Ultimately, we want to get to a place where these complexities are a distant memory, where it is simple and straightforward to create differentiated services. For that to happen, OSS/BSS needs to change and change massively
Automation is key
We must move away from siloed, static, scheduled processes where the vast majority of things still happen manually to automated end-to-end management.
So, if you got a parameter changing in the slice, it’s understood in the network. It’s measured, delivered and rated and billed at that level, and the customer service department will be able to see all of that, end to end, from customer to network, and network to customer, which is a massive change in the way things work, and it’s also got to be dynamic.
This makes sense. Consumer requirements will not change on a quarterly or annual or multi-annual basis. At present, traditional mobile contracts last a long time, with upgrades and changes every 12 to 18 months, but the needs of customers may well change from hour to hour, and providers need to move towards these on-demand requirements.
To get there, we need automation
You can automate existing process whereby you take a process, break it down into steps, and say you’ll automate it. That helps to a certain extent, but in the dynamic world of 5G where you need to support multiple slices, this doesn’t work very well. Different functions should be able to react to things happening in other parts of the network, and apply policies as necessary (“if X happens, I do Y”, etc). But this is difficult because you still have the manual process of defining what the processes are that will need to react to new policies.
There’s still a human bottleneck.
To move beyond this bottleneck, operators must apply machine learning in their networks, so that network functions, for example, can learn from experience what’s happened, the subsequent reactions, and what needs to be done next. The final step here is applying prediction-based machine learning to predict future trends, something that Haslam admits we’re still a long way from. But for CSPs to understand that this is where we are heading, and guide 5G strategy in that direction, is a step in itself.
Selling 5G isn’t all technical
5G technology must be understandable by more than just techies and telcos, and must translate into something, say, a manufacturing company that wants to automate its plant it can understand.
They’ll want an SLA [service level agreement] that shows performance characteristics. They want to know the button that switches your machines off will work every time. It’s about turning those [technical] things into the kinds of propositions that people understand. This is a very different way of selling, and a more difficult proposition. It’s not saying “How many do you want? How much? For how long?” It’s saying “What are your problems? And how can we solve them?”
Sales training will be important and so will building a flexible SLA blueprint that can be developed and backed by flexible contracts.
Everyone knows contracts are very time consuming, even templated contracts take time and effort to get through, so I think this will be of major importance. When you start to look at developing and trying to sell differentiated services, you’ve got to start with those contracts and the SLAs in line with them. Without them, it could take too long and could cause the failure of what could ultimately have been a successful service.
How CSPs are taking services to market
Providers know they cannot do everything themselves. They know they have to work with others to both develop and distribute services, but there’s a huge amount of uncertainty in how to go about this.
Half of the providers surveyed were only sure about the primary routes to market for enhanced broadband, fixed wireless access, improving coverage, and extending their consumer propositions with high definition video or immersive technologies.
They understand those business models and how to build them. But they don’t yet fully understand the new flexible ecosystems and partnerships that will be required to deliver 5G services. How will partners be onboarded? How will operators deal with settlements across the supply chains if multiple partners are involved? Are they moving towards a platform model?
In a previous 5G report about the importance of platform and platform being a killer proposition for 5G – that showed very clearly that you really need to be able to develop those kinds of platforms. What the survey showed this time when we applied it to those verticals, is that we’re moving toward more flexible solutions, partnerships over joint ventures which are much more structured and set in stone, platforms over traditional wholesale channels which are rigid and inflexible.