By 2018, the UK is expected to have the highest share of online retailing in the world, with online sales accounting for 21.5% of the retail sales overall.
With fewer and fewer physical retail outlets around, and the proliferation of devices and connectivity to effectuate online purchases, online support is now often the first port of call for customers. Far from being a nice to have, it is increasingly becoming a defining channel for the customer experience.
Without direct personal contact to the customers, online support is challenging at the best of times. It is not unusual for a consumer tech company, for example, to have hundreds of thousands of support pages covering every possible scenario. No wonder customers often get lost in the maze of information.
In the words of Jeff Bezos of Amazon.com: “In the offline world . . . 30% of a company’s resources are spent providing a good customer experience and 70% goes to marketing. But online . . . 70% should be devoted to creating a great customer experience and 30% should be spent on ‘shouting’ about it.”
This applies to customer support as much as the overall customer journey.
A call says more than a hundred clicks
Online support can be a win-win for both sides: customers get immediate attention and, ideally, a quick resolution to their problems and companies can save a lot of money, which they would traditionally spend on contact centres.
But as with the latter, it is critical to ensure that customers are getting the best service. However, many businesses are struggling to figure out how to measure the effectiveness of online customer care. Why is that?
With call centre support, there is a clearer, natural diagnosis and feedback loop. A phone call determines the issue a customer is facing and helps analyse customer records to establish if the issue was resolved, what solution was recommended, and whether there was a follow up required.
In case of online platforms, businesses only get information about people visiting the support section of the website but not necessarily what drove them there and whether their problem was resolved.
Despite the millions of pages with searchable FAQ sections and self-care drop-down menus, the insights gleaned remain limited or potentially untapped. It may have a survey, but less than one per cent of people actually fill it in. So, how do businesses ascertain the effectiveness of their online support?
The trouble with web analytics
The other challenge is that traditional web analytics – clickstreams, visitors, time spent on site – are often used to analyse online support. This may seem like a no-brainer but this approach suffers from two serious limitations – not being able to discern the customer’s intent and not being able to confirm resolution easily.
People use the online platform for a variety of reasons – be it for research, comparing prices, evaluating feedback, and perhaps making a decision. However, there could be a wide variety of intentions for customers to use a support site, depending on the issues they are facing. Therefore, online support does not follow the traditional sales funnel in the same way that an ecommerce site may do.
If a customer encounters a problem with a product they may not go directly to the supplier’s or retailer’s website. The customer might try to Google a solution first, and failing that, move on to the company’s site. In this case, the company will have no visibility of the customer’s previous Google searches. All they will see at this point is that somebody has visited their site and looked at certain content pages. They won’t know what the customer searched for previously, what steps led him to the site, and what the customer’s problem was in the first place.
Furthermore, clickstream data only reflects the pages a customer visited, not the particular content they may have viewed on these pages. Nor is there any way of ascertaining whether the content helped the customer resolve their issue,- beyond maybe a short feedback question at the bottom of the page.
Discerning customer intent
Big Data capabilities such as text analytics and unstructured data analysis enable us to reverse-engineer the customer journey on a support site. Big Data tools can help analyse all the content on the pages, all the keywords customers used to search, and the topics they searched for. Based on these insights, we can deduce what the original problem might have been and the solution that the customer found on the website.
By simulating the user experience and figuring out the problem, it is possible to identify the user’s original intent when visiting the site, improve the content provided on the site to better match common customer issues, and optimise the visitor path to make site usage more effective.
Traditional contact centre metrics such as first time resolution can be adapted to the online world by measuring repeat visits and tracking the online resolution rate, meaning the percentage of visitors who have their issues resolved online. Since this metric cannot be directly measured online, attempts must be made to decipher information by collecting the device details, tracking at which point the customer goes to the contact page, etc. The site can also have a provision to display a dedicated support number to users who have spent time trying to resolve an issue.
Once the necessary resolution rate is understood, the content on the site can be audited to see if it matches up in response – in terms of time and relevance. The results can then be compared to what the company’s call centre would offer under the same situation.
Validating support site content
We can also look at whether the ‘right’ content is available to solve some of the most common problems. This can be achieved by simulating typical problems (such as how the visitor would use the search option, the keywords, the clicks and navigation).
Another line of investigation would be to assume that the right content is available on the site, and explore the ease with which customers can access it. In this scenario, variance analysis is used to compare the paths customers take to find the right content. The results of such analyses often perplex the companies concerned – even if it is large Fortune 500 firms – as they realise how lost visitors get on their support site.
If it is indeed taking customers a long time to get to the content they are looking for, examining where and why they are getting lost could help. It may come down to using the wrong keywords in their search, or clicking on the wrong links. Accordingly, search terms and links can be amended and moulded to the most common customer navigation paths through the site, and content can be revised to make site usage more effective for the user.
Conclusion
The question arising is whether Big Data analytics can contribute to the creation of automated web self-care analytics, which can be deployed and adapted on an ongoing basis? The answer is that it certainly can.
The best-practice approach would be to build an analytics framework that measures the support site continuously and channels results into an easy-to-use analytics dashboard that will flag up areas of concern.
In setting up a real-time measurement framework, organizations should go beyond tracking URL visits alone. To get the complete picture, they will need to revisit customer intent using text analytics tools, improve on-site resolution rates by adjusting the content provided, and optimize visitors’ navigation paths continually.
A cohesive and integrated customer support mechanism, monitored by real-time analytics, can go a long way toward strengthening a brand – beyond e-tailing peak periods like Christmas.
Sridhar Turaga is delivery leader at Mu Sigma
You are in: Home » Guest Comment » GUEST COMMENT Why great online customer support is so important
GUEST COMMENT Why great online customer support is so important
by Sridhar Turaga
This is an archived article - we have removed images and other assets but have left the text unchanged for your reference
By 2018, the UK is expected to have the highest share of online retailing in the world, with online sales accounting for 21.5% of the retail sales overall.
