Good afternoon everybody. Welcome to the next presentation.
My next presentation is by Richard Evans, who works with COS,
and COS is an industry leading agile systems integrator providing comprehensive tech
solutions that cover collaboration, connectivity, networks,
security, infrastructure, and audio visual, and this afternoon,
Richard's going to be talking about how to move critical calling safely to the cloud.
Thank you, Richard. Fantastic, thank you very much.
Oops. So I said, I'm Richard Evans, I'm CTO at COs,
so we're a UK based systems integrator specialising in communications for broadly the
UK public sector, but also, more specifically, blue light and emergency services.
Um, we also look after, as anyone can see closely enough,
my badge says Police Scotland. Police Scotland were due to join me today,
but due to a funeral in force, they can't make it.
But together with Police Scotland, we've migrated the first police force 999 service
fully over to the cloud.
So what I'll do as a result of not being able to hear is I've pivoted ever so slightly,
and I'll take you through the steps that we went through to make that a seamless process
for them, make it risk free, and generally go through the reasons why that's been a roaring
success. First thing I'd do though is um address the
elephant in the room whenever I speak to any force about this subject is um.
The alternative, which is doing nothing and there's a perception that kicking the can down
the road and just upgrading or maintaining an existing premise platform is a risk averse
strategy. So all I really wanted to go through here is
just to highlight that. In my opinion, that is not the case.
So what I've got here is a general life cycle of any system,
right? Specifically here we're talking about your
communications platform that's operating nines and critical coms,
but ultimately what you have is over on the far left hand side,
the general availability. This is where the manufacturer makes the
product, they sell the product and they support the product.
Now what they will do eventually is they will announce end of life,
which means they will first of all go end of sale, which means they're not going to sell it
anymore, they'll still support it.
And then eventually they'll go end of support, which means they're not going to support it
anymore, which means if it breaks, you're on your own.
Now the idea with any premise platform is try and stay towards the left hand side of this as
much as possible, and the only way to do that is constant updates,
upgrades, patches, all of that kind of stuff.
So the point I'm making is with the premise system, every single one of those updates and
upgrades is a risk point. It's downtime, it's impact to live services,
and then the other thing along the top is any new features that become available from this
platform along the time that you'd like to take advantage of,
that's another upgrade. So what we've got is this constant bumps in the
road along the way, just to stay towards the left hand side and then ultimately the further
right you get, the more the risk, the more the risk increases.
So how do we address that then? The first one then,
and I've tried to break this down into 3 steps just to articulate the point.
So the first one is to choose a cloud service aligned with your objectives and largely the
objectives around critical coms.
And the government kind of did half the job here um over 10 years ago in 2013 when they
launched this cloud first policy.
Now I'm aware that within critical calling and within the public sector,
there's a very, very clear brackets after that which says where appropriate.
But ultimately, this is what they're trying to do, and it's created this dynamic at the bottom
we've found with pretty much every single customer we've spoken to,
and this isn't unique to blue light, this is the same in NHS,
it's the same in government.
You've got a kind of strategic pull versus the tactical limitations and the actual
reality of trying to make that happen.
So from an organisational strategic level, you've got the government initiatives and the
organisation in general trying to pull everyone towards the future,
and behind that is AI, it's digitization, it's cloud, all the good stuff that you're here
talked about in any kind of presentation of Earth salt,
but specific to critical calling and specific to the um UK public sector,
we have those two main things holding it back.
Um, the first one is compliance, so here we're talking about stuff like UK sovereignty of data.
Beyond that, we're talking about security compliance, we're talking about being able to
meet government security classifications or sensitivity of data,
that type of stuff, and then you've got emergency, which broadly just means it needs to
always work. And what you'll find is with pretty much every
single cloud service out there, what they will offer is at best a 49s availability,
so 99.99% availability, which is a few hours per year,
but is good enough for most kind of retail and commercial customers.
When we're talking about critical coms, it's not good enough to meet that.
So as a result then, we've built our COs cloud platform specifically for Blue Light and
emergency services, and it's to meet these kind of key points around it.
So the first one I'll draw your attention to over on the far left hand side is high
availability. We built this with 99.999, so 5 9s availability,
that's about 5 minutes a year downtime.
The reality is in the 4.5 years this has been available, it's 100% up time,
it's never been down. We have acute hospitals on this,
every single one of those has got cardiac event 22, 22 call systems on there,
and there's a really, really simple 1 to 1 ratio in terms of call not happening and a life
being lost. So there's a reason it can't fall over.
There's a more cynical reason in that we're on the hook for service credits for a lot of these
as well, so if it does fall over, we owe people money.
Um, the other one in the middle, UK sovereign, it's hosted from our London and Manchester data
centres. It makes sure no data leaves the UK.
Every single member of staff at CNOs that looks after it is a UK based member of staff,
and every single member is at least DV cleared.
