Can ChatGPT really develop a digital product strategy?
ChatGPT (Generative Pre-trained Transformer) is a cutting-edge technology that has the potential to revolutionize the way we work. By harnessing the power of advanced machine learning algorithms, ChatGPT can assist with a wide range of tasks, from writing and editing to data analysis and customer service. However, as with any new technology, it’s important for people to be aware of its capabilities and limitations in order to use it effectively in their work. Being aware of ChatGPT can help professionals to automate repetitive tasks, increase efficiency and productivity, and ultimately, stay ahead in their field.
I’m going to be honest, I didn’t even write that first paragraph, I had ChatGPT do it for me. I feel a bit intimidated. How can my skills match that of the machine? I better make this second paragraph a good one! I wonder if that also means ChatGPT co-authored this blog post — I hope I don’t need to pay them any royalties. Thinking about it, ChatGPT isn’t that personable of a title, so for the remainder of this blog post I’ll refer to them as Gavin (for no particular reason, other than for the way they communicate, they deserve a proper name). If you haven’t met Gavin yet, you can check them out here: https://chat.openai.com/chat/. The potential impact of having a friend like this could be huge on a range of tasks that we normally have to slave over or at least spend a whole bunch of time researching, then thinking, and finally creating.
With my product hat on, I wanted to see how Gavin could support me in coming up with a digital product strategy for a fictitious health insurance company [check out the transcript at the bottom of this blog post].
How I used it
I wasn’t trying to do anything particularly deep. I was just pretending that I was working for a health insurance company and then asking Gavin questions like a product manager might, to help develop an understanding of the current trends in the industry and get some ideas that I could leverage in developing my digital product strategy.
My observations
It’s hard not to be impressed. Gavin will remember the context of the conversation, they have no difficulty understanding my questions and can clearly articulate their responses. There were probably only a couple of questions where Gavin struggled, for example, I was looking for specific numbers that would identify how big my potential target customer segment was — but I was able to work around that by asking more specific questions i.e. getting an age breakdown and then a health status breakdown separately — they obviously didn’t have access to a data source that had the combined information. At the moment, Gavin is only being trained on data up to 2021, so the information is not the latest, but it’s certainly enough to get you started. But it was easy to tweak the responses to make things of greater relevance, you just have to ask them.
The results
Obviously, you don’t want to be developing strategy in isolation and Gavin won’t replace the need for doing customer research or getting good data to analyse. But I can see key advantages and time savings for how you might do your research on competitors or the macro environment and trends. In the space of about 10 minutes, Gavin gave me a lot of good ideas to consider for my strategy which I could then take back to the team as well as the associated risks with implementing those ideas. I liked the idea of the personalised pricing strategy and they were even able to give a rough guide to the type of data I would need to be able to implement that strategy and how to make it profitable.
But don’t take my word for it — I’ll let Gavin give their two cents on how they can best help with a product strategy:
ChatGPT can assist in product strategy development by providing insights into customer needs and preferences, market trends and competitor analysis, as well as by helping to identify potential risks and opportunities. Additionally, ChatGPT can assist in the development of a product roadmap, the prioritisation of features, and the generation of ideas for new products or features. ChatGPT can also assist in the development of a business case, by providing insights on market size, potential revenue and costs, and by conducting simulations and forecasting. Furthermore, ChatGPT can help with natural language communication and documentation of the product strategy.
[I’m just disappointed they weren’t referring to themselves as Gavin]
At this point, even though I have better insights thanks to Gavin, I’d still have a lot of outstanding assumptions around my strategy. The way we’d typically address this at DiUS would be to start running some experiments, and if I did go down this data and AI-led approach, it would need a lot more validation. I really can’t see any way Gavin could help with this, other than to provide some insights into similar experiments that might have been undertaken. Things we’d need to address would include concerns about whether customers would trust us with access to all their information, how best we could provide transparency in how we determined their premium, and what range of premiums would resonate with our target customer segment.
