Tip:
Highlight text to annotate it
X
[applause] Ladies and gentlemen, from EE, please welcome
the head of customer technology & innovation, and customer service, David Walker.
[Music]
Good afternoon, I think it's just me standing
between you and your lunch now so hopefully you will find this interesting. It's the story
of our journey of knowledge transformation with eGain in EE. First I want to start with
a little bit of context around what is EE, for those of you guys who don't know, and
a little bit about customer service. It just gives you a bit of an idea of scale that we
have done the transformation on. So firstly, we're the biggest and fastest network in the
UK. We are now part of the BT Group which is the largest and most advanced digital communication
company in Britain, delivering mobile and fixed communications. We have around 550 retail
stores and service more than 31 million connections across mobile and fixed networks. EE runs
the UK's biggest and fastest mobile network, pioneering the UK's first, super-fast 4G mobile
service in October 2012. We are the first European operator to surpass the 14 million
4G customer landmark. EE's 4G coverage today reaches more than 95% of the UK population,
with unique, double-speed 4G reaching 75%. EE's 2G coverage reaches 99% of the population
and 3G coverage reaches 98%. EE's super-fast fiber broadband covers around 80% of the UK
population and ADSL 98.7%. A brief history of EE, it all started off in Bristol and Borehamwood.
EE was formed in 2010 following the merger of Orange and T-Mobile, at that point in time
we were called Everything, Everywhere. T-Mobile began as Mercury one2one founded in Borehamwood
in the early 90s and then rebranded to one2one before being purchased by Deutsche Telecom
in 1999 and re-branding to T-Mobile. Orange was founded in Bristol in 1996 and became
the youngest ever company to enter the FTSE 100. It was acquired by France Telecom in
2000 and bought Wanadoo to add fixed broadband alongside the mobile offering. We're investing
more than 15B building Britain’s biggest network. Just a couple of stats, we are UK's
first 4G. We have 4G in more than 600 cities & large towns and more that 6000 small towns
& villages with 95%+ population coverage. We are expanding 4G coverage by around 2000
square miles each month and have more than 14 million 4G customers. We have the UK's
biggest 3G network also and we are continuing to invest and grow month per month. A little
bit about customer service now, we have 6 in-house sites dotted around the UK. We have
partners in the UK, Ireland, and India and we have around 10 thousand advisors. 80% of
those are within the UK & Ireland with the remaining 20% in India. Let's move on to the
really interesting topic which is customer service transformation with eGain knowledge.
I'm going to run you through the challenge, so why did we do it? What was the reason we
embarked on this journey? Second, what we did & what we delivered. Third, what results
have we seen? I've got some results to share with you on the success we've seen. What we
learned and what's next. So to get started, the first challenge we had was we had a huge
variation in performance. We looked at that and why we were getting such a variance because
performance is a mixture of people & process. Is it the people delivering great service
regardless of the process or is it the process delivering great service regardless of the
person? That's to say when we looked at range in performance we had huge variance on customer
satisfaction. We had huge variance on handle time of calls, we had huge variance on answers
given to customers. We had a little bit of a problem because customers were calling us
back to check if what we have said was correct. Because of the variance you sometimes would
get a different answer when you phoned in. So you could ring in three times and get three
different results, which isn't great because there is only one right answer generally for
these things. What we found, and this was really helpful later, was our best people
could crack it. So our best agents were absolutely bringing the numbers in with the right handle-time
and delivering a great customer experience. We used this later on in the story as I continue
through because using that knowledge we got from the best agents really helped to create
really compelling content to look through. The second challenge was, as I said the journey
where we've gone through T-Mobile & Orange combining, we had four knowledge tools. We
had a T-Mobile knowledge tool, we had an Orange knowledge tool, we did diagnosis in Orange,
and we had also introduced a fourth tool to add to that. All of these tools were based
around searching predominately with the exception of the diagnostics. What we found when we
measured it was we were getting around 10% consistent use of knowledge. So our advisors
were using knowledge correctly and accurately and consistently around 10% of the time. The
knowledge information was inconsistent so when we crossed the different systems, the
same process might have a number of different instructions. We were using tick sheets to
capture contact reasons and we found those, I don't know if you guys found the same, pretty
problematic. Years ago, we thought we had a major problem with the answer phone. Just
so happens it was at the top of the list of the selected reasons. So we were worrying
about that and in reality that wasn't the case at all. We had, believe it or not, over
20,000 articles across the knowledge base. Now when I first started we had a flip chart
on the desk that had twenty pages in, with all of the help and knowledge you need. Now
20,000 articles, if you printed those out and taped them together you can hang them
from top of Big Ben, you can hang them on the BT Tower, you can hang them on Tower Bridge.
