AI perspectives from Tractable.

Back to blog list: See all posts

22 Mar 2021

UK insurers must prepare for the post-pandemic roads

Image credit: Mike Malone

By Adrien Cohen, Tractable co-founder and President

Within three weeks of the UK’s first national lockdown being imposed, there were up to 80% fewer car journeys on Britain’s roads. AA president Edmund King even said traffic volumes in Britain are “now akin to those in the early 1970s”. 

While numbers have picked back up since, the (mostly) consistent ‘stay-at-home’ order means many drivers have gone an entire year without getting behind the wheel.

Yes, this has had positive effects on air quality (although the long-term impact is questionable), but will also have implications for insurance policies. The number of motor claims in 2020 dropped by a fifth, with the total payout falling by 6% on-year. 

Be warned: this is the calm before the storm. 

Millions of drivers being out of practice means more road accidents – and more severe ones. In addition, learners have had to put their lessons on hold, so the driving experiences of an entire generation of new road-goers has been disrupted. 

In short, insurers must get ready – a surge in claims is on the way! 

Claims: a crash course

So, more claims are coming – but that poses a major problem for the repair sector. 

Why? Because even at the current low numbers, it still can’t keep up with the number of vehicles it needs to fix and ensure policyholders get back on the roads as quickly as possible.

Today, while the time it takes to repair a car might only take a few hours, it probably still takes weeks to actually return a car to its driver. Delays occur when an assessor is needed to inspect the damage and estimate the cost of repairs – but one isn’t available immediately, and might take days to schedule. 

In turn, the repairer can’t start to fix the vehicle until the insurer gives the go-ahead, dependent on the assessment it needs to receive. Every lost hour could cause a potential chain reaction down the line – whether in the availability of required expertise, or in physical factors like space in the garage of forecourt, or parts.  

For an insurer, reviewing a repair after it’s been carried out is also time and labor-intensive. For example, It can take about 30 minutes for an assessor to manually review a claim. Given that a major insurer might process millions each year, it simply doesn’t have the time or capacity to check them all. 

As a result, insurers often end up checking only a certain percentage, often based on specific rules – for example, they might look at expensive accidents, or ones containing certain damage patterns. However, that not only means that some repairs won’t be carried out to the correct specifications… but the insurer will have no way of checking.  

Does AI hit the mark?

To overcome these shortcomings, many insurers are exploring how to best use technology to automate elements of the claims process and increase their oversight of every repair, ensuring that everyone receives a fair price for their work. 

For example, AI can be used to evaluate damage at scale, enabling insurers to look in detail at many more claims than they were previously able to see. 

Unfortunately, not all AI is made equal. While an accurate, reliable AI that operates at the level of a human assessor will help an insurer be more efficient, a solution that doesn’t meet those standards will create more problems than it solves. And in a logistics process like insurance, every mistake – whether that’s incorrectly sending a vehicle for scrap, or replacing a part that ought to be repaired – adds up. 

For context, in 2018, there was a total of over £7bn of motor insurance claims in the UK… which means that every 1% gain (or loss) in efficiency is worth £70m to the ecosystem. 

In other words, accuracy doesn’t just matter – it’s the key differentiator in determining whether a tool like AI is worth the investment, or not. 

What should insurance companies look for in an AI solution?

So you’ve come to the (right) decision to automate the claims process. Excellent!

But how do you decide which one is going to provide the greatest value? Here are some key questions you need to ask yourself: 

  1. Is its performance equal to human assessors?
  2. Does it reduce the key 2 key times?
  3. Is there a clear level of transparency for how the technology makes decisions?
  4. Does it cover any car make and/or model?
  5. Does it provide paint verification?
  6. Can it identify fraudulent claims, e.g. unrelated prior damage?
  7. Does it provide a complete list of repair vs replacement costs and labour hours?

If the answer to each of these questions is ‘yes’, then you can be confident that the platform can provide the level of efficiency and accuracy that you need. 

With the end of England’s lockdown looming, now is the time for insurers to invest in the right AI solution and elevate the claims experience.


Tractable is already being used by 21 of the world’s top 100 insurers and is live in 14 countries. This equates to over $1bn (and counting!) auto accident claims being processed and accelerated recovery for over a million households. To learn more about how the technology works, book a demo with the team today

Previous post: Webinar: How Claims Automation is Transforming the Appraisal Experience

Back to blog list: See all posts

Next post: AI and claims management: what really matters?

Get updates