Learnings

Outcomes

 Phase 1: current trends and expert thinking

During this phase of the research DXC produced prototype solutions which illustrated the existing capabilities of current technologies and how they could be adopted and developed by DNOs to significantly enhance their customer servicing capabilities in the future.

The employee engagement workshops carried out in this phase provided information to allow DXC to produce a series of “user stories”, a set of user-centric requirements identifying: 

·       the persona that will benefit from the requirement 

·       the action created or generated by the requirement 

·       the value provided by the implementation of this requirement 

Analysis of the user stories identified consistent themes and overlaps, which allowed DXC to group the requirements and ideals into nine broad, ‘high-level’ user stories described as ‘epics: 

1.     As a vulnerable customer, I want to receive extra support during an outage. 

2.     As a customer, I want to use backup