Another important aspect of a strategy is to get buy-in from those within the company, so I asked Gavin to give me a paragraph on what I should put into my business case:
Personalized pricing, which uses data and machine learning algorithms to set different prices for different customers based on their individual characteristics, can provide significant benefits for your health insurance company. With personalized pricing, your company can better price its products and services, which can help to improve profitability and competitiveness. Personalized pricing can also help to attract and retain low-risk customers, by offering them more competitive pricing, which can help to increase customer loyalty and reduce customer churn. Additionally, by identifying patterns and predicting future claims, your company can proactively target higher-risk customers with personalized products and services, such as disease management programs. This can help to reduce costs associated with claims and improve the overall health outcomes for your customers. Overall, personalized pricing can help your company to improve its financial performance and customer satisfaction, which can be key factors for success in the Australian health insurance market.
But I feel it’s important to reinforce the point that we shouldn’t be developing a strategy in isolation, so I asked Gavin to give me some thoughts on that too:
A collaborative approach to product strategy development brings different perspectives and expertise to the table, which is critical for aligning the strategy with customer needs, market conditions and business goals. By involving customers in the development process and gathering feedback on existing products, teams can better understand customer needs and preferences. This will help ensure that the product strategy is aligned with customer needs and that the product will be more likely to be successful in the market. Additionally, by involving cross-functional teams and aligning objectives with overall business goals, product strategy will be more likely to align with market conditions and support the overall success of the business.Moreover, a collaborative approach allows for better identification and management of risks, improved efficiency and effectiveness, fosters innovation and adaptability and increases commitment and buy-in from stakeholders. By working together, teams can better coordinate efforts and make better use of available resources. Additionally, by encouraging open communication and the sharing of ideas, a collaborative approach can lead to more innovative and creative solutions. This can be particularly important in a fast-paced and ever-changing market.
Gavin’s just popped out for a few minutes, so I’m going to have to close with my own thoughts.
It’s hard not to be in awe (and maybe just a little bit nervous) of the technology behind Gavin (who’s still learning). The case I looked at here was around developing a digital product strategy, but Gavin could just as easily help out with copywriting or even writing code.
Actually, Gavin’s just popped back in so I asked them how they would like to be perceived by us — it seemed like an appropriate way to finish up:
I would like to be viewed as a reliable and helpful tool for providing accurate and informative responses.
I can’t argue with that!
The transcript
My company sells health insurance. I am a digital product manager and need to understand the trends that might affect our customers in the next few years. Can you suggest what trends I might be able to relevant for the insurance industry?
There are a number of trends that may have an impact on the health insurance industry in the next few years. Here are a few:
Telemedicine: With the rise of virtual consultations, telemedicine is becoming increasingly popular. Customers may expect more flexibility in how they access healthcare services and may demand more options for virtual consultations.
Consumer-driven health: As healthcare costs continue to rise, consumers are becoming more cost-conscious and are looking for ways to take more control of their healthcare spending. This may lead to a shift towards high-deductible plans and increased use of health savings accounts.
Wearable technology: Wearable devices, such as fitness trackers, are becoming more common and are helping consumers to better understand and manage their health. Insurance companies may look for ways to incorporate data from these devices into their products and services.
Data and Analytics: with the increase of data being generated from electronic health records and connected devices, the insurance companies are looking for ways to use this data for risk assessment, fraud detection and personalized pricing.
Artificial Intelligence and Machine Learning: As AI and ML technology advances, insurance companies may look for ways to use these technologies to improve underwriting, claims processing, and customer service.
Digital Experience: With the rise of digital technologies, customers expect a seamless and personalized experience when interacting with insurance companies. This includes easy-to-use mobile apps, quick and responsive customer service, and the ability to access and manage policy information and claims online.
These are just a few examples, and trends can vary depending on the region, culture, and economy. However, keeping an eye on these trends may help you understand the direction of the industry and identify opportunities for your company to innovate and stay ahead of the curve.
Can you give some examples of health insurers using AI or ML to create innovative products?
There are many examples of health insurers using AI and ML to create innovative products. Here are a few:
Predictive modeling: Many health insurers are using predictive modeling to analyze data from various sources, including claims data, medical records, and demographic information, to identify patterns and predict which customers are most likely to have certain health conditions. This can help insurers to proactively target those customers with personalized products and services, such as disease management programs.
Fraud detection: Health insurers are using AI and ML to analyze claims data and identify patterns that may indicate fraudulent activity. This can help insurers to quickly and effectively identify and investigate potential fraud, which can ultimately help to reduce costs and protect policyholders.
