Retail marketers can benefit from AI.
Much has been written about applying artificial intelligence (AI) and machine learning (ML) in complex retailer back end systems.
AI can enable retailers to optimize supply chains, better predict inventory needs, and implement personalization in user experiences. However, these can be significant, large-scale projects that take time to implement and test before they go live.
With an imminent code freeze, it’s more than likely that any major technology deployments won’t go live until after the holiday shopping peak. Despite the current focus on holiday campaigns, retail marketers can still modernize many tried and true tactics by using generative AI to streamline workflows.
High Impact Generative AI Strategies
Optimize email marketing performance. Many retailers see significant revenue uplift attributable to email; and there’s always a need to test and optimize every piece of the consumer experience in email, from subject lines to message content to the call to action. With that in mind, marketers can use generative AI to create test options for email.
For example, use a generative AI tool to rapidly produce multiple versions of subject lines, the call-to-action, body copy, etc., then conduct testing on performance. Send out test candidates to a segment of your email list, then monitor effects on the metric you’re optimizing for, and finally, deploy the winning subject line to the remaining audience after determining a winner.
However, because time is of the essence for holiday events such as Black Friday, you won’t necessarily have the luxury of time to test specific segments before deploying the full send. Use the data you’ve already gathered about successful tests to deploy the “most likely to succeed” version for time-sensitive emails.
Social proof and product descriptions. Many shoppers use reviews and star ratings to assess a product. Marketers can use AI to glean insights from existing reviews to better convey the benefits or positive features about a product up-front. It’s as simple as copying and pasting the text of positive reviews into a generative AI tool.
Then, ask the tool to summarize the entered text with a prompt such as, “You are an e-commerce merchandiser at [store name], tasked with increasing the sales conversion rate for this item. Read the attached existing item reviews and summarize or highlight what other buyers mentioned as their favorite features or what appealed to them the most.”
Elaborate on this prompt as much as you think it needs; typically, the better the context, the better its output. You can weave the outputs from these prompts into product descriptions, into your merchandise gift guides, into email messages, into onsite copy, or as social or search engine marketing ad copy.
Similarly, you can use AI to assess negative reviews and prepare your customer service team with advance notice of potential complaints, so they can proactively determine how to address unhappy customers.
Bonus tip: If your catalog has some product parity with other retailers, you may consider a similar exercise for competitors’ items to capture the most appealing product features to add into your description, especially if a given product is brand new and has no customer reviews yet.
General copywriting. Building on the concept above, generative AI tools are getting better at assessing inputs to answer questions for marketers, helping them to improve their writing skills. To that end, some other ways that marketers can leverage generative AI to improve customer-facing copy include:
- Tone analysis: Evaluate product marketing copy for tone, reading level, and emotional sentiment to ensure messaging aligns with brand voice and resonates with your core shopper.
- Readability assessments: Provide readability grades and feedback on copy to enhance clarity and simplification. Make sure your copy is not overly complicated for your target audience.
- Grammatical error checking: Proofread copy for spelling, grammar, punctuation issues.
Competitive analysis. Generative AI tools like ChatGPT and Claude.ai can review and compare product copy from major competitors in a particular e-commerce category to help streamline competitive analysis. For example, if a retailer sells fitness gear, the tool can be used to analyze product listing and other content from competitors like Nike, Under Armour, etc., and provide insights on messaging themes or styles. This can identify weaknesses and iterate on how your product and website are differentiated (or not).
Wrapping Up: Never Let Generative AI Have the Last Word
A word of caution – make a point to review AI-generated content to make sure that it makes sense. Generative AI still has a number of flaws, including well-publicized hallucinations. Carefully evaluate any copy the AI tool generates for accuracy, ensuring it passes the sniff test and matches your brand’s tone and approach.
As with all new technologies, taking “baby steps” with generative AI will allow brands to see how useful it can be for their business. AI promises to create new efficiencies by automating routine and creative processes and drive ongoing, incremental improvements to e-commerce marketing efforts – even after a code freeze goes into effect.