We have people that are MPPV cleared and SC cleared to look after it so we can ring fence
them if we need to.
But ultimately what we've tried to do is build a cloud that nips all of these problems,
all these um these kind of speed bumps that were stopping people getting to where they
needed to. Um, the other thing as well, it supports
natively out of the box government security classification of official and official
sensitive and through partnerships we can push that to secret and top secret if you really
want to. Um, and then the last one over on the right
hand side we've got open standards. So we based the entire platform on Cisco and
it's a joint proposition we brought to the market with Cisco.
It's a Cisco powered service which without boring you to death,
means we've been through a big long audit process with Cisco to enable and we kind of
have to jump through a whole load of hoops for that for them to rubber stamp it and carry the
badge which says Cisco powered on it, and then it represents a joint proposition with the
brand. Um, the reason why we chose Cisco, it's a
question I get asked when we talk about this with customers all the time,
is, for example, the elephant in the room is why didn't you choose Microsoft Teams,
for example, you know, with alignment to the NEP and that kind of thing.
Um, open standards, that's the main reason. So Microsoft,
for example, uses proprietary codecs in its technology, which means if you want anything to
play nicely with it, it has to be on an approved Microsoft list,
or you have to go and sweet talk Microsoft into getting it added to that list.
What Cisco does is align to industry open standards, which means anything else that
aligns to industry open standards, you can just integrate with it out of the box.
And one of the key drivers for that was talking to forces,
well, police forces, fire and ambulance across the board who all said this needed to be an end
to end solution. You can't just drop a phone system in and walk
away, it needs to talk to all the other systems in, in place in the control room.
And so what we wanted to do was make sure you had the best possible platform for onward
integration. Now the two speech bubbles, I'll draw your
attention to as well, is these are the two key like kind of benefits that we were keen to
drive as part of it. So almost a byproduct of it,
but specifically for control room and nines. So the first one,
the need to meet SLA.
So we are a SIP PSTM provider, which means we peer directly into the BTIP exchange.
So in the context of 999 after a member of the public calls 999 and speaks to the BT call
handler who says what service.
Once they release that call, it's straight to us.
There's no middleman. There's no third party provider in the way.
It goes straight to us and we can just pass that straight to the operator in the control
room. It means you hit SLA every single time with
every single call. And there's a context for that as well in that.
If you do have an SLA that's how many, you know, answering 90% of calls within 10 seconds,
for example, and BT have just chewed up 5 to 6 seconds with their first call handle a bit of
it, it just means you've now got even less time to do that,
so it's more important that we don't have all of these different parties in the way that the
call needs to be handed off to along the path before it gets to the agent.
Uh, and the last one, staff need help. One of the main things we get told all the time
in this space is that control room staff are expensive to train,
it takes a long time to train them.
And the other thing as well as soon as they finish training,
go straight onto the front line and take what could be the most harrowing call of their life
on the first day. We've got the tools and systems in place to
give them AI support on that. We can suggest next action to them.
The AI can type up all of the call notes for them automatically and just give it to them to
cut and paste. But similarly, we can use elegant tiered
routing which makes sure if you are first day off your training,
you don't go straight on the front line. You're perhaps tiered below more skilled agents,
so a call will go to a skilled agent first and then go to your your more junior member of
staff. But all the time as well, it gives all of the
staff the ability to tap out as well, so let's face it,
the staff go have a hard time of it as well. They have the ability to just tap out and take
5 minutes after a call if they need to collect their thoughts,
and then the call will reroute to someone else.
The idea being though is what we're doing is we keep you,
you know, we're trying to say to the customers here is that,
and, and especially with Police Scotland, it was, we hear you,
we get where you're trying to head, but let us support you on that journey by recognising the
past and the existing investments and the existing systems you've got in place along the
way. Step 2 then is choose a provider with a proven
track record, and this is basically going through how we made this happen for Police
Scotland. So what I've got here is two systems.
The first one on the top is the current communication system.
That could be any system that's in place at the moment, and then the one along the middle is
our cloud for blue light.
Now the first thing we do when we look to implement a new cloud service is we build it.
That sounds fairly commonsensical, but the idea is we're building it in parallel with the
existing platform. That means we can then implement it alongside
the existing one, but then this is the key bit.
We can do all of the integration, all of the testing, all of the training done with the two
systems in parallel, and what you'll notice is the phones and the agents are still on the old
system. They don't even know anything's happened yet.
You can get the entire system in, all the tyres kicked, all the kinks ironed out,
absolutely everything done before you then implement migration and move them over.
And the beauty of it is with the with the systems both in place,
integrated, talking to each other.
Um, you've got a really, really slick rollback option as well.
If there are any problems, you just move those back over to the old system,
it's still there. But ultimately, because the arrow ending there,
that's your point where you have a go, no go period with the customer and then you can say,
right, OK, that's where we want to switch off the old system and move ahead.
