Simple Steps to Save Time and Leverage AI

Using AI can be a bit frustrating at times. So much is said about its potential, but it's often hard to figure out where to spend your effort to get the return everyone is talking about. For each area of application, there seems to be a multitude of options, which can be overwhelming. To help cut through the noise, we'll use a realistic example to show how you might use these tools effectively.

To get more familiar with making use of these tools, it's good to take some baby steps. Use simple but real examples in practice to build your confidence and understanding. We’ll demonstrate this by developing a story around the risk advice process, which might be used in client materials or on your website.

This may look like we’re going a little overboard – but the idea isn’t to suggest you carry out each of the steps, or use all these platforms – but just to provide you with some ideas and perspectives on making a start with using them.

Key Points We'll Cover:

1. Grabbing an example to use – in this case, articulating the risk advice process for clients.
2. Looking at the basics – starting with a rough outline.
3. Using different AI tools to enhance and refine the content.
4. Bringing these outputs together for a better constructed and formatted end result.
5. Modifying or leveraging this content for various uses – client communication, website content, or referral sources.

We’ll work through the use of a few tools, which can help contrast and compare the differences. For demonstration purposes, we've used paid versions of these tools (ChatGPT Plus, Anthropic Claude Pro, and Perplexity.ai Pro). Free versions are available, but they generally offer fewer features and capabilities, and their options for information protection are also lacking.

Step-by-Step Example: Crafting Your Advice Process Story

Step 1: Grabbing an Example

For this demonstration, we’ll start with a basic outline of the advice process. The scenario here is about crafting a simple overview of the process to use with clients, to help them understand the process you’ll (as an adviser) be walking them through. You could start this process with AI assistance, but I suggest starting with a simple brain dump of your own thoughts – as notes, steps, or current promotional material you have on hand (which you have rights to use).

Step 2: Looking at the Basics

We'll jot down this rough idea, giving us a foundation to build on. This simple outline helps us identify the key steps in our process. The details below are a quick summary of the primary steps (my own brain dump). Your approach may very well differ – but as noted, this is simply for demonstration purposes:

• Contact: This is where we gather some basic information from you to get started.
• First Meeting: We discuss your goals and needs to understand your situation better.
• Research: We take the time to work through what's needed, explore the options, identify potential hurdles based on your situation, and evaluate product options.
• Your Advice: We present you with a detailed summary of options and our recommendation tailored to your needs.
• Implementation: We assist with submitting the necessary paperwork to the insurers.
• Underwriting: There is still some work to do at this stage. We work to ensure you get the best terms of cover.
• Policy Issue and Servicing: Once everything is in place, we begin the servicing process. We work with you to handle any changes needed in the future and are here to assist if a claim arises.

Step 3: Enhancing with AI Tools

ChatGPT: Testing the Output

We'll use ChatGPT (model 4o in this example) to turn this outline - well actually only the key headings - into a more polished, client-friendly guide.

Here's a sample prompt:

Prompt: "I need help to create a client-friendly guide for our financial advice process, and more specifically Risk Insurance advice. Here’s a rough outline as a starting point: Initial Contact, First Meeting, Research, Your Advice, Implementation, Underwriting, Policy Issue, and Servicing. Please expand on each step to make it engaging and easy to understand for clients."

As you can see, this is a pretty good start. You could use this as a base and further refine it. In this case though, we're going to compare the different output (from different platforms) based on the same prompt.

Anthropic Claude: Testing the Output

Next, we’ll take a look at the output from Anthropic Claude using their Sonnet 3.5 model:

Prompt: "I need help to create a client-friendly guide for our financial advice process, and more specifically Risk Insurance advice. Here’s a rough outline as a starting point: Initial Contact, First Meeting, Research, Your Advice, Implementation, Underwriting, Policy Issue, and Servicing. Please expand on each step to make it engaging and easy to understand for clients."

The output from Claude is good, although it’s missing a little context (Australian spelling and the market), but it’s otherwise pretty good in tone and style.

