Agile Marketing Personas: Use AI to Stay Up to Date With Your Audience

Any good marketer knows that the key to a successful strategy is understanding your audience. Personas are one of the best tools we have to get to know our audiences better because they give us an opportunity to dig into our customers’ needs, motivations, and emotions.

But audiences change quickly, especially today. It’s hard for teams to keep personas updated because it takes so much time and effort to do deep research into audience behavior.

That’s where AI tools like ChatGPT and Bard come in. Both Bard and the latest update of ChatGPT, which now comes with a “Browse with Bing” option, can now access current information, which got us thinking: what if you could use AI to help keep your customer personas agile and responsive to new shifts and updates that could impact marketing performance?

The right prompts are key: those questions and queries will help you use Al to layer in fresh information from reviews and other sources and discover new trends and opportunities to ensure your campaigns connect with your target audience effectively. Let’s dig into the best practices around gathering new persona information from conversational AI and then check out three distinct examples of persona-focused prompts and outputs.

Setting a strong audience foundation with marketing personas

While overall audience insights are great for general market research, a persona-based approach helps marketers understand how different activations might work for subsections of their overall audience. Personas put a (made-up, often alliterative) face and a name to large audience groups, helping teams zero in on their exact target customer so they can tailor creative, messaging, and campaigns to those audiences. 

Marketing personas should be built from a mix of third-party research, quantitative data, and qualitative insights. Brands gather quantitative information from large data sets to get insight into their target audience’s preferences, behaviors, and purchasing patterns. Then marketers can add qualitative data from surveys, polls, interviews, and more to take their audience’s motivations, attitudes, and beliefs into account. 

As useful as personas are,  they can end up doing more harm than good if they aren’t up to date. Relying on older data and assumptions about your audience puts your brand at risk of misunderstanding your audience’s needs today.

But because it takes so long to do the requisite deep research and analysis of quantitative and qualitative customer data, many brands can only commit to updating personas annually or quarterly, or even less often–leaving the most relevant information on their audience’s media consumption, interests, and motivations on the table.

How AI can help you keep marketing personas agile

Generative AI tools can help you keep your personas up to date in between those large-scale, high-effort regular updates. You can use tools like ChatGPT and Bard to bring in new information and refresh your personas on a regular monthly cadence or when you’re kicking off a new campaign.

AI tools can help your team pull insights like:

• Customer sentiment: These tools can scour the internet for reviews of your products or services to gauge how your audience feels about your brand
• Seasonal changes: AI can keep an eye on seasonal changes in your online audience by quickly processing the available information and how it changes throughout the year
• New ideas: AI is great for brainstorming future personas and finding opportunity audiences
• Outdated assumptions: Since your robot helper doesn’t come into a persona-building session with any preconceptions, it can “think” outside the box and challenge outdated or incorrect ideas your team might have built into your current personas

It’s important to note that AI tools have a known problem when it comes to hallucinating information and surfacing false results. All information you source from gen AI chatbots should be checked against your core persona information; if something seems off or totally misaligned with your existing personas, dig a little deeper and do some digging to confirm that information.

Because of these risks, we recommend using these AI-generated persona insights inform creative and platform tests; these campaigns should come out of the 10% of your brand’s budget you should already be setting aside for testing. That’s where you need to be trying our new ideas and strategies to see what works for your brand before deciding to bring them into your business-as-usual (BAU) campaigns.

Prompting AI to get the insights your team needs

Now that you know what AI tools can do for your audience research, you need to take the right approach to make the most of them. If you’re using ChatGPT, start by making sure you’re using the right version: ChatGPT 4 will search the internet with Bing for information, but older versions (like 3.5) don’t incorporate more recent information.

Source: The Verge

We mentioned the importance of prompts in this process: you need to give clear and detailed instructions for this to work. Get specific about what you want the AI to do and what you expect as an output and you’re more likely to get more accurate responses.

You should break down complex tasks and give the AI time to think things through to improve the quality of the outputs you get. If you’re giving it multiple tasks at the same time, clearly separate different parts of your instructions into chunks with symbols or markers like “—” or “#” to help the AI understand what you’re looking for more clearly.