Underwriting: some insurers are using AI to analyze data from electronic health records and other sources to quickly and accurately assess the risk profile of potential policyholders. This can improve the underwriting process, making it faster and more efficient, and can help insurers to offer more competitive pricing to customers.
Personalized pricing: Some insurers are using machine learning to personalize pricing for customers based on their individual health risks and lifestyle choices. This can help insurers to offer more competitive pricing to low-risk customers while still maintaining profitability.
Chatbots and virtual assistants: Some insurers are using AI-powered chatbots and virtual assistants to improve customer service and make it easier for customers to access information and manage their policies. This can help insurers to improve customer satisfaction and reduce costs associated with call centers.
These are just a few examples, and the use of AI and ML in the health insurance industry is constantly evolving. However, by keeping an eye on these and other examples, you may be able to identify new opportunities for your company to use AI and ML to create innovative products and services that meet the needs of your customers.
Which specific companies are using these approaches?
There are a number of health insurance companies that are using AI and ML to create innovative products and services. Here are a few examples:
UnitedHealthcare, a subsidiary of UnitedHealth Group, is using predictive modeling to identify customers who are at risk of developing chronic conditions, such as diabetes and heart disease. The company then works with those customers to develop personalized care plans to help them manage their health and reduce costs.
Humana, a health insurer based in the United States, is using machine learning to personalize pricing for customers based on their individual health risks and lifestyle choices. The company is also using AI-powered chatbots to improve customer service and make it easier for customers to access information and manage their policies.
John Hancock, the US-based life insurer, is using AI-driven data analytics to improve underwriting for life and long-term care insurance.
AXA, the French insurer, has integrated AI into its claims process to help reduce the time and costs associated with processing claims.
MetLife is using AI to automate underwriting and claims process, it’s also working on using AI for personalized customer engagement.
These are just a few examples, and there are many other health insurance companies around the world that are using AI and ML to create innovative products and services. Additionally, many insurtech companies, who focus on innovation and technology, use AI and ML as a core part of their products.
Any Australian companies using AI in health insurance?
Yes, there are several Australian health insurance companies that are using AI to create innovative products and services. Here are a few examples:
Medibank, an Australian health insurer, is using machine learning to analyze claims data and identify patterns that may indicate fraudulent activity. The company is also using AI-powered chatbots to improve customer service and make it easier for customers to access information and manage their policies.
nib, an Australian health insurer, has integrated AI into its claims process to help reduce the time and costs associated with processing claims.
Bupa, an Australian health insurer, is using AI-powered predictive modeling to help identify customers who are at risk of developing chronic conditions, such as diabetes and heart disease.
HCF, one of the largest not-for-profit health fund in Australia, is developing a virtual health assistant using AI and natural language processing to help customers understand and navigate the health system.
NIA, an Australian insurtech, is using AI to automate the underwriting process for life insurance, making it faster and more efficient for customers.
These are just a few examples and many more Australian companies are investing in AI and ML to gain competitive advantages and offer better products and services to their customers.
What is the market share for health insurers in Australia?
The health insurance market in Australia is competitive, with a number of large players and smaller companies vying for market share. According to the Australian Prudential Regulation Authority (APRA), as of 2021, the top five health insurance providers in Australia by market share are:
Medibank Private: ~30%
Bupa: ~28%
nib: ~9%
Australian Unity: ~5%
HCF: ~5%
This is based on the number of policies held by each provider. It’s important to note that these numbers can change over time as the market evolves and companies merge or acquire other companies. Additionally, there are smaller health insurers that hold a smaller market share, but still offering competitive products and services.
It’s also important to note that the Australian market also has a significant portion of the population who are covered by the government-funded Medicare, which covers most of the citizens and provides a “safety net” for those who can’t afford the private coverage.
In summary, the market in Australia is dominated by Medibank and Bupa, with nib, Australian Unity and HCF following closely behind. Other smaller players are also present in the market.
What do customers see as Medibank Private’s strengths and weaknesses?
Medibank Private is one of the largest health insurance providers in Australia, with a significant market share. As such, it has a wide range of products and services that are designed to meet the needs of different types of customers. Some of the strengths that customers may see in Medibank Private include:
Wide range of products: Medibank Private offers a wide range of health insurance products, including hospital cover, extras cover, and combined cover. This allows customers to choose a product that best meets their needs and budget.
