Calling the shots


The relentless rise of motor insurance websites has resulted in them now picking up about 50% of the quotes generated, but has the customer experience grown equally as rapidly? Stephen Jacobs tests the quality of the major internet portals to destruction

Over the past five years, Direct Excellence has been monitoring the call centre performance of all the UK's major direct insurers. During that time, the internet as a sales channel has grown from practically zero to approximately 50% of sales, according to January research into renewal behaviour. This massive shift in buying patterns has prompted the development of a web monitor report - an annual review of direct insurers' websites. The first report is focused on motor insurance with reports coming out later this year for home and travel insurance.

The report is designed to provide a detailed examination of each website's functionality and reviews the relative strengths and weaknesses of each company's website using four different types of measure. The first measure comes from consumer focus groups, which involve an in-depth qualitative study of consumers using different websites and comparing them. But they also provide an understanding of how these sites compare to consumers' best online experiences and examine their attitudes to the process of shopping around and switching.

The second measure is in the form of an expert review of each website, based on four different categories: clarity, functionality, convenience and the transaction. Then there is a technical evaluation - a review of the website's fitness to handle insurance quotations, based on five categories: underwriting requirements, the quotation, payment, compliance and support.

Finally, each website's search engine performance is reviewed.

Of the four measurement areas, only the expert and technical reviews are used to score the relative performance of the different websites and create the league tables (see above). It would be dangerous to score focus group feedback because it is qualitative by nature. Equally, the search engine performance results, while interesting, reflect the relative media-buying strategies of the different insurers but this does not impinge upon the relative merit of a company's website.

The four different expert review categories are made up of 25 different measures; the five technical review categories are made up of 10 different measures. Each measure is defined, then scored on a scale of one to five and given a weighting. The total of these weighted scores produces a category score that itself is weighted to give an overall expert or technical score.

This is where the focus group work comes to the fore because many of the weightings are decided by the consumer feedback from the focus groups.

In deciding which companies should be included, the decision was taken to focus on those that appear - from the research into renewal behaviour - to be currently winning most new business in the sector. However, this is obviously not necessarily the same as those companies with the biggest books of business.

Two leading aggregators are also included - namely and Money Supermarket, which currently enjoy a dominant position in the aggregator market with nearly 90% market share. Finally, the decision was taken to include Swiftcover due to the company's heavy promotion of itself as an internet-only business. Therefore, it should arguably be setting the benchmark for the online market having had the opportunity to build a business from scratch.

The same scoring and weighting is used for all the sites - but the aggregators, who do not actually take the customer's money, inevitably score lowest, because how a website takes a customer through the purchase produces quite a large percentage of the overall score. Consequently, there are, in effect, two league tables - one for direct insurers (including brokers) and one for aggregators.

Focus group insights

Feedback from the focus groups reveals that, in terms of renewal, consumers are far more likely to respond to a specific offer - such as a discount or of a quote price, but not in the form of a vague 'we'll beat your current quote'. This is especially the case if it is delivered around six weeks before their renewal date. The preferred way to receive these offers is by post, with e-mail the next best option, and phone being a distinct 'no-no'.

Consumers are also easily driven away from sites by problems and irritations - they are only too aware that other providers are only seconds away (see example comments, p27).

In addition, consumers like to feel in control. This means that assumptions built into a motor insurance website that restrict their choices irritate them. Anything that causes them to lose control of the process will bother them. Consequently, error messaging needs to be very specific, and give customers a clear idea of all the options they have for proceeding. Where a quote is denied, it needs to be explained carefully. Where a transition is made to a partner site, the transition - and the reasons for it - should be explained. Equally, an option to amend what is the source of the transition should also be offered.

Consumers also like short quote processes, but not too short. While they enjoy the convenience, consumers like to feel they are getting a 'proper' insurance product. Needless to say, they really dislike long quote processes. However, customers will tolerate this if they know up front they are getting a range of quotes, or a very competitive one. They will be particularly intolerant if the questions seem pointless or irrelevant, or if the navigation makes the process confusing or demotivating, giving them no sense of how long it will take.

Finally, repeated entry of the same data drives consumers absolutely crazy, and should be avoided at all costs.

League tables

So what distinguishes the winners from the losers in the overall league table? The first thing to note is the domination of the Royal Bank of Scotland brands at the top - taking four of the top five slots. This is partly because it has a product that has been tailored to be sold over the web, not an off-line product that has been inadequately shoe-horned down a different sales channel. It also reflects the distinct possibility that the websites have the same quote engine sitting behind them - therefore, if it works for one of them it must work for all. However, there are differences that enable the brands to be separated in the scoring process. Interestingly, the one RBS site that does not make it into the top five is Tesco; its overall score was dragged down by poor policy information on the website.

The only non-RBS brand in the top five is newcomer Swiftcover, which was given a neutral reception in the focus groups. However, with its designed-for-the-web motor product (just nine questions to get a quote for instance) the website did well on the measurement criteria.

Of those companies occupying the bottom five slots in the overall league table all is not lost. Only two websites received negative feedback from consumers in the focus groups; the feedback on two of the others was neutral, and Budget's was actually positive.

