If you do a quick Google of ‘TalkTalk customer service reviews’ it doesn’t take long to realise that the British telco doesn’t have the best reputation. Broadband.co.uk, which allows customers to vote on the service of their telco provider, literally has thousands of complaints from customers that have been poorly served by the company.
And this is something TalkTalk readily admits itself. Alex Birtles, head of customer loyalty strategy, was speaking at Pegasystems’ annual user conference in Las Vegas this week, alongside her colleague Andrew Storer, lead architect at TalkTalk, where she said that the company had always had “a lot of big disruptive ambitions”, but had was “not always there with the quality”.
We don’t have the best reputation for service experience. We know our customers have a choice, they don’t have to remain with us. [Thanks to new regulations] they don’t even have to talk to us if they want to leave us. We know we have to do better by our customers if we want to see lower churn.
The pair explained that a big part of TalkTalk’s growth has been via acquisition, which has meant that the company has inherited a lot of different products, price plans and contract types, which has resulted in a lot of complexity. This has meant that there is a long-tail of legacy products that have to be managed and there is a lot of complexity in the company’s systems.
[This is also a challenge] for our agents, they have to understand all this. What that means is that it restricts our agility, restricts our ability to innovate. And worst of all it means that our customers don’t get the best customer experience, because they don’t get a consistent experience.
Making TalkTalk simpler
Storer explained that for all the challenges the company has, the systems it does have are relatively new and aren’t some of the mainframe applications that you’d perhaps see in similar organisations. He said that TalkTalk has a clear separation between its applications and that there is a ‘robust’ service layer in place, meaning that integration between applications can be done quite easily.
However, there are obviously some specific challenges. And although the online platform for customer service and the company's desktop agents are using the same set of services and accessing the same set of data, the processes that they use are very different. Storer said:
The process you go through online is exposing a different customer experience when you speak to an agent. We’ve also got some gaps in our capability in the agent desktop, which means agents having to ‘swivel chair’ to go into some of these back-end systems where they haven’t integrated. We don’t have a case management capability in the systems we have there.
We established a programme called ‘Making TalkTalk Simpler’ to address these challenges.
However, Storer said that this programme was too technology focused and wasn’t focused on developing value to the customer quick enough. So TalkTalk reassessed and started its ‘consumer revolution’, which meant that any technology change could only be successful if it was grounded in delivering value for the customer.
Fundamentally value for money means that it’s a reliable service that works, not just a cheap price point. They do get that we get stuff wrong, but they want us to fix it, and fix it the first time they call us. They don’t want to have to call us repeatedly. If we can spot there is a problem before they have a problem, that’s better.
Most of our customers don’t want to talk to us. When we do talk to them, they want it to be relevant, they don’t want us to continue to flog them products. That’s really where we start from when we think about omni-channel, that’s our service aspiration.
A broken journey
Storer explained that because of the different processes within the different channels, customers have previously experienced a poor customer journey if they ever were to have a problem and use multiple channels of support.
He also said that TalkTalk’s old approach was to create a self service portal online for customers, which he believes fell into the problem of making the solution a ‘tech one’ rather than one that served the customers better.
They’re not necessarily bothered about online, they just want it fixed. What we did was take a step back and say, okay let’s not talk about omni-channel just yet. What we need to do first is build our customers a robust, reliable and efficient repair journey and get them back online as quickly as possible.
The next step is then to say, okay now we have got that really good process, we can now put that available to other channels. IVR (interactive voice response) was the next one for us, which is where the customer phones us and the IVR system can say straight away that the customer has got a fault, they’ve told us they’ve got a fault, what we will do now is escalate a case, initiate some diagnostic processing and initiate a fix straight away without having to speak to an agent.
Third stage is to go full omni-channel, which means the customer can choose a channel. The customer can talk to an agent, can initiate from IVR, do it online, and they can channel hop between all those. The customer can choose how they want to interact with TalkTalk. This means that we have one set of data, one case, one process, one customer experience. That’s what omni-channel means to us.
However, ultimately, Storer and Birtles said that the ultimate goal is to make TalkTalk proactive and try to predict when customers have challenges. TalkTalk wants to be closely monitoring the state of a customer line, so when there is a problem or the line is degrading, then TalkTalk can issue a repair before the issue hits the customer.
This is all being done via the implementation of the Pega 7 platform, starting with MyDesk, which is an application based on a customer service for communications framework and should give agents a full 360 degree view of the customer. Storer said that Pega gives TalkTalk a platform to run the customer processes it needs, handle the cases and gives the teams the management capability.
Pega 7 also means improved integration for TalkTalk, which should solve the problem of agents having to ‘swivel chair’ and dive into disparate back-end systems to solve problems.
Interestingly, Birtles said that TalkTalk doesn’t want to specify what channels its customers should be using. She said that over the years the company has adopted ‘digital first’ or ‘self serve’ projects, which have been about attempting to get the customer to use online in order to save money. Brittle said:
Actually a lot of it, if we were really honest with ourselves, was rebranded cost cutting. It’s about shifting customers into a cheaper channel and being able to take out heads. The reality is it’s not always appropriate for a customer to self serve online.
One of the complaints I get from customers quite frequently is: don’t try channel shift me when I’ve got a broadband fault, I’m calling because I can’t get online. It’s going to take time to build trust in our self-serve channels until our customers feel confident to use them, and we need to respect that.
What we mean by digital first is building a process for digital, for a customer to self serve and then putting that back into the processes used by the agent. Because what we used to do was design for agents. And you compromise on things because you had a person that could tape bits together and flip between systems. Whereas if we design digital first it will be just as good for the agent and allow the customer to move through channels that work for them.
Storer concurred and added that for TalkTalk the most important thing regarding the project has been
sticking to the ‘vision’. He said that it takes a long time to develop that vision, but what he didn’t want to do was put a hold off progressing improvements to customer service whilst that vision was refined.
As such, rather than wait for the strategic vision to fall into place with Pega, TalkTalk made the decision to implement tactical changes alongside the big strategic one. A good example of this was the decision to implement order management exceptions (OMX), which was focused on getting customers that weren’t happy with their service back on to a ‘happy’ customer journey. Storer said:
Our customer journey is really good when things don’t go wrong. It’s nicely integrated, it’s nicely automated. But when the customer drops off the happy path, when there has been some issue with that provisioning journey, we are not so good there and we do have a lot of challenges around how we recover that customer.
What we said we would do is that when a customer drops off that happy path, straight away we will create a case in Pega and have a dedicated case handler. And that individual is incentivised to make sure that that customer gets to a happy state and a fully provisioned working line. It’s also a number of weeks after that that they will still be accountable and look after any early life problems.
That illustrates a really important point to the consumer revolution, because that’s quite an expensive thing to put in place. To have that dedicated handler, it’s quite an inefficient way to manage your resources. But it was the right thing to do for our customers.