Strong network of providers: Medibank Private has a large network of providers, including hospitals, doctors, and other healthcare professionals. This makes it easy for customers to find a provider that is covered by their policy.
Rewards program: Medibank Private has a rewards program that rewards customers for taking steps to improve their health. This can include discounts on gym memberships, health assessments, and other wellness services.
Online services: Medibank Private offers a range of online services that make it easy for customers to manage their policies and access information. Customers can also access a virtual GP service and other online health resources.
However, as with any company, Medibank Private also has some weaknesses that customers may see. Some of these might include:
Premiums: Medibank Private’s premiums may be higher than some of its competitors, which may make it less affordable for some customers.
Limited coverage options: While Medibank Private offers a wide range of products, some customers may find that their specific needs are not fully covered by the company’s policies.
Complaints handling: like with any company, despite Medibank Private’s efforts, some customers may have negative experience when trying to resolve a complaint or claim and may not be satisfied with the service provided.
It’s worth noting that these are just examples and experiences can vary depending on individual customers and their specific circumstances. It’s important for customers to research and compare different health insurance providers in order to find the one that best meets their needs.
Is there anywhere that you can see that my health insurance company can get a competitive advantage in the australian health insurance market?
As a Digital Product Manager, you have the opportunity to leverage technology and data to gain a competitive advantage in the Australian health insurance market. Here are a few examples of how your company can use technology to differentiate itself from competitors:
Personalized pricing: By using machine learning and AI, your company can personalize pricing for customers based on their individual health risks and lifestyle choices. This can help your company to offer more competitive pricing to low-risk customers while still maintaining profitability.
Predictive modeling: By analyzing data from various sources, including claims data, medical records, and demographic information, your company can identify patterns and predict which customers are most likely to have certain health conditions. This can help your company to proactively target those customers with personalized products and services, such as disease management programs.
Telemedicine: By offering virtual consultations, your company can give customers more flexibility in how they access healthcare services. This can help to improve customer satisfaction and reduce costs associated with traditional in-person consultations.
Customer engagement: By using chatbots, virtual assistants, and other AI-powered tools, your company can make it easier for customers to access information and manage their policies. This can help to improve customer satisfaction and reduce costs associated with call centers.
Fraud detection: By using AI and ML to analyze claims data, your company can quickly and effectively identify and investigate potential fraud, which can ultimately help to reduce costs and protect policyholders.
Digital Experience: By offering a seamless and personalized experience through mobile apps, easy-to-use website and easy claims processing, your company can differentiate itself from competitors and improve customer retention.
These are just a few examples, and there may be other areas where your company can leverage technology to gain a competitive advantage in the Australian health insurance market. It’s important to keep an eye on industry trends and to be open to new ideas and innovative solutions.
What are the potential risks of using personalised pricing?
Personalized pricing, which uses data and machine learning algorithms to set different prices for different customers based on their individual characteristics, can have a number of potential risks. Some of these risks include:
Fairness and discrimination: Personalized pricing can be perceived as unfair, particularly if it results in higher prices for certain groups of customers, such as those with pre-existing conditions or those in lower income brackets. It could also be perceived as discriminatory if it results in higher prices for certain groups based on factors such as race, gender, or age.
Privacy concerns: Personalized pricing requires the collection and analysis of large amounts of data on customers, which can raise privacy concerns. Customers may be concerned about how their data is being used and may be hesitant to share personal information with your company.
Legal risks: Personalized pricing may also raise legal risks, as it could be seen as a form of price discrimination, which is prohibited in some jurisdictions. It’s important to be aware of and comply with any relevant laws and regulations when implementing personalized pricing.
Customer dissatisfaction: Personalized pricing may also lead to customer dissatisfaction if customers feel that they are being charged more than others for the same product or service. This can lead to negative customer reviews, complaints, and reduced customer loyalty.
Reputation risks: if personalized pricing is not handled correctly, it could lead to negative press and damage your company’s reputation.
It’s important for any company considering using personalized pricing to carefully weigh the potential benefits against these risks and to have proper governance and controls in place to mitigate them. It’s also important to be transparent with customers about how the pricing is determined and to give them the option to opt out of personalized pricing. Additionally, companies should ensure they are meeting any legal requirements and ethical standards related to personalized pricing.
