.

WoodWing can be valuable to players in the retail sector in many different ways. Interested to find out how? Click here, or contact our retail experts.