There are lots of different ways to prompt these tools: you can write from the point of view of a marketer looking for insights from the AI or ask the AI to act as your persona to simulate a conversation. The RGIO (Role, Goal, Input, Outcome) formula can help make sure you get the output you want.

If you’re looking for market research, for example, an RGIO approach could look like this:

• Role: You are a creative strategist.
• Goal: Your goal is to assist your client in understanding consumer preferences.
• Input: You analyze ad data from Meta that explains how consumers are interacting with a specific ad.
• Output: Provide a detailed analysis highlighting emerging trends and preferences based on the data from each ad.

If you’re not satisfied by the initial output, don’t stop: continue querying the AI to get the result you’re looking for by pointing out what doesn’t work in the response, asking to get more specific in a certain area, or rewording your request to be more specific or clear.

Once you’ve got the output you’re looking for, remember to have human team members double-check any research the AI has done for you. Always get a second pair of eyes on these responses before using them in your marketing.

Examples of marketing persona prompts and outputs from generative AI

No two prompts will be the same for any two scenarios, but there are some general formats you can follow to get the responses you’re looking for. To help your team create the best possible marketing personas, here are some examples of possible prompts and outputs from AI tools.

Example 1: Prompt as marketing manager looking for new insights on an emerging persona for a legacy retail brand

Input:

I am the marketing manager for StrideActive Footwear. We specialize in high-performance shoes designed for active individuals aged 18 to 45. Our company focuses on crafting innovative, durable, and stylish footwear tailored to support various activities, from running and hiking to gym workouts and everyday wear. We prioritize comfort, functionality, and style for both men and women who lead active lifestyles.

You are a persona research specialist. Your role is to assist me in understanding our new audience. You review data based on the persona that I explain to you to provide a detailed analysis highlighting the habits, pain points, and selling points of our new audience. 

The persona is men aged 18 to 25, residing in Colorado, who are deeply passionate about trail running. They are adventurous spirits who find joy and fulfillment in exploring the rugged terrains of Colorado’s trails and are seeking footwear with both high performance and durability. They are highly engaged within their local trail running community, often participating in races and actively sharing their experiences and gear preferences on social media platforms.

Output: 

Habits:

• Active Lifestyle: They prioritize outdoor activities, especially trail running, and engage in it regularly, seeking new challenges and routes.
• Community Engagement: Actively involved in the local trail running community, participating in events, races, and sharing experiences through social media platforms like Instagram, Strava, or dedicated forums.
• Gear Enthusiasts: Keen on researching and investing in high-quality gear that enhances their performance and endurance on trails, including footwear, apparel, and equipment.
• Explorers and Adventurers: They value experiences and seek rugged, challenging terrains to explore, requiring footwear that provides excellent traction, support, and durability.

Pain Points:

• Durability and Performance: Concerns about the durability of footwear due to frequent use in rugged terrains. They seek shoes that can withstand harsh conditions without compromising performance.
• Comfort and Support: Need for comfortable shoes that provide adequate support and cushioning, ensuring their feet stay protected during long runs.
• Style and Aesthetics: Desire for footwear that not only performs well but also looks stylish and aligns with their personal fashion sense.
• Price vs. Quality: Balancing the cost of high-quality, durable footwear against their budget is a consideration, as they prioritize value for money.

Selling Points:

• Performance-Oriented Design: Highlighting shoe features that improve performance on trails, such as superior traction, durable outsoles, and lightweight construction.
• Durability and Reliability: Emphasizing the shoe’s construction materials and design that guarantee durability, reinforced toe caps, protective features against rocks, and abrasion-resistant elements.
• Comfort Technology: Showcasing advanced cushioning systems, breathable materials, and supportive midsoles designed for long-distance runs, ensuring comfort throughout the journey.
• Community Engagement and Validation: Leveraging user-generated content, testimonials, and reviews from local trail runners who have had positive experiences with the shoes to build trust and credibility within the community.

Example 2: Prompt queries as an individual persona for a finance brand

Input: 

I am a marketing manager for HarborGate Financial, a financial institution that is focused on integrity, customer empowerment through knowledge, and customer-centric innovation. 

Your role is to assist me by role-playing as one of our target personas. Your goal is to explain any pain points of our business or any features that help you achieve your goals. I will explain the persona that you are to play below and you will engage in a conversation with me and answer my questions about the company. 