What sort of data would be needed to be able to implement personalised pricing?
In order to implement personalized pricing, your company would need to collect and analyze a variety of data on customers. Some of the types of data that would be needed include:
Demographic data: Information on customers’ age, gender, income, education level, occupation, etc. which can be used to identify patterns and segment customers into different groups.
Health-related data: Information on customers’ health status, medical history, and health-related behaviors, such as smoking or exercise habits. This data can be used to assess customers’ health risks and to personalize pricing accordingly.
Claims data: Information on customers’ past claims, including the type of claims, the cost of claims, and the outcome of claims. This data can be used to assess customers’ likelihood of making future claims and to personalize pricing accordingly.
Behavioral data: Information on customers’ online and offline behavior, such as website browsing history, purchase history, and social media activity. This data can be used to understand customers’ preferences and to personalize pricing accordingly.
Geographic data: Information on customers’ location, which can be used to understand regional trends and to personalize pricing accordingly.
Economic data: Information on customers’ economic situation, such as credit score, income, debt, and employment status, which can be used to understand customers’ ability to pay.
This data can be collected through various means such as surveys, forms, digital trackers, third-party data providers, or even through the usage of connected devices and applications. It’s important to note that the data collection, handling, and usage should be done in compliance with relevant laws and regulations, such as data protection and privacy laws. Additionally, customers should be fully informed about the data collection and usage and have the option to opt-out of providing certain data.
What type of customers would benefit the most if we implemented personalised pricing?
Customers who would benefit the most if a health insurance company implemented personalized pricing would be those who are considered low-risk. These are customers who are less likely to make claims, have fewer pre-existing health conditions, and have healthier lifestyles. Personalized pricing would allow your company to offer these customers more competitive pricing, which would make your products and services more affordable for them.
Examples of customers who would likely benefit from personalized pricing include:
Younger and healthier individuals
Non-smokers
Individuals with no pre-existing medical conditions
Individuals who live in low-risk geographical areas
Individuals with higher income or good credit score
It’s important to note that personalized pricing would not only benefit the low-risk customers but also the company, as it would help the company to better price its products and services, and ultimately, can improve profitability.
It’s also important to consider that while personalized pricing may benefit certain low-risk customers, it may not be as beneficial for higher-risk customers. In these cases, it’s important to ensure that your company has other products and services in place to meet the needs of higher-risk customers and to ensure that they are not disadvantaged by the personalized pricing.
Of the types of customers you mentioned above — can you tell me roughly what percentage of the market they make up?
It’s difficult to provide a precise percentage of the market that low-risk customers make up, as it can vary depending on a number of factors such as the region, the population characteristics, and the company’s target market. However, generally speaking, low-risk customers tend to make up a relatively small percentage of the overall market.
Younger and healthier individuals tend to make up a larger percentage of the market, as they are less likely to have pre-existing conditions or chronic health issues. However, as they get older and start to develop health issues, they may become higher risk customers.
Non-smokers also tend to make up a larger percentage of the market, as they are less likely to develop smoking-related health conditions. However, the percentage of non-smokers in the population can vary depending on the region and culture.
Individuals without pre-existing medical conditions are also a smaller percentage of the population, as many individuals have some form of pre-existing conditions.
Individuals who live in low-risk geographical areas also tend to make up a smaller percentage of the market, as certain regions may have higher rates of certain diseases or conditions.
Individuals with higher income or good credit score, also tend to make up a smaller percentage of the market as those.
Within Sydney what percentage of customers would be under 40 and healthy with no existing health conditions?
It’s difficult to provide an exact percentage of customers within Sydney who would be under 40 and healthy with no existing health conditions, as it can vary depending on a number of factors such as lifestyle, occupation and access to healthcare. However, it’s likely that they would make up a relatively small percentage of the overall market, as many people develop health conditions as they age or have pre-existing conditions.
It’s also worth noting that the percentage of individuals in this category can vary depending on the region, as some areas may have higher rates of certain health conditions than others. It’s also important to note that the percentage of this category of customers can change over time as the population ages, and lifestyle changes.
It’s recommended to conduct market research and gather data specific to your target market to get a better understanding of the percentage of this category of customers that you could expect in Sydney.