Perplexity: Testing the Output

Lastly, we’ll examine the output from Perplexity. This tool isn’t necessarily the best for this particular demonstration, but it does provide some useful additional resources and help on many topics. This output is using the Pro version:

Prompt: "I need help to create a client-friendly guide for our financial advice process, and more specifically Risk Insurance advice. Here’s a rough outline as a starting point: Initial Contact, First Meeting, Research, Your Advice, Implementation, Underwriting, Policy Issue, and Servicing. Please expand on each step to make it engaging and easy to understand for clients."

And the close-out text produced in this one wasn't bad either:

Step 4: Bringing It All Together

As you see from the above examples, the quality, appropriate context, and style of output can differ dramatically between the various tools. This is impacted by a number of factors, such as the quality and detail in the prompt, as well as training on the model itself. ‘Pro’ versions of the tools will generally provide improved results.

The three tools used here might be simply summarised:

• ChatGPT: A good all-round tool, dramatically improved with efforts to help it learn (pre-training).
• Anthropic Claude: Great off the bat with limited training, generally does a better job at tone and style of writing.
• Perplexity: Best for pure research and factual inputs.

None of these tools are targeted for pure ‘writing’, but they each can be highly useful for written content regardless. And if you’re keen to minimise the use of too many apps and platforms, they might just do the job okay. Depending on your purposes and willingness to venture into using varying tools, consolidating content from multiple tools can be quite valuable.

(take a look at the AI Essentials module on AI tools for writing here)

Now, we'll combine the output into a more constructed and formatted end result. This involves ensuring consistency in tone, style, and structure. To do this, we’ll use Anthropic Claude, mainly due to its quality of written content.

Prompt: "I need help to create a client-friendly guide for our financial advice process, and more specifically Risk Insurance advice. I am a Financial Adviser in Australia. I will provide you with my outline of the core steps and a little detail on each of these steps. I’d like you to retain this core detail but expand and refine the steps based on the further information and notes that follow. Provide the output in the form of a table.”

I then followed this request (within the same prompt) - the original Advice steps I'd put together, as well as each of the output/content blocks retrieved from ChatGPT, Claude, and Perplexity (noted above).

This provided the output you see below (pre-formatting). I did not adjust any of the actual content, this is as produced.

I then took a few minutes to copy this output into a Microsoft word table, and gave it a little formatting:

This is a simple view of the quite comprehensive detail we managed to gather from the various platforms. This might be good to use for a simple graphic, but you may also want to create a more substantial promotional document.

The good news is, this chat has now been well-informed with all this information – so we’re ready to modify the content further as we like. You can return to this ‘conversation thread’ later on, and add to it, or request other output. The next step will draw out some of the variations you might want for the output.

Step 5: Modifying or Leveraging the Content

Now, let's look at how you might modify this content for different purposes or audiences:

For Client Communication:

Prompt: "Provide a lead-in paragraph to be used just prior to the table or graphic."

For Website Content:

Prompt: "I need a paragraph to provide a very short overview of the process, which encourages the viewer to click through for more information and detail on the Risk advice process, for use on my website."

For Referral Sources:

Prompt: "I need a version of the lead-in for use with referral sources, helping them understand the detailed but helpful process a client of theirs will be taken through (by our practice), which will be followed by the detailed process table or graphic."

Conclusion

By using these AI tools, you can efficiently create polished, professional content tailored to various audiences. Starting with a simple outline and refining it through strategic AI applications not only saves time but also enhances the quality and clarity of your communications. These tools are designed to assist and amplify your efforts, making your efforts more efficient and impactful. And now that you have this content that's easy to modify, the creation of graphics or flowcharts for marketing materials is far simpler.

While these examples demonstrate the potential of multiple platforms, you don't need to use them all. Instead, use this guide to understand how different tools can be leveraged based on your specific needs. For more information on the use of AI tools, check out our growing AI Essentials program on Risk Hub. We're continually adding new tools and resources to support you.

Best regards,

Marc Fabris

Risk Hub Founder

About Risk Hub

Risk Hub is a central source of tools and resources designed to enhance the delivery of risk advice. Our platform provides quick access to insurer information, a comprehensive service provider directory, and essential resources like PDS links. With growing educational support, including programs like AI Essentials, we're committed to empowering advisers with the tools and guidance that make a real difference in their practice and for their clients.