With fewer and fewer physical retail outlets around, and the proliferation of devices and connectivity to effectuate online purchases, online support is now often the first port of call for customers. Far from being a nice to have, it is increasingly becoming a defining channel for the customer experience.
Without direct personal contact to the customers, online support is challenging at the best of times. It is not unusual for a consumer tech company, for example, to have hundreds of thousands of support pages covering every possible scenario. No wonder customers often get lost in the maze of information.
In the words of Jeff Bezos of Amazon.com: “In the offline world . . . 30% of a company’s resources are spent providing a good customer experience and 70% goes to marketing. But online . . . 70% should be devoted to creating a great customer experience and 30% should be spent on ‘shouting’ about it.”
This applies to customer support as much as the overall customer journey.
A call says more than a hundred clicks
Online support can be a win-win for both sides: customers get immediate attention and, ideally, a quick resolution to their problems and companies can save a lot of money, which they would traditionally spend on contact centres.
But as with the latter, it is critical to ensure that customers are getting the best service. However, many businesses are struggling to figure out how to measure the effectiveness of online customer care. Why is that?
With call centre support, there is a clearer, natural diagnosis and feedback loop. A phone call determines the issue a customer is facing and helps analyse customer records to establish if the issue was resolved, what solution was recommended, and whether there was a follow up required.
In case of online platforms, businesses only get information about people visiting the support section of the website but not necessarily what drove them there and whether their problem was resolved.
Despite the millions of pages with searchable FAQ sections and self-care drop-down menus, the insights gleaned remain limited or potentially untapped. It may have a survey, but less than one per cent of people actually fill it in. So, how do businesses ascertain the effectiveness of their online support?
The trouble with web analytics
The other challenge is that traditional web analytics – clickstreams, visitors, time spent on site – are often used to analyse online support. This may seem like a no-brainer but this approach suffers from two serious limitations – not being able to discern the customer’s intent and not being able to confirm resolution easily.
People use the online platform for a variety of reasons – be it for research, comparing prices, evaluating feedback, and perhaps making a decision. However, there could be a wide variety of intentions for customers to use a support site, depending on the issues they are facing. Therefore, online support does not follow the traditional sales funnel in the same way that an ecommerce site may do.
If a customer encounters a problem with a product they may not go directly to the supplier’s or retailer’s website. The customer might try to Google a solution first, and failing that, move on to the company’s site. In this case, the company will have no visibility of the customer’s previous Google searches. All they will see at this point is that somebody has visited their site and looked at certain content pages. They won’t know what the customer searched for previously, what steps led him to the site, and what the customer’s problem was in the first place.
Furthermore, clickstream data only reflects the pages a customer visited, not the particular content they may have viewed on these pages. Nor is there any way of ascertaining whether the content helped the customer resolve their issue,- beyond maybe a short feedback question at the bottom of the page.
Discerning customer intent
Big Data capabilities such as text analytics and unstructured data analysis enable us to reverse-engineer the customer journey on a support site. Big Data tools can help analyse all the content on the pages, all the keywords customers used to search, and the topics they searched for. Based on these insights, we can deduce what the original problem might have been and the solution that the customer found on the website.
By simulating the user experience and figuring out the problem, it is possible to identify the user’s original intent when visiting the site, improve the content provided on the site to better match common customer issues, and optimise the visitor path to make site usage more effective.
Traditional contact centre metrics such as first time resolution can be adapted to the online world by measuring repeat visits and tracking the online resolution rate, meaning the percentage of visitors who have their issues resolved online. Since this metric cannot be directly measured online, attempts must be made to decipher information by collecting the device details, tracking at which point the customer goes to the contact page, etc. The site can also have a provision to display a dedicated support number to users who have spent time trying to resolve an issue.
Once the necessary resolution rate is understood, the content on the site can be audited to see if it matches up in response – in terms of time and relevance. The results can then be compared to what the company’s call centre would offer under the same situation.
Validating support site content
We can also look at whether the ‘right’ content is available to solve some of the most common problems. This can be achieved by simulating typical problems (such as how the visitor would use the search option, the keywords, the clicks and navigation).
Another line of investigation would be to assume that the right content is available on the site, and explore the ease with which customers can access it. In this scenario, variance analysis is used to compare the paths customers take to find the right content. The results of such analyses often perplex the companies concerned – even if it is large Fortune 500 firms – as they realise how lost visitors get on their support site.
If it is indeed taking customers a long time to get to the content they are looking for, examining where and why they are getting lost could help. It may come down to using the wrong keywords in their search, or clicking on the wrong links. Accordingly, search terms and links can be amended and moulded to the most common customer navigation paths through the site, and content can be revised to make site usage more effective for the user.
Conclusion
The question arising is whether Big Data analytics can contribute to the creation of automated web self-care analytics, which can be deployed and adapted on an ongoing basis? The answer is that it certainly can.
The best-practice approach would be to build an analytics framework that measures the support site continuously and channels results into an easy-to-use analytics dashboard that will flag up areas of concern.
In setting up a real-time measurement framework, organizations should go beyond tracking URL visits alone. To get the complete picture, they will need to revisit customer intent using text analytics tools, improve on-site resolution rates by adjusting the content provided, and optimize visitors’ navigation paths continually.
A cohesive and integrated customer support mechanism, monitored by real-time analytics, can go a long way toward strengthening a brand – beyond e-tailing peak periods like Christmas.
Sridhar Turaga is delivery leader at Mu Sigma
Read More
You may also like
Register for Newsletter
Receive 3 newsletters per week
Gain access to all Top500 research
Personalise your experience on IR.net