Before you got arrested for doing that, essentially. It's massive, it's absolutely massive. It's
completely where technology comes in and helps us here. Why did we choose eGain? Obviously
we went through a RFP process as we would normally do. The key reasons for eGain: Number
one, it had the capabilities we wanted to drive us forward. Secondly, we really felt
the eGain team had a desire to work with us and help us realize our ambition. Thirdly,
to collaborate as we grow. This isn't a journey that we want to stop, we want to continue
with [it]. So we wanted some collaboration with that, on the route. That's what we feel
eGain brings to the table. What we did and what we delivered next. So we got the inspiration
from the NHS and the processes they use on the grounds that we cannot afford to get it
wrong. We had a real problem with getting it wrong and we absolutely wanted to crack
that problem. We looked at what other organizations, such as NHS, were doing to actually get it
right. Like most things, I don’t know if you guys do the same, but we have a habit
of naming systems and our advisors love to be part of that process as well. We called
it Albert, no abbreviation for anything it is after Albert Einstein because we felt it
was real expert knowledge in one place. The key features we've implemented, summarized.
We pass in customer and device information into the knowledge. We give the advisors the
ability to pause and resume sessions. We do this so that if we need to break a process
with a customer, so we need them to stop and go away and do something, we can allow them
to do that and come back without having to start explaining things again which was a
problem we had previously. We also save the guided help session into our CRM system so
anyone across the company, wider than customer service, can see what has happened there.
This is a quick shot of the system as the advisor lands on it. There is a step before
this where the advisor puts the phone number into a system we call Toolkit that our IT
teams developed and then they are launched into ALBERT. What that does is that it transfers
the customer's information into the knowledge system and we start a journey there. There
are a number of different sections you know, customer service, billing, network. It's the
primary issue the customer is ringing about initially that we go through. As I mentioned,
pre-answering questions is really important because we use it for two purposes. Number
one, it helps us minimize the number of questions we need to ask our customers. Our customers
think we are daft if we keep asking them things that we already know about them. The second
reason, it's really useful for reporting at the end because we've got a snapshot in time
of some key information there. What plan is the customer on, what is the balance, those
sort of things. That becomes super helpful later, which we will go through. What we present
to the advisors is pretty much a three column screen. The first column, quite a dry example
here but nothing contentious in it. The first column is the question. There are two or more
answers to the question and the advisor goes through and answers a series of questions
which drives, ultimately, to a resolution. The second column is the advisor help. Now
what we haven't done is we haven't taught our advisors to work like robots and script
everything. It was very important to us that we didn't do that because it creates and unnatural
flow of conversation. There are things at times, such as regulatory information, that
we need to read out to cover it often in which case we do script it but predominately we
give the advisor the ability to deliver great customer service and give an engaging flow.
What we have in there is some useful hints and tips for the conversation. We also then
help the guys how to do it in the system, so the second bit, our CRM system is called
Excalibur. The instructions are there on what to do within the CRM and that's really helpful
in terms of saving training time because you've got what you need to do right on the screen
there. We also cover any other useful hint and anything else for the customer in that
middle section. On the right-hand side, we have some control buttons. Feedback I'll talk
about in a moment. Restart, so if you've gone completely down the wrong route you can start
right back again and move through. We also have what we call "Albert Breaking News."
This isn't information on content, these are things to add a little bit of context around
it. We make sure that's visible all of the time in case there is any relevant that comes
up outside of the guided help. Onto feedback now, what we've found really important is,
it is absolutely imperative that we have the content correct. It's really, really important
on that and if there is something wrong we need our advisors to tell us so we can fix
it. I'll certainly review it and see whether it is incorrect. So we have a feedback button
on every page so if there is anything wrong or anything needs improving on those articles,
you simply hit the feedback button and that feeds back through to our knowledge team.
What our knowledge team do there, and we have a very tight SLA on that so typically we will
get back to an advisor within 24 hours. When the advisor feeds the feedback in, that is
reviewed by our content team and they make a decision. Is it something that is in error
in the content or is it something simply that needs to be fixed or is it something that
has changed recently that we haven't updated. If it's a straightforward change, the content
is updated and a response is sent back and we start that cycle again. If it's a more
complicated change that needs further review, we hold sessions with a more senior audience
and process teams to make sure we thoroughly review it. Again, we don't want to get it
wrong, it's vitally important we've got the right information here. We find that really
effective and the feedback loop is really important because it encourages people to
feed back. We get several hundred bits of feedback a week of which 10-15% of those actually
result in change, mostly it is for clarity of information. On to what we delivered effectively
through the rollout. We deployed the product and we deployed the content and we trained
over 10,000 advisors and three retail stores. So we used this for customer service & retail.
