If headlines and hype are an indication, artificial intelligence could transform ecommerce marketing.
AI-based technology can now predict consumer needs, target ads with greater accuracy than a human, personalize content for each user, chat with online shoppers, optimize product recommendations, analyze customer data for promotional insights, write marketing copy, and much more.
AI tools such as Jasper and OpenAI’s ChatGPT and DALL-E 2 have dominated my recent conversations with a select group of four businesses — from an ecommerce company with $1 million in annual sales to a SaaS business with close $30 million.
The sample size, to be sure, is small. But my conversations with those owners and managers followed a similar pattern.
First, I ask about the business’s marketing plans. The answers inevitably lead to how the company has used an AI tool. A marketer mentions how amazing the tool is before saying something like, “But I did have to regenerate some of the answers. It is so important to ask the questions in the right way.”
Hold that thought about the right questions and consider three examples of how these folks have used AI.
Automated email marketing messages are table stakes for many ecommerce operations. A prospect who signs up for a newsletter, abandons a cart, or performs another trackable event triggers an automated email sequence.
The messages in that sequence might combine human-written promotional copy and dynamically-inserted product info.
One of the marketers in my sample addressed his “welcome” sequence to new subscribers.
His process is this. He asked Jasper to identify why consumers might buy the company’s products and to generate several variations of the email messages based on those reasons. Next, he tested the variations with customers to learn which produced the most sales.
Certainly a human could have written the variations. But it’s doubtful, as this marketer pointed out, owing to time constraints and disinterest. Assigning the task to an AI tool is much more efficient.
The second example is unrelated to sales. A marketing manager used ChatGPT to help complete his own annual review.
This individual meticulously provided info to ChatGPT about how the company graded employees. He also shared the HR team’s skill matrix for identifying candidates for a promotion.
The manager then asked ChatGPT to formulate questions to help identify deserving candidates. He composed answers to those questions and instructed ChatGPT to grade him relative to the skill matrix.
Eventually, he fed the company’s self-review questionnaire to ChatGPT, which generated responses that made him an obvious choice for a promotion.
Then there was the marketer who asked ChatGPT to help improve his campaign forecasting. The marketer was responsible for estimating how a multichannel promotional campaign — affiliates, pay-per-click ads, email, on-site banners — would perform at various spending levels.
This individual asked ChatGPT to identify the best ways to build the forecast model and compose the Google Sheet formulas for that model.
Businesses seeking AI efficiencies will need similar specialists. Call them “AI operators.”
Observers have predicted AI uses for ecommerce businesses to include chat for conversational selling and complex data analysis for personalization. Both appeared in “Emerging Roles of Artificial Intelligence in Ecommerce,” a 2020 paper by Vishal Dineshkumar Soni of Campbellsville University in Kentucky and published in the International Journal of Trend in Scientific Research and Development.
Ecommerce companies could employ data scientists or developers to create AI-driven marketing tools. But using those tools presumably requires expertise, too.
Should companies hire technicians to optimize AI? I asked ChatGPT that question. Here’s the response:
To be successful using AI to improve marketing, ecommerce companies may need to hire specialists, depending on the complexity of the AI tools they are using.
Implementing AI in ecommerce marketing can be done in-house by data scientists, software engineers, and marketing specialists who have experience working with AI tools. These team members will be responsible for designing and implementing AI models, as well as analyzing and interpreting the results.
However, if the company is using more advanced or specialized AI tools, it may need to hire outside experts with specific knowledge and experience in those areas. For example, if a company wants to use advanced image recognition AI, it may need to hire an expert in computer vision.
Additionally, it is important for the company to have a dedicated team that can manage, maintain and monitor the performance of the AI system.
Overall, the need for specialists will depend on the complexity and scope of the AI tools being used, and the company’s existing capabilities in terms of data science and AI expertise.