Persona: 

Amanda Johnson – The Young Professional

• Background: Amanda is a recent graduate in her mid-20s, eager to start her career in marketing. She’s enthusiastic about learning to manage her finances effectively, seeking guidance on building a savings plan, managing debt, and exploring investment opportunities.
• Needs & Goals: Amanda is looking for guidance and mentorship to navigate financial complexities. She values educational resources and seeks advice on building a solid financial foundation for her future.

Output:

Great, I’m stepping into the shoes of Amanda Johnson, the young professional eager to delve into the world of finances! How can HarborGate Financial help me out?

Input:

Hi Amanda! Can you explain any pain points you have encountered with HarborFinancial as you seek to achieve your goals?

Output:

Hey there! Well, navigating the financial world can be a bit overwhelming, especially as someone starting out. One thing I’ve noticed is that sometimes the information available can be a bit too technical or complex. I’m eager to learn about savings, managing debt, and exploring investment opportunities, but I often find it challenging to get clear, understandable guidance that fits my level of understanding. So, having resources or guidance tailored to someone just starting out like me would be incredibly helpful.

Input: 

I’m sorry that the information is too complex. We can fix that. Outside of the issues with how we present our information, can you explain the benefits you have experienced while banking with us?

Output: 

Absolutely! I’ve really appreciated the emphasis on integrity and customer empowerment through knowledge that HarborGate Financial stands by. The commitment to these values gives me a sense of trust and security when it comes to managing my finances. Also, the customer-centric innovation aspect is intriguing—I’m curious to know more about how this innovation translates into practical tools or resources that could help someone like me understand financial concepts better or manage my money more effectively. 

Example 3: Prompt to conduct product research for a retail brand

Input:

I am a product engineer for a company that creates outdoor gear for individuals who live a very active lifestyle and have extensive experience with camping and outdoor activities. 

We are creating a new cooler that will act as our flagship product. Your role is to assist me in creating this new product by using data on other coolers meant for camping and rugged outdoor activities. Your goal is to explain any specific features we absolutely need to add to our product as well as explain any features we should stay away from adding. I have explained the product that we are creating above. Explain how we should construct our cooler to appeal to the audience mentioned above. 

Output:

Creating a flagship cooler for the outdoor enthusiasts you’re targeting requires meticulous attention to detail and a deep understanding of their needs and preferences. Here are some key features and considerations based on data and insights from coolers designed for rugged outdoor activities:

Features to Incorporate:

Durability is Paramount:

• Materials: Opt for high-grade, rotomolded polyethylene or similar materials known for durability. They should withstand rough handling, impacts, and harsh weather conditions without compromising performance.
• Insulation: Incorporate thick insulation (at least 2 inches) to ensure maximum ice retention, especially for extended outdoor trips.

Portability and Handling:

• Size and Weight: Balance capacity with portability. Aim for a spacious interior while keeping the weight manageable for easy transportation.
• Handles and Straps: Ergonomic, reinforced handles or straps are vital for comfortable carrying, even when the cooler is fully loaded. Consider options for single-person or team carries.
• Wheels (Optional): For larger coolers, durable, terrain-conquering wheels can enhance mobility across rugged landscapes.

Features to Avoid:

Fragile Components:

• Thin Walls or Weak Seams: Avoid using materials with thin walls or weak seams that may crack or split under pressure. Thick, reinforced walls with strong welding or bonding techniques are crucial for longevity.
• Fragile Handles: Stay away from handles made of brittle plastics or weak attachment points. Instead, choose handles constructed from sturdy materials like rubberized grips with reinforced anchor points for maximum durability.

Complexity without Utility:

• Overcomplicated Latching Mechanisms: Avoid intricate latching systems that might confuse users or require excessive effort to operate. Opt for straightforward, foolproof mechanisms that provide a secure seal effortlessly.
• Over-Engineering: Avoid over-engineering aspects that don’t directly contribute to the cooler’s core purpose, like incorporating excessive structural reinforcements that add weight without substantial benefits to durability.

Looking for more AI insights? Check out our white paper The Game Changing Power of AI to learn everything you need to know about AI in search.