Tag Archives: VA

How to Use AI, Starting With Distribution

How can insurers meet increased customer expectations at a lower cost? AI-powered care delivers on a future vision of customer service with an opportunity for savings of 30% by, for example, driving customers to digital experiences. In this post, I will explore how to apply AI using an intelligent customer engagement (ICE) framework.

How can your insurance company increase its artificial intelligence quotient (AIQ) with a balanced innovation strategy? In this blog series, I’m exploring the myriad ways in which AI adds value to financial services in general and the insurance value chain in particular. In my previous post, I defined the term AIQ and revealed and discussed three key ingredients to building a strong AIQ: technology, data and people.

In this post, I’ll take a close look at one of the key areas in the insurance value chain—sales and distribution—and explain how AI-related technologies can add value to this function. But first, I want to reiterate the value of AI and why it’s important to transform your business into an AI business.

Why a strong AIQ is vital for your business—and why you need a strategy first.

Most of what’s written about AI relates to cost-cutting and job losses, but as we saw with the example given with regard to the health industry in the previous post, AI is a much more optimistic story. Its greatest benefits are not only efficiency and productivity but innovation, improved customer and employee experiences and the development of new sources of value and growth, especially when they augment human capabilities.

However, to gain these benefits and to identify relevant use cases, it is necessary to develop a cross-enterprise AI strategy that clarifies the strategic goals: the whys, the hows and the whats of the business model leveraging our “AI strategic approach,” as outlined below.

See also: 3 Steps to Demystify Artificial Intelligence  

Once the strategic goals have been clarified, the potential use cases for AI can be identified and prioritized according to the impact and estimated implementation effort they have on supporting the achievement of these goals (such as enhanced operational efficiency or improved customer experience) along the insurance value chain:

How can insurers use AI in sales and distribution?

As mentioned in my previous post, there are numerous use cases of AI that can be applied along the insurance value chain. In this post, we focus on AI in marketing, sales and distribution, including:

  • Enabling intelligent customer engagement
  • Workload balancing/lead allocation for agents
  • Machine learning insights to support customer segmentation
  • Automated data extraction from PDF reports and comparison against various policy combinations
  • Automated demand analysis and generation of new product offerings
  • Intelligent reporting and visualization
  • Customer personality and tone analysis
  • Automated creation of targeted marketing materials and promotions
  • Enablement of intelligent self-service product research for customers
  • Automated product recommendations and natural language question answering

When it comes to deciding which AI to employ, insurers need to focus on the things that AI and humans do best together. When AI is combined with human ingenuity across the enterprise, it can help solve complex challenges, develop new products and break into and create new markets.

Data analytics for better customer engagement

In sales and distribution, insurers can use data analytics to improve customer engagement.

Virtual assistance (VA)

Accenture’s virtual adviser Cathy (Cognitive Agent to Help You) is a self-learning virtual agent that responds to customer queries by extracting information from a back-end database. Cathy is always learning more as it consumes human-agent interactions and stores knowledge on its database, enabling it to make automated product recommendations based on customer profiles. If a more complex customer request arises, Cathy seamlessly transfers the request to a human agent.

Machine learning

Insurers can boost their sales and distribution by using machine learning to analyze customer personality and tone. Machine learning makes selling (and buying) insurance easier than ever—virtual agent Amelia, for example, can give customers a motor insurance quote immediately, without the need to speak to a human being.

What are the benefits of AI for sales and distribution in insurance?

When humans and machines work together, they create the opportunity for growth and innovation. Within insurance sales and distribution, AI-related technologies can help to enable:

  • Increased lead generation — data analytics helps insurers identify and reach potential customers. The insight derived from data analytics can drive and constantly improve the sales team’s effectiveness at generating leads.
  • Efficient leverage for cross- and up-selling — AI such as data analytics and VA gives insurers invaluable knowledge about their customers, making it easier to convince them to buy a comparable higher-end product (up-selling) or a product that is related to the ones they already have (cross-selling).
  • Increased service quality — self-learning virtual advisers like Cathy interact with customers and absorb information about their needs. This valuable feedback drives personalization of products and improves the quality of services.

Use case: intelligent customer engagement (ICE)

With intelligent customer engagement, insurers can strategically deflect issues that need to be addressed from human agents to machine chatbots. They can predict why customers are calling and approach them effectively. Humans now take on the new role of knowledge engineer: They take over where the AI ends and curate the knowledge corpus over time.