The results of this study, including the focus group feedback, create a number of golden rules for selling motor insurance online but the first and most important rule will come as no surprise: use a short quote process, or rather one where a quote is delivered within the first two or three screens. However, this also implies the need to have an insurance product/underwriting criteria designed for the web channel.

Speed is one of the essential qualities valued by consumers when looking to buy insurance online. The factors that have most influence over speed can be displayed via a traffic light system (see table above) of red, amber and green to make it easier to see who is doing well and who less so.

What stands out is that the RBS brands are covered by a sea of green. However, there are only two companies - Norwich Union and Saga - which impress post-quotation in terms of speed to payment screen, keeping the questions at this stage to a minimum.

The broker and aggregator sites suffer from having to collect so much information and many of the traditional insurers still appear to be getting to grips with the realities of the demands of selling on the internet.

Overall site performance

When it comes to site performance, green and red colours indicate whether the company was above or below the average score in the category concerned (see table, p28).

In the pilot studies for the web monitor report, the clarity category was called 'usability'. However, as it covers such items as plain English; clarity of text; instructions and navigation; how the site deals with second drivers; collecting occupation; and its use of graphics, clarity was felt to be a better descriptor.

The RBS brands do particularly well in this category due to consistency across all the measures. Companies which did badly did so for different reasons. Two scored badly for the way they collect occupation and for the clarity of their text and instructions. One broker got marked down for not using plain English across the whole site and one direct brand for its use of graphics. The aggregators' use of graphics also gave them a reduced score.

Included within the functionality category are automatic postcode to address generation and car registration to car details generation - which almost every website has now got right - in addition to a range of different measures around customer service where significant numbers of companies are failing to provide good service.

E-mail confirmation is one of the functionality measures and is an area that has seen huge improvement in the six months since the pilot study was run. However, of the 15 brands that did send a confirming e-mail, only six got top marks for the quality of that e-mail.

Customer support

Motor insurance websites were rated in terms of support given to consumers. Red means the service is not available, green that it is available and accessible from the quote screen, and amber that it is available but only accessible from the home screen (see table, right).

Every brand reviewed offers telephone support except one aggregator which will only talk to customers via e-mail. Most sites have lists of frequently asked questions but quite a lot of these are not easily accessible.

What is interesting is that one RBS brand is now offering live chat on its site, the only RBS brand to do so. Is this a test before it goes live across all the brand websites? And interestingly, support is one area where the different RBS brands take different approaches. Only two companies offer to call the customer back, but live chat appears to be becoming more widely used - in this study four out of the 21 websites reviewed have this facility.

Search engine analysis

A chart has also been produced to plot 'weighted' (by the market share of the search engine) and 'unweighted' first page returns for each motor insurance website, with the sites ordered by rank from left to right.

The unweighted plot gives some clue as to the types of strategies employed by specific sites. The black line is a hand drawn trend-line (as a calculated regression line might strike as overkill) that gives a sense of which sites are coming out above or below the trend in terms of their unweighted scores (see graph, left). Generally speaking, an above trend unweighted score may indicate a web-marketing strategy oriented to smaller search engines, and a below trend score towards the larger engines, especially Google.

What emerges clearly from this chart is that Money Supermarket and Direct Line are way out in front of the rest of the pack in terms of their web-promotion via search engines. It is also interesting to note that brands such as Churchill, Sheilas' Wheels, More Than and Confused - all of which have a strong presence via television advertising - do not necessarily have a strong search engine presence to match. This suggests that the web is being treated as a separate, and currently secondary, marketing channel by these companies - or that they do not see search engines as the most significant point of enquiry for the consumer. Separate research by Direct Excellence on the use of aggregators and companies' prominence on them gives a good insight into this but it is something to watch as web-marketing continues to grow now that the use of the internet for quotes is about 50%.

So, it would appear there is a definite gap between the very best motor insurance websites and 'the rest'; however, the differences in the scores of the middle-ranking ones could easily be negated by a change to any one of a number of measures or weightings that would see them constantly change places. However, it would not get them a seat at the RBS and Swiftcover table, to do that they need to re-engineer their product to make it tailored for selling online.

When this exercise is repeated in 12 months' time, that's exactly what is expected - seeing more products being sold that have been specifically designed with the needs of the web channel in mind.

This will be harder for the brokers to achieve as they need to collect so much more data, but it is up to them to work with their partners to get them to provide more web-friendly products and help them drive down the time and effort it takes to get a quote. In addition, they could help their own cause by promoting the fact that they are a broker and that the 'reward' the customer will receive for the 'work' they are being asked to complete is a brokered offering.

Stephen Jacobs is general manager of Direct Excellence.


"I think if I was doing this at home I would give up. I mean, there are so many insurance companies, you can just go to someone else."

"It couldn't read the information that we put in, so we changed it and it couldn't read it again. We changed it for a third time and it cancelled and shut down. It just killed the quote and said 'ring us', which was frustrating. Well, I couldn't be bothered."

"I don't like pop ups. Because when it comes up on my screen I have to allow them and then go back. Then you have to see if they are still open and then you have to close them."


Anyone interested in obtaining a copy of the full web monitor report should contact: Stephen Jacobs on 020 8843 7600 or [email protected]

  • LinkedIn  
  • Save this article
  • Print this page  

You need to sign in to use this feature. If you don’t have an Insurance Post account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an indvidual account here: