Using AI to automate plain language optimization of member communications

In a recent article, I discussed the value of conveying the right sentiment in member communications and how AI can help. However, there’s another critical element that determines whether your communications resonate and drive action: the clarity of your content.

The hidden cost of complex communications

While credit unions strive to be accurate, transparent and member-focused, many send out member communications—letters, notices, disclosures— that are very difficult for the average person to understand. They often contain jargon like “collateral” and “overdraft,” which may be unfamiliar to those with low financial literacy. Long sentences, dense paragraphs, legalese and overly formal phrasing are also common and create structural barriers that make it harder for members to understand and act on the information presented.

This is especially concerning when you consider that over half of American adults read at or below a sixth-grade level, 22% don’t speak English at home, and 25 million have limited English proficiency.

This disconnect can have real consequences. Members may make poor decisions based on partial understanding, miss out on beneficial financial opportunities, or disengage altogether due to confusion. Operationally, it can lead to increased call center volume from frustrated members seeking clarification, increasing costs and straining internal resources.

Plain language as a framework for clarity

Plain language requires more than just using simpler words, it’s a structured approach to creating communications that are clear, accessible, and actionable for nearly all readers. The International Organization for Standardization’s (ISO) plain language guidelines emphasize the use of familiar terms, shorter sentences and paragraphs, headings, and direct language through the use of active voice and personal pronouns.

For example, instead of saying, “Your share draft account has been assessed an overdraft fee due to insufficient available balance,” a plain language approach would be: “We charged you a fee because you didn’t have enough money in your checking account to cover a recent payment.”

The standard also provides guidance on how to structure communications—for instance, placing the most important information first and using clear headings to guide the reader through longer or more complex content. The benefits of using plain language standards for customer-facing communications are well-documented. A 2022 study found that businesses that adopt plain-language writing had lower rates of confusion and higher trust within their customer-base, as well as lower overall support costs.

AI as a catalyst for plain language adoption

For most credit unions, applying plain language principles across all member communications would be an overwhelming challenge. Manually rewriting every letter, notice, and statement would require time and resources that most teams simply don’t have. Thankfully, credit unions now have the opportunity to automate plain language optimization using AI.

AI models are now sophisticated enough to analyze entire communication libraries, flag content that falls below defined readability thresholds, and suggest rewrites aligned with plain language standards– all in a fraction of the time it would take a human.

AI can also be leveraged proactively, analyzing new content as it’s authored and recommending improvements before it’s published. This enables credit unions to move beyond one-time content clean-up efforts and maintain plain language standards over the long-term.

Lastly, while individual writers may interpret plain language principles differently, AI systems apply them in a consistent way across all communications, so members receive clear, effective messaging, no matter which team or department wrote it.

Implementing AI safely and effectively

While AI offers unprecedented opportunities for plain language adoption, the path to implementation matters enormously. Many teams may start with standalone tools like ChatGPT, but these tools come with challenges and risks particularly when applying them to sensitive member communications.

To get quality results from these conversational tools, users must provide detailed, well-structured prompts that guide the response from the AI. This requires a level of prompt engineering expertise that most communications teams don’t have. Given the complexity of plain language writing and the precision required for regulated content, it’s often difficult to generate desired results, especially on the first try. In many cases, the time spent refining prompts can exceed the time it would take to rewrite the content manually.

More importantly, credit unions should be cautious about pushing sensitive member or corporate content into open AI platforms without controls and guardrails in place. Copying and pasting content out of communications management systems into public AI models is fraught with risk, not to mention inefficiencies.

Modern customer communications management systems offer credit unions the opportunity to leverage generative AI in a controlled way, from within the same secure environment where their member communications are managed. These platforms enable credit unions to take advantage of the speed and efficiency of AI language while maintaining a high level of data security, regulatory compliance, and operational control. Some solutions now offer pre-built prompts, specifically designed for plain language optimization, saving users from wasting time crafting their own prompts while ensuring consistent results.

Plain language is a win-win

By adopting plain language across all member communications, credit unions strengthen their core value proposition of being focused on what’s good for their members. Members make better financial decisions when they clearly understand their options. Call center volume and costs decrease when communications answer questions instead of creating them. Most importantly, the cooperative principles at the heart of credit union membership are reinforced when members feel understood and supported by their institution.

Article originally posted on CU Insight