How businesses can leverage Gen AI and get to personalized marketing content @ scale

By Gopi Vikranth

Gen AI ushers a new dawn of personalized content which can help brands revolutionize the way they engage their customers with deeply persuasive and relevant content. This can enable brands to maximize outcomes from data driven content which understands and anticipates the core needs of the customers.

There are additional benefits from a cost efficiency, time, and effort it takes to produce this form of personalized content at scale. At the heart of the matter is this question – can a company which has a fraction of the budget of Nike, produce content at similar scale and quality? Gen AI says yes. Moreover, large companies can provide an infinite variety and future marketers can test these variants against an objective and get to the right content for the right audience. For this to be realized, there are 4 core aspects that marketers need to consider.

Four aspects of n=1 personalized marketing

Let’s examine these four aspects and how specifically Gen AI/AI can help enterprises.

Content Insights and Intelligence

Gen AI large language models are extremely good at deciphering various aspects of content. This typically needs AI looking at existing data and content, extract intelligence from the content and infer insights

• Ingest data from existing system, type of files include — PDF, Word, Image, Video, Ads and audio files and extract information

Content/ Asset attributes

• Where applicable against existing content, we can also take in user input like prior objective, ROI or proforma information, Brand regulatory filings in some industries.
• Gen AI can also look at publicly available competitor information as a contrast mechanism, including offers, ads to understand what themes competitors are using.
• This is relatively easy to perform on a periodic basis to gain perspective into. For example, Gen AI LLM analyzed the below marketing copy and identified a number of details

Sample Email Copy with Content Tags generated using Gen AI

Content Variant Creation

These insights can in turn help marketers define what new content needs to be created and it can be approached in multiple ways.

1. The objective is similar to past campaign, then the Gen AI agent can look at the past campaign as seed input, take additional variation instructions via prompt and generate new content
2. The objective is new and a marketing agency has been asked to create a new campaign against it (this can be a few pieces of content instead of large collection). A Gen AI agent can take this new theme and scale it to a 100 or 1000 variants
3. The objective is net new, but Marketer wants to provide a prompt template against which gen AI will generate ideas for marketer to inspect and chose a select few which can act as the seed to generate additional content.

Gen AI agents can create a lot of variety of content — as of today they can create Emails, Messages, Tag lines, catchy Summary, Image + captions, Full marketing Copy (has title, image and some text), Banner Ads, Content Ads, 1 page white papers.

In future, GIFs or 10 second videos seem to be just on the horizon (example: RunwayML lip sync).

Image 1: Seed image; Images 2, 3, 4 are Gen AI generated Images

Content Prediction

Once content is created, AI can also help marketers assess the content via a predictive model (this is a bit more of a combination of classical AI and gen AI).

1. Assess similarity of a content piece to others and show the options to user on where all its similar and what all other content pieces is it related to. This is especially useful in regulated industries to assess the approval process (ex: healthcare MLR processes)
2. Predictive model which takes, content, target audience, any past data and predicts a score the new content available — ex: of the taglines generated, one option might have a higher resonance with target audience which is captured via the score.
3. Model can explain the content components — and the reasoning behind its scoring where needed.

Content Experimentation

Marketers can now experiment with the range of content available and learn from these findings. These experiments will also generate valuable data quickly to feed into the content insights and content generation process.

1. Marketers at large firms can set up some guidelines/campaign ideas and can rely on gen AI to generate the variations. Once brand aligned content is finalized, they can test these to see a much improved conversion across the board
2. As content creation costs go down, customer can expect the creative content that uniquely appeals to them based on who they are. This will further extend to a very personalized customer journey.
3. As customers respond to this personalized content, each such response in turn creates new psychographic data within enterprise which captures customer preferences and behavior. Such data about customers is currently non-existent in most enterprises. Such insight is highly valuable for future purposes as enterprises launch new products and offerings to their customers. This in turn provides better content intelligence and ideas for new content creation

All in all Gen AI is ushering a new era. Micro communities and personalized engagement will define the future of customer experience.

Future of Customer Experience (Images generated by Gen AI)

About the Author:

Gopi Vikranth is a Principal at ZS and leads AI SaaS Products and Platforms, Customer Experience Analytics and Personalization.

Gopi, along with Arun Shastri, hosts a podcast called “Reinventing Customer Experience”. Gopi and Arun chat with digital, analytics, technology and customer experience leaders from various industries and organizations to understand how they are engaging their customers. The podcast is available on Apple and Spotify.