See also: Are Insurers Ready for Voice Search?  

In our future vision for insurance, AI-powered care gives insurers the opportunity to save 30% of their customer service costs by:

  • Driving customers to digital experiences;
  • Providing conversational interactions that increase digital adoption and containment;
  • Leveraging AI to automate and deliver consistency across channels.

Insurers are looking to better connect with their customers and to strengthen relationships with experiences that delight them—while reducing the cost to serve. Technology enables them to do this by, among others, shifting the mix of customer contacts:

It’s time to put your AIQ to work

When you combine human ingenuity with AI—such as data analytics, virtual assistance and machine learning—to improve the sales and distribution function, you will see results improving.

AI presents the opportunity for business transformation by enabling intelligent processes in the value chain and intelligent products and services in the market. Success will depend on how well your organization can harness the combined power of technology, data and people.

In my next post, I’ll look at how you can use AI to augment underwriting and service management. Get in touch to find out how you can boost your company’s sales and distribution function, as well as others within the insurance value chain, or download our report on How to boost your AIQ.

Dangerous Confusion on ‘Painandsuffering’

What is pain? According to Merriam-Webster, it is “the physical feeling caused by disease, injury or something that hurts the body.” Which is different than suffering: “to become worse because of being badly affected by something.” Often, these words are treated as synonyms (or as a single word, “painandsuffering”) when they are actually quite different. Pain is what happens to you. Suffering is how you handle it.

The confusion of these two terms can create issues.

The American Pain Society in 1996 described “pain as the fifth vital sign” (giving it equal status with blood pressure, heart rate, respiratory rate and temperature). The phrase created a perfect storm because it coincided with the message being delivered to medical schools and the healthcare industry that doctors had an opioid phobia and were under-treating pain. That was followed in 2000 by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) establishing standards for pain assessment and management. Then the Veterans Health Administration incorporated the new emphasis into its national pain management strategy. So, regardless of where a patient was treated and by whom, the (subjective, patient-driven) assessment of pain was one of the first questions asked and often drove treatment plans.

Then the new approach began to be questioned. A 2006 study by the VA found quantifying pain “did not increase the quality of pain management.” In June 2016, the American Medical Association recommended removing “pain as the fifth vital sign” and connected the idea to the beginning of over-prescribing of opioids. Opponents of the change say it will “make it even more difficult for pain sufferers to have their pain properly diagnosed and treated.” Proponents of the change say “pain is not a vital sign, but more of a symptom, and cannot be measured.”

So far, pain is still the fifth vital sign.

See also: Health Startups Go After 3 Pain Points  

The biggest problem is unrealistic expectations – patients often are told or come to believe they will be pain-free. When they’re not, and their condition becomes chronic, it sows doubt in the mind of both the patient and clinician.

The second biggest problem is that often the circumstances beyond their physical pain is ignored. I am convinced that dealing with what happens between the ears and at home is as important as what is physically wrong with the body (i.e. the biopsychosocial model).

So how is “pain as the fifth vital sign” measured? Sometimes it’s a scale of frowny face to smiley face. But often it’s a comparative pain scale, from 0 to 10. The Health Organization for Pudendal Education (HOPE) offers the best description:

  • 0 – No pain – Feeling perfectly normal.
  • 1 – Very mild – Barely noticeable pain, like a mosquito bite or a poison ivy itch. Most of the time, you never think about the pain.
  • 2 – Discomforting – Minor pain, like lightly pinching the fold of skin between the thumb and first finger with the other hand, using the fingernails. Note that people react differently to this self-test.
  • 3 – Tolerable – Very noticeable pain, like an accidental cut, a blow to the nose causing a bloody nose or a doctor giving you an injection. The pain is not so strong that you cannot get used to it. Eventually, most of the time you don’t notice the pain. You have adapted to it.
  • 4 – Distressing – Strong, deep pain, like an average toothache, the initial pain from a bee sting, or minor trauma to part of the body, such as stubbing your toe really hard. So strong you notice the pain all the time and cannot completely adapt. This pain level can be simulated by pinching the fold of skin between the thumb and first finger with the other hand, using the fingernails and squeezing hard. Note how the simulated pain is initially piercing but becomes dull after that.
  • 5 – Very distressing – Strong, deep, piercing pain, such as a sprained ankle when you stand on it wrong, or mild back pain. Not only do you notice the pain all the time, you are now so preoccupied with managing it that your normal lifestyle is curtailed. Temporary personality disorders are frequent.
  • 6 – Intense – Piercing pain so strong it seems to partially dominate your senses, causing you to think somewhat unclearly. At this point, you begin to have trouble holding a job or maintaining normal social relationships. Comparable to a bad non-migraine headache combined with several bee stings, or a bad back pain.
  • 7 – Very intense – Same as 6 except the pain completely dominates your senses, causing you to think unclearly about half the time. At this point, you are effectively disabled and frequently cannot live alone. Comparable to an average migraine headache.
  • 8 – Utterly horrible – Pain so intense you can no longer think clearly at all, and have often undergone severe personality change if the pain has been present for a long time. Suicide is frequently contemplated and sometimes tried. Comparable to childbirth or a really bad migraine headache.
  • 9 – Excruciating, unbearable – Pain so intense you cannot tolerate it and demand pain killers or surgery, no matter what the side effects or risk. If this doesn’t work, suicide is frequent because there is no more joy in life whatsoever. Comparable to throat cancer.
  • 10 – Unimaginable, unspeakable – Pain so intense you will go unconscious shortly. Most people have never experienced this level of pain. Those who have suffered a severe accident, such as a crushed hand, and lost consciousness as a result of the pain and not blood loss have experienced level 10.

How many times have people said their pain is a 9 or 10 (or a 47) when they’re conscious, sitting upright and drove themselves to the doctor’s office? I have seen that manifold times in hundreds of chronic pain workers’ comp claims since 2003. But it’s easy to succumb to that kind of self-assessment …

I had the flu in February and went to a CVS Minute Clinic. One of the initial questions the nurse practitioner asked me (having been prompted to do so by her practice management software) was my level of pain. I truly felt miserable — body aches, high temperature, sneezing. For a brief moment, because I wanted to ensure a prescription of Tamiflu, I wanted to catastrophize (“an irrational thought a lot of us have in believing that something is far worse than it actually is“) and say I was a 9 or 10. But then I remembered all the times I had argued against that approach. And I remembered exactly what a 9 or 10 meant. So I resisted the urge and gave myself a 5 rating. I still got the Tamiflu that started the journey to recovery.

See also: Better Outcomes for Chronic Pain  

Pain is complicated and individual, so there is not a single answer for quantifying and treating it appropriately. However, I have three high-level suggestions:

  • Re-calibrate the scale. The clinician should educate patients on the true meaning of 0 through 10 and help them decide on a lower number that better describes their pain. That would require an actual dialogue between the clinician and patient. I understand that pain is unique and personal. But if patients can convince themselves their pain is a 6 instead of a 10 (or a 47), then managing it seems much more achievable.
  • Be honest. If there is going to be residual, chronic pain, the patient should know it. And own it.
  • Manage the pain. In my opinion, “pain management” is a term that is often misused. You can’t manage your pain if you’re comatose (i.e. sedated on opioids, benzos, muscle relaxants, et al.). Yet we often see “pain management” as a series of pills or injections that are passive and repetitive (in some cases, I think pain management clinics have become “addicted” to the repeat office visits). At some point, patients need to manage their pain rather than allowing the pain to manage them, and be taught how to do that. That could mean yoga, an active lifestyle, better nutrition, biofeedback, proper sleep hygiene, deep breathing exercises, mindfulness, volunteer work or any number of other methods in combination or isolation that work for the patient. The key is an internal locus of control (“he or she can influence events and their outcomes“).

I’m not saying pain isn’t real. For those dealing with chronic pain, it is very real. But I’ve chatted with and observed too many people with significant chronic pain who overcome it on a daily basis to live productive and happy lives. I know that chronic pain does not have to win. Instead, we need to re-define pain, re-define suffering and help people take back control of their lives.

I will finish with this wisdom from Dr. Stephen Grinstead:

  • Thoughts cause feelings
  • Thoughts + feelings = urges
  • Urges + decisions (choices) = actions
  • Actions cause reactions
  • Reactions could help or hurt management of pain

In other words, how you think about pain influences how much power pain has over you. So think differently.