The key thing from that point onwards then is this is where you then execute or rather um
manage service kicks in with continual service improvement and we can roll out the AI and
digital service on top of that.
Now what we've done is structure this into these two phases of transition and transform.
So what we're trying to do. I transition over to a new cloud platform that
as much as possible mirrors the existing system, because we found that if you change too much
day one, it'll just bamboozle all of the users and they won't accept it very well.
So we keep it as close as possible to the current platform and then once we get to the
transform stage of it, which is the 3rd arrow of digital services down here,
that's when we Start rolling out the bells and whistles and the extra added value features.
This is where AI omnichannel overlay, so this is where you can start doing stuff like for 101
services adding AI enabled chatbot to field some of those and skimming off the top of the
999 queue, some of those less urgent calls that can perhaps be self-served or dealt with by by
AI. Oop oops.
Step 3 then is the service you choose needs to have scope for expansion.
So once again it can't just be something that gets dropped in it's a phone system and that's
all it'll it's ever going to do.
So ultimately what we've tried to do is build a suite of services that then complement that.
So first and foremost, going back to the open standards for integration,
we play nicely with all of the control room systems in place,
so we have partnerships with Sopra, with Saab, with Zettron,
all of those systems that would be in place from a CAD or command and control system
perspective. When it, whether it's a third party calling
platform, so using Police Scotland as the example, they had 180 different phone systems
that we needed to integrate with, having that solid base of open standards meant we could
integrate all of them from day one and have this centralised orchestration point.
The managed service element of it is part of the kind of dynamic as you move from a
premise-based system to a cloud is that you're now moving to a system that needs to grow and
evolve with your organisation and with the public services that you're trying to serve as
they evolve. So with our managed service and our continual
service improvement part of that, which is where we make sure that it's always on the
latest version, always got the features that you need,
any new features that do come out and you might be of use,
that they get rolled out.
With that, it just makes sure that a system that's rolled out on day one is still
applicable on day 4000 or whatever it might be.
custom application deployment. So with Cisco as the basis,
we have a whole developer team here that's dedicated towards building applications that
are customised for the environment or customised for the use case that a customer
might have. The example we've got here is we've built a
negotiator app for policing that allows a quick.
Aggregation of people when they need to get together in a negotiator scenario.
So the situation we were given by the police force there was that currently when a
negotiated situation needed to happen and you need to bring negotiators,
translators, all those types in, it's a very manual process of conferencing lots of people
in and then if the perpetrator on the other end does hang up the call,
then everyone has to dial in and be conferenced in again.
We've built an app, or we're in the process of building an app in order to address that
specific requirement. But the idea being is that you've got this
solid base of Cisco open standards on which we can execute however we need to.
Control room AV, so half of the COs businesses were a control,
uh, so we're an AV systems integrator, and we've done control rooms in numerous police
forces and airports across the country, as the picture suggests here,
so in addition to the kind of back end communications, we can handle the actual
in-room situational awareness piece.
Um, video and content collaboration. So the first bit that I've got standardisation
of video spaces. We've done this for a few forces now because
again they've come from a standpoint of we need to align with an NEP programme and roll out of
Microsoft Teams for meetings.
We've standardised all of those meeting spaces from really large conference rooms right the
way down to small huddle spaces along the lines of Microsoft Teams.
Mobile and body worn video, including meetings. So Police Scotland,
for example, they've rolled out a video app on all of the mobiles across the police force.
So for example, um, when some of the officers had to go out to Euro 2024 to keep tabs on the
Scottish fans over there, um, they had a video app on their mobiles which they were using to
message, meet and communicate all while working in another country,
and they found that to be a really, really big success, and we're actually seeing a huge
uptake on the number of people rolling that out on the mobiles now.
Uh, and then the last one is, um, we have another app that we've built there for um
domestic abuse. This allows, um.
A domestic abuse victim to then start a video workflow that can be however you want to.
You can start it via a URL or you can start it by QR code,
however you need that to kick off, but it'll start a discreet video session with um someone
within the police force, but that it's kind of aware enough to know that in those situations,
normally the abuser is within earshot. They're normally sat next to them.
So for example, the person on the other end of the call can push discrete messages to the
video feed saying can you speak, can you not speak?
They can take a statement during the call and get them to e-sign it while they're on the call.
And similarly, if that, if the phone gets snatched out of their hand and the video
session drops, it just um refreshes to BBC website as well.
So if they do snatch it out of their hand, it's just BBC that we're browsing.
All of that though together means that what can start off as a telephone platform can then
start adding AI, start adding video, start adding standardisation,
and then grow to bring all communications into one piece.
And I think that's about me.
If you'd like to learn a bit more about our whole suite of services,
as I say, we've built an entire cloud for this rover on Stan J1 over the far side.
Thank you, Richard, any questions for Richard?
Any Thank you, Richard, a round of applause to Richard.
Thank you. Thank you.
That concludes this section.