Is there any market research that can give me a breakdown of various customer cohorts based on age and health in Australia?
Yes, there are a number of market research companies and organizations that provide data and insights on customer cohorts based on age and health in Australia. Some examples include:
Australian Bureau of Statistics (ABS) — ABS provides a wide range of data and statistics on the Australian population, including information on age and health. They also regularly release reports and publications on specific health-related topics, such as the Australian Health Survey, which provides detailed information on the health status and risk factors of the Australian population.
Australian Institute of Health and Welfare (AIHW) — AIHW is a government-funded organization that provides data and information on the health of Australians. They regularly release reports and publications on specific health-related topics, such as the National Hospital Morbidity Database, which provides information on hospital admissions and separations.
Private Market Research Companies — There are private market research companies which conduct and publish research on various segments of the population, including health, age and demographics. They use various data collection methodologies such as Surveys, Online research and focus groups.
Private Health Insurers — Private Health Insurers also have access to a large dataset of their policyholders, which can give them information on demographics and health conditions of their customers. They might also conduct market research to understand their customer base and potential new customers.
It’s important to note that the data and information provided by these organizations and companies may have some variations depending on the source and methodology used. It’s recommended to review multiple sources to get a more comprehensive understanding of the market. Additionally, it’s also worth noting that the data and information may have some lag and might not reflect the current situation.
Can you write 1 paragraph that can clearly summarise why we should be using personalised pricing that I can put into my business case?
Personalized pricing, which uses data and machine learning algorithms to set different prices for different customers based on their individual characteristics, can provide significant benefits for your health insurance company. With personalized pricing, your company can better price its products and services, which can help to improve profitability and competitiveness. Personalized pricing can also help to attract and retain low-risk customers, by offering them more competitive pricing, which can help to increase customer loyalty and reduce customer churn. Additionally, by identifying patterns and predicting future claims, your company can proactively target higher-risk customers with personalized products and services, such as disease management programs. This can help to reduce costs associated with claims and improve the overall health outcomes for your customers. Overall, personalized pricing can help your company to improve its financial performance and customer satisfaction, which can be key factors for success in the Australian health insurance market.
can you give me a breakdown of australians by age group with each age group being a 5 year increment?
I apologize, but as my knowledge cut off is in 2021, I am unable to provide you with a breakdown of the Australian population by age group with a 5 year increment. The data on population demographics is constantly changing, and I do not have the most up-to-date information. I recommend checking with the Australian Bureau of Statistics (ABS) or other government bodies that might have more recent data on the population demographics in Australia. They typically release census data every 5 years and also regularly update the population estimates. These sources would be able to provide you with a more accurate and current breakdown of the Australian population by age group.
For 2020 can you give me a breakdown of Australians by age group?
I apologize, my knowledge cut off is in 2021, so I do not have the most recent population demographics data. However, I can give you an idea of what the population breakdown was like in 2020.
According to the Australian Bureau of Statistics (ABS) as of 2020:
0–14 years: 15.3%
15–24 years: 12.9%
25–34 years: 12.7%
35–44 years: 12.5%
45–54 years: 14.0%
55–64 years: 12.5%
65–74 years: 8.9%
75–84 years: 4.9%
85+ years: 2.0%
It’s important to note that these numbers are an approximation and can change over time. Additionally, the population distribution may vary depending on the region and can be affected by migration patterns and birth rate.
What proportion of Australians would be considered healthy?
It’s difficult to provide an exact proportion of Australians who would be considered healthy, as it can vary depending on a number of factors such as age, lifestyle, and access to healthcare. Additionally, the definition of “healthy” can vary and can be a subjective measure. However, according to the Australian Bureau of Statistics (ABS) around 69% of Australians aged 15 and over reported their health as being “good” or “very good” in the 2019–2020 period. Additionally, the ABS reported that in the same period, that around 12% of Australians aged 15 and over reported their health as being “fair” or “poor”.
It’s important to note that these numbers are an approximation and can change over time. Additionally, the data may have some lag and might not reflect the current situation. Factors such as the outbreak of COVID-19 pandemic and its aftermath may have affected the proportion of Australians considered healthy. It’s recommended to check with the ABS or other government bodies for the most recent data and updates on health status of Australians.
Originally published at https://dius.com.au on January 17, 2023.
