Blog | Salsify

How Generative AI Works With Ecommerce Technologies | Salsify

Written by Doug Bonderud | 11:00 AM on August 26, 2025
Three years ago, artificial intelligence (AI) was the big ecommerce business buzzword. Now, it’s the next iteration of these intelligent tools: generative AI (GenAI).
 
According to Gartner, in 2022, just 2% of outbound marketing messages were personalized using GenAI — but that number is projected to reach 30% by the end of 2025. Additionally, 45% of companies say they’ve increased their AI investments since the launch of tools such as ChatGPT, and 70% are investigating AI’s role in their digital commerce efforts.
 
Rising interest leads to a key question: How does GenAI work with ecommerce technologies? Does the evolution of intelligent tools mean the end of traditional sales and marketing? Or is AI simply too big for its digital britches?
 
The answer lies somewhere in the middle. Here’s how GenAI works with current ecommerce technologies — rather than against them — to deliver improved outcomes.

Where GenAI Fits Into Existing Digital Commerce Systems

GenAI uses a combination of large language models (LLMs) and natural language processing (NLP) to parse and respond to user queries.  Combined with access to multiple data sources and the ability to learn over time, GenAI has rapidly become a powerful force in ecommerce.
 
GenAI has most notably impacted search results. Many Google queries now come with an AI-generated summary before the first search listing. If companies can create content that gets found and used by GenAI tools, they can improve market exposure and drive customer engagement.
 
This process is known as generative engine optimization (GEO). Similar to its search engine optimization (SEO) counterpart, it prioritizes getting content noticed and displayed by search engines.
 
GenAI also fits into other functions, including the following.

GenAI Streamlines Product Information Management (PIM)

PIM tools ensure that product information, such as pricing, shipping, materials, and other relevant data, is accurate and up-to-date on all product detail pages (PDPs).
 
For ecommerce companies that sell hundreds or thousands of products across multiple geographic locations, the scale of data management required makes PIM a time and labor-intensive process.
 
GenAI can streamline this process. Companies can set up automated evaluations of product pages and use GenAI solutions to fill in the blanks and keep data current.

GenAI Makes Digital Asset Management (DAM) Easier

Ecommerce companies now manage a host of digital assets, including product pages, videos, images, and podcasts, among others. DAM tools provide content repositories for the sales and marketing team.
 
Augmented by GenAI, DAM solutions can both create new assets using current data and make it easier for users to find the image, video, or other asset they need.

GenAI Improves Content Personalization

According to the “Ecommerce Pulse Report: Q4 2024” from Salsify and the Digital Shelf Institute (DSI), 37% of shoppers buy more often due to personalized product recommendations. GenAI excels at improving these recommendations by combining data such as consumer purchase and service history with market trends and larger-scale factors such as economic shifts.

GenAI Empowers Data Analytics

Finally, generative tools can help empower data analysis. While traditional analytics tools have proven their worth for in-house data evaluation, the scale of data required to understand customer behavior requires the speed and scale of GenAI.
 
For example, GenAI can combine local data with cloud-based customer information, then combine these data sources with publicly available assets to help give ecommerce brands the big picture.

When To Use GenAI vs. Discriminative AI

The application of AI in digital commerce isn’t new. Solutions such as discriminative AI have been a part of ecommerce since its inception.
 
Here’s a quick overview of both.

Generative AI Can Perform Many Functions

GenAI uses large models and massive amounts of training data. As a result, it can have millions, billions, or hundreds of billions of parameters.
 
These models are often used for new content generation or to power conversational user experience (UX) tools such as chatbots.
 
For example, GenAI tools are what make second-generation chatbots possible; their ability to parse not only context but, in many cases, sentiment makes them ideal for customer-facing portals.

Discriminative AI Performs Specific Functions

Discriminative AI uses much smaller data sets to perform specific functions. As noted by the Gartner report, discriminative AI excels at classifying data and predicting events. This makes it ideal for product categorization, customer segmentation, and predicting customer churn.
 
While GenAI is arguably more powerful given its ability to create net-new content, it is not inherently superior to discriminative AI. Instead, it’s about selecting the right tool for the job.

How GenAI Supports Ecommerce Workflows

GenAI supports rather than supplants current ecommerce workflows. But what does this look like in practice?
 
Consider SEO. While the rise of GEO has added a new layer of optimization, it doesn’t replace keywords — companies must still track relevant industry terms, keyword densities, and keyword trends. GenAI can help companies better understand consumer preferences and sentiment, in turn improving their overall SEO strategy.
 
Other areas for GenAI application include the following.

GenAI Enhances Content Enrichment

GenAI can improve current content by adding relevant details, such as personalized offers or tweaking content to better match customers’ preferred tone.

GenAI Helps With Guided Selling

Guided selling helps buyers find exactly what they’re looking for. It may take the form of product recommendations based on past purchases, online comparison tools that show key product differences, or agents equipped with real-time and historical customer data.
 
The advent of LLM-driven chatbots makes them an ideal starting point for guided selling. Customers can connect with chatbots to start the process, then move to agents at the decision-making stage.

GenAI Can Generate Quotes Quickly

For companies that generate product or service quotes, GenAI can be used to generate numbers that align with market value and account for customer priorities such as quality, speed, and total cost. Using AI tools, these quotes can be generated quickly and accurately, allowing brands to respond as customer needs evolve.

What’s Next for GenAI in Ecommerce?

The evolving role of GenAI in ecommerce doesn’t create an either/or scenario for organizations. While taking a humans vs. AI stance may seem useful in the moment, it only hurts brands over the long term.
 
Consider an ecommerce retailer looking to boost customer engagement. Taking a human-only approach requires substantial time and effort dedicated to collecting data, conducting analysis, and pinpointing trends.
 
Using AI exclusively, meanwhile, can result in marketing and outreach copy that lacks heart or character. It comes off as AI-generated, which can make customers feel like profit points rather than people.
 
To improve digital ecommerce, brands need to leverage the strengths of each approach. For example, companies can use GenAI to generate campaign ideas based on buyer demographics, then let marketing teams design the content.
 
Brands also need to determine if they’re best served by building or buying AI. According to the Boston Consulting Group, companies with limited amounts of proprietary data looking to create a unique customer experience should consider buying point solutions that provide specific functions.
 
Those with more data on hand, meanwhile, may want to build proprietary systems to handle product page content and personalized marketing outreach.

Going the Distance With GenAI

Ignoring GenAI won’t make it go away. GEO is now an essential part of search engine rankings, and advanced AI-driven chatbots let companies streamline customer service without sacrificing consumer satisfaction.
 
Instead, brands need to find a middle ground, one where humans and AI work in tandem across a structured ecommerce system. This is the key — it’s not about shifting to an exclusive AI approach or eschewing it for human expertise. Instead, it’s about using both sides of the sales and marketing coin to enhance existing processes and open the door to new ecommerce opportunities.
 
GenAI doesn’t replace human expertise, and human experience can’t compete with AI speed. Going the distance in ecommerce means taking a tag-team approach that plays to the strengths of both smart tools and intelligent humans.