From a knowledge content perspective, from our 20,000 individual articles to ten guided
help case bases. Ten is not necessarily a number you want to aim for but we've found
is that it is the right balance in terms of the number of case bases to set up. We did
a complete overhaul and rewrite of our knowledge content and set standards up for that. That
ensures that we've got consistency in the way it is presented and that really helps
our front-line advisors. We have clear content ownership and accountability. Previously we
let the process teams write a lot of the content but what we found is it wasn't particularly
customer or advisor-savvy. We actually used our best advisors to help us write this, back
to that quartile slide, there are people there that really know how to do it. They also know
how to articulate it so if you share that information you can make a huge difference.
The accountability for knowledge was really key here. The reason we had the challenge
previously within consistency is accountability had drifted off over time. We have a single
knowledge content team for which we have a core team and we bring in advisors to make
sure we are fresh. The core team also spend time on the phones because it is vastly important
that we keep that knowledge fresh and keep the people fresh to what's going on within
the contact centers. We also now capture automated contact reasons. We retired two, previously
widely used, capture tools. All of this is done automatically, there is no box to fill
in, there is nothing to click, nothing to get wrong, it is done as a part of the process.
More on that in a little while. What we found. In the frontline we found that first contact
resolution had improved. We did a case study to measure this specifically because we've
done a lot of transformation and change at the same time. As you guys know, it's quite
difficult sometimes to actually work out what has driven what change. When guided help is
used correct we were measuring FCR on 85% of calls. When it was not at all, the old
way, 62% so a huge increase there of 23%. We found also that guided help contributed
to a 20% increase in net promoter score. We measure, as we discussed earlier, customer
satisfaction. At the end of each call we send them a text and use that to measure customer
satisfaction and we use NPS as one of the methods for that. We've also seen complaints
reduced as a result of this. From a training perspective, when we get new hire and new
intake, guided help has reduced our training time from at least 14 working days down to
8. We've not compromised the training there, we are just teaching the guys how to use the
system rather than cramming their heads with processes, procedures, devices, and things
like that. We've lightened that up now because all of the answers are in the guided help.
We've improved speed to competency, this is the time from a new advisor joining to being
competent at serving customers effectively, from six months, which was often extended,
down to three months. A 50% improvement there. This has been huge in terms of training. We
found we had another problem then, we needed to get our logins done much quicker because
we were getting people through training and had to get our HR set up. It's interesting,
one problem fixed actually drives the next opportunity. From a frontline perspective,
just to summarize what we saw. From a customer's perspective, consistent information is given,
if you ring back up to check it's correct you will get the same message. We have a better
customer experience as a result of that because we are giving the right information at the
right time through a structured method. That has really improved our customer service,
the results we covered a few minutes ago. From a business perspective, we've improved
first contact resolution significantly. We've aligned the handle time; we've moved from
massive variance in handle time. We've reduced complaints, and we've reduced training. That
sounds a little bit negative "reduced training" doesn't. it but we've made training much more
effective and made the people more effective as a result. From an advisor perspective,
they have the knowledge at their fingertips and their KPIs are achievable. If you follow
the guided help, you will succeed. Very straightforward. Advisor confidence is much stronger because
we are not scrambling around trying to search for things and hunting through the fourth
page of search results. We found search pretty challenging from that perspective because
you really needed to know how to search and you no longer need to. Beyond the front line,
this is probably the one that, personally I thought we knew we were going to get this,
it was common sense this is what we were going to get. Until you've actually got this data,
you don't realize how powerful it is. We use the call reasons. So the agents follow the
guided help, the data gets transformed, we attach some additional metrics to it, so it's
joining data together. We attach the call center metrics, the satisfaction survey, etc.
Then we've got insights that answers to particular problems. Number one, it gives businesses
awareness of call drivers which we didn't have effectively previously. Secondly, people
and performance management to make sure we can continually evolve and make sure the advisors
are working to best-in-class. We have company-wide program and we focus on what we call "propensity
to contact" (PTC), The number of times a customer contacts us each year. We effectively split
these into two categories, one is what does customer service own, because that is our
problem to fix, and what other products, services, and propositions that are business owned.
We do regular weekly tracking and we have a company-wide program to actually drive these
initiatives and make sure we are making improvements and to make sure we are making the right decisions
from a customer service perspective when we are launching propositions. Previously our
marketers would have fantastic ideas but we didn't always think through the customer service
element. The data that we get out of the back of eGain joined with other information is
super powerful. We have clear accountability for this and that's one theme we've had all
the way through from content to improvement. It's all about real, clear accountability.
What we found through the drive to reduces PTC, we grouped the call reasons into business
functions and tracked them weekly. Very frequent tracking, very real-time information. Established
for each of those, the clear accountability and ownership. We got 100% buy-in across the
business, we had top-down buy-in to make sure that happened. People couldn't hide away anymore
if something happened, they had to follow-up and fix things when we had any issues. They
key learning we found was, we had the most successful period of reducing PTC following
implementation and making sure we had the accountability because we had the tangible
data there to make it happen. That really helped us from that perspective. We also allow
analysts across the business to deep dive call reasons, so everyone can get under the
bonnet and see. Previously we did not have this capability to do that. Now that we have
this rich information we can really get down into the detail. You can see the difference
between particular handsets on performance for example. We had a great piece come out
very early on this. We had picked up a variance on performance with a certain handset from
a certain manufacturer. We pulled that out, contacted the manufacturer and said text message
problems seem to be completely out of the norm for your particular devices, it’s completely
different from what we are seeing across the rest of the range. We fed that in and we also
fed in some insight that we got from it with a little bit more detail. Within around a
week, that manufacturer had updated the software on the handset, pushed it out to the customers,
and fixed the problem. You would never find that previously because amid 40 million other
calls per year it's a needle-in-a-haystack, we are talking a few hundred calls here and
you could find that data and that has been super helpful. It allows you to pick things
up that you just wouldn't find previously. We essentially week to week, month to month,
rank the call reason information that we get out of eGain tied to the other data. We put
that in a league table to figure out what we need to focus on to get the biggest impact
because you can’t change everything at once. We break it down into the key things that
are going to give the biggest value return. We cut that by first contact resolution and
customer satisfaction results and that helps us prioritize our continuous improvement activity.
What we've done as well is we have linked the contact reasons to the call recording.
Joining up the data gives you massively more powerful information and we allow our agents,
our managers, and our quality teams to listen to those call recordings. One of the key things
we did as well is we gave our advisors the ability to listen to best practice calls.
So with the contact reasons now, we can pick out calls that are really good quality in
a certain specifically complicated thing. It might be a signal issue for example. Those
tend to be quite complicated processes because there are a lot of variable in there but we
can actually demonstrate to the advisers what good looks like. By joining data up it makes
a huge difference in what we can achieve there. What we learned. These are the things we kind
of picked up that we weren't expecting as we came through. First point was, as I mentioned
earlier we did a wide customer service transformation program at the same time, one thing we did
as a part of that was implement dual screens. Dual screens really helped us here in combination
with the knowledge transformation because it allows the advisor to work on the left
or the right screen with the guided help, walk through there, and execute the transactions
on the other screen. That was a really great partnership with guided help to make that
work. Secondly, as you can probably see, reporting was something that was really, really amazing.
We knew we were going to get it we just didn't know how powerful this was going to be to
help us both within customer service and in the larger organization. Completely transparent
reporting back to the advisors to ensure they understand how they're performing. We have
a target of 90% guided help usage, which is regularly exceeded. There are obviously odd
calls where it’s just a random call that you wouldn't use it for but we set targets
there so we help the team managers and the advisors ensure they are following that correctly.
The other thing we found and it was probably a real biggie that we didn’t expect to be
such a challenge, we moved from search to guided help and it was a huge cultural shift,
especially for our long serving advisors, to move between search and guided help. It
did take quite a lot of work and there was quite a lot of stick there to keep it moving
forward. What we did find when we measured it, you could prove it statistically time
and time again, the people, long-term or new starters, that were following guided help
had significantly better performance metrics than those that didn't follow it correctly.
It kind of solved itself then using the metrics back to the point above. The final point is,
IT resiliency. People who have been around awhile still have quite a lot of knowledge
in their head but when we get into IT resiliency, it's really important that if anything goes
wrong, if you mess up your content deployment or anything like that, you've got some form
of standby that you can switch to because the guys are flying blind if they lose the
system. Really, really important there to make sure you've got the contingency in place
if anything does happen to go wrong. So what next? This is our kind of logical next steps,
some of this is. So the first thing, I mentioned earlier, we're really keen not to deploy something,
say hey this is great, and sit on it and effectively not move it forward. Our current plan is we're
looking to upgrade to the latest version and make sure that that's effective for us, we
exploit any new capabilities, and keep it current. The second, how can we leverage guided
help for our customers. We've done some great things within the contact center, the logical
next step is, how can we give that to our customers and what will we need to do to do
that. So that's something we are really, really super keen on at the moment. As I mentioned,
they key theme is let's keep current, let's continue to improve, and let's keep exploiting
what we've done and what we've achieved. We can't sit on this one, we've got to keep it
moving forward and continue to build on it. That's what our next steps with eGain are.
Thank you very much. [applause]