Starting an insurance system replacement project.

Starting an insurance system replacement project from scratch is easier than you think.

Here are some of our top tips.

By Martin Stewart : 14th of December 2020

To maintain an old legacy system, or to start again? While the benefits of new technology are manifold, many insurance executives are daunted by the prospect of implementing new systems and have delayed the decision forever.

A BCG survey of 1800 companies
across nine industries found
Insurers scored poorly
on digital maturity.

A BCG survey of 1800 companies
across nine industries found
Insurers scored poorly
on digital maturity.

It’s no surprise then, that a BCG survey of 1800 companies across nine industries found that insurers scored only 47 out of 100 when it comes to digital maturity—one of the lowest scores of any industry.

The downsides of a legacy system

The older your system, the more costly it is to maintain. It becomes more difficult to find good software developers who are skilled in the coding language your system was written in. It may require specific hardware and infrastructure, costing more to maintain than a modern cloud-based solution.

However, the real problem isn’t just the core legacy systems itself but the myriad of supplementary components tacked on the outside to provide the functions missing in the core. Simple changes such as adding a new rating factor can feasibly mean updates to 10 or more systems.

Also, this mishmash of interconnected systems will likely be burdened with code supporting decades of product variations and business processes. This means what may be a straightforward business change may ultimately turn out to be complex to implement and expensive. One McKinsey survey found IT costs for legacy systems can be as much as 41% higher per policy—yet having been involved in developing insurance systems since 1983 I know the real number is often much higher.

It’s also worth considering that modern software development is at least 10 times faster than it was 30 years ago.

Your legacy system is like an old Nokia.
It makes calls and sends texts,
but it can’t hope to compete
with a smartphone.

Your legacy system is like an old Nokia.
It makes calls and sends texts,
but it can’t hope to compete
with a smartphone.

Think of your legacy system like an old Nokia phone. It still makes phone calls and sends texts, but it can’t hope to compete with a smartphone.

Similarly, while a legacy insurance system may be able to sell policies and process claims, today’s tools offer so much more than that. The ability to interface with phone systems allow your customers direct access to their policies and claims, support online chat, generate documents and email them out and perform data analytics requiring a digital solution.

So not only could your legacy system be costing you to run, it could also be costing you your business. McKinsey also found that players with modernised IT systems write 40% more policies than competitors with legacy systems.

So, do you update your old system or start again?

Imagine that you’ve inherited an old house. The foundations are sagging. The roof is leaking. If you tackle it like many systems upgrades, you would patch up the critical issues first and then update the house room by room over several years. The end result is a compromised design, that takes a lot longer and costs way more than you first expected. And, by the time you get to what you thought would be the end you’ll need to start again.

Whereas, if you start from scratch you can ensure that the design is exactly what you want, not limited by the original floorplan. You can preserve the best features of the old designs like the bay windows and high ceiling while reconfiguring the floor plan to suit modern living. You’ve the ability to add in features such as underfloor heating, ducted air conditioning and the latest lighting and appliances. Best of all, you can incorporate passive sustainable design principles into the build to save you serious money over the next few decades.

The same applies when looking at old insurance technology.

Adding on enhancements is complex,
risky & costly. And you will never
fix the major issues like
poor user experience.

Adding on enhancements is complex,
risky & costly. And you will never
fix the major issues like
poor user experience.

Adding on enhancements is complex and risky, with a high price tag. Those add-ons will never really fix the major issues like poor user experience, because the underlying foundations of the system are still slow and outdated.

In contrast, a new system allows you to design what you need, from the outset.

But before you start, consider:

  • Rationalising your products to remove duplicates and near-duplicates.
  • You’ve probably added many products and cover options over the years. One insurer we worked with had over 100 car insurance products, while a commercial insurer had a different set of rating spreadsheets for each product for each state because the underwriters for each state couldn’t agree on a standard.

    Take time to rationalise your products and pare the options down to the minimum number needed. Keep it simple. And leave any legacy products back in the old system.

  • Whether standardising technology is really necessary for your business
  • Over time you’ve no doubt added lots of new systems to do new things: CRM, a data warehouse, new reporting tools, phone apps, etc. If your business has been involved in mergers you’re likely to have a large amount of duplication in systems and data. Usually in merger planning there will be lots of promised savings from consolidation, but rarely is the consolidation completed.

    If you’ve a maze of different systems like this, it may seem attractive to simplify things by selecting “best of breed” software packages so you get back to having just one of each.

    While there are solid reasons to use special tools (perhaps for code quality checking) and libraries (say a well-supported JavaScript and CSS) it’s different if you’re looking at untangling a jumble of systems. While it may be a great goal to have a single CRM or a single workflow tool across all your technology are you really going to be able to implement them across all your systems and is it worth it?

    For example, to implement a shared company-wide CRM means you need to identify clients replicated in multiple systems and then validate and merge the data. You’ll also need to implement a complex set of system-to-system interfaces and work out how you’ll manage updates and synchronise the data. Many companies have started this process with visions of the value of a common client record but give up well before they achieve it (but after they’ve spent a fortune!).

    Designing great solutions isn’t for the purists. What you need¬—to quote Einstein—is “the simplest solution that works, but no simpler”. So don’t be afraid to compromise on architectural purity and go for what works best for the business.

  • Build in flexibility to the core design
  • Whether you’re building a house or a system you’ll know a lot more about what you need after you’ve finished. So rather than try and design the perfect system up front it’s better to focus on where you’d like additional flexibility. Do you want to be able to update premium rates quickly, be able to run campaigns and make special offers? Do your products have hundreds of variations—say building construction, occupations or business categories? Build that flexibility in from the very beginning.

  • Re-examine your processes and policies.
  • How many of your processes exist
    because that’s the way you’ve
    always done it? Are you
    still using paper forms?

    How many of your processes exist
    because that’s the way you’ve
    always done it? Are you
    still using paper forms?

    How many of your processes exist because that’s the way you’ve always done it? Are you still using paper forms, requiring documents to be sent by fax, or insisting on wet signatures? Do you customers have direct access to their policies and claims? Can they do things like change cover, change their address, make a payment or lodge a claim themselves? Use the opportunity of a new build to look critically at everything you do. By streamlining you can save on operational costs and speed up the turnaround time while improving the experience for customers.

    The benefits of modernisation are certainly compelling. However, it can seem like a long and challenging journey. However, starting from scratch can be easier than you think.

    Start with a pilot program

    Since your core legacy systems—by definition—are critical to day-to-day operations you can’t simply take them down and spend a few months building a replacement.

    While the eventual goal is to go completely digital, rest assured that you don’t have to scrap what you have now in one go. In fact, going digital is easier than you think precisely because it can be done in stages and the new technology is immediately more productive.

    We suggest beginning with a pilot program. The full modernising program will take several years and you’ll want to be able to show results sooner than that. Moving one area of operations to a digital platform is an effective way to demonstrate the benefits of digitisation to the ultimate decision-makers.

    When making your choice, look at what’s motivating you to modernise. Deloitte suggests that the top three business drivers are likely to be product strategy, technological relevance and service enablement. Time and cost are also significant factors. Every insurance company is different, so ensure that the project aligns with your overall business strategy.

    For many insurers, the ideal candidate
    for a pilot is online quoting,
    replacing paper applications
    and spreadsheet rating.

    For many insurers, the ideal candidate
    for a pilot is online quoting,
    replacing paper applications
    and spreadsheet rating.

    For many insurers, the ideal candidate for a pilot is an online quoting page that replaces a process that’s currently handled through paper application forms and with rating done in a spreadsheet.

    Another great candidate for a pilot is anything that’ll make life easier for your agents and customers. Analyse calls to your service desk to see what could be made available on the internet via self-service. You may find this includes things that are effortless to accomplish, for example the ability to reprint a policy or generate a certificate of currency.

    Set a clear target that you can achieve and verify. An obvious one is simply to be better than the competition. If you can create a design that’s easier and more pleasing for customers to use than your competitors and which saves time in back-end processing, it’s easy to demonstrate real benefits.

    While you may feel pressure to prove a real cost benefit case, remember this is a pilot to test out options. Many of the benefits will be intangible or difficult to predict and measure, but that doesn’t mean that the pilot lacks value. At the same time if it doesn’t work, don’t be afraid to throw it out and start again. I’ve seen too many projects that were doomed to failure almost from the start but the teams kept going because it took so long to get the projects approved no one wanted to see them cancelled.

    A pilot program is small in scope by design. However, it should still be a priority for the team assigned to work on it. Ideally, you should have a dedicated, full-time team and insulate them from other tasks that could pull their attention away. Be clear on the limited scope of the project (don’t let managers chuck in some new tech they just read about), but also keep in mind that it’s only the first phase of a higher-level plan to move your systems to a 100% digital platform. Stay agile, so you can adapt quickly as you learn more.

    Progressively replace your systems

    After the pilot, assuming it’s successful, you’re on your way. Since it’s not feasible to put your entire system out of action at once, start with one area of operations and then progressively replace one or two systems at a time. That way, you’ll minimise the disruption to your normal business and keep costs down. It’s best to start at one end so you can progressively take out slices of the old systems and minimise the number of interfaces needed to build between the old and the new.

    The time it takes will depend on your budget. Trust me—it’ll get faster and more efficient as you take the lessons forward to the next system.

    However, my experience is the speed isn’t the most critical element. On completion of each area, you’ll have gained further insight into the business needs and technology. Your team will become more experienced and better trained.

    Modern digital solutions, add value
    through machine learning,
    artificial intelligence
    and data analysis.

    Modern digital solutions, add value
    through machine learning,
    artificial intelligence
    and data analysis.

    In addition, while progressing to modern digital solutions, you’ll be able to add value through advanced technologies like machine learning, artificial intelligence and data analysis. You’ll gain better insight into your operations, customer behaviour and risk factors. These insights will help you adapt your future transformation efforts to deliver the greatest benefits.

    Insurtech can help you on the journey to digital transformation

    For smaller insurers that may never have the budget to build a modern IT system in-house the best option may be to partner with an insurtech company. It can bring in the technical expertise to compliment your insurance expertise. My final tip is to consider giving the insurtech company a real stake in your transformation to ensure it’s aligned with your goals. Its staff may be willing to work for low fees with extra rewards once the projects are successful, simplifying your capex approval processes and enabling you to move much faster.

    Rethink, don’t reboot your commercial insurance systems

    Whether you’re looking for an innovative platform to support business growth, you’re over capacity and need to outsource a project, or you’re in a jam and need a rapid solution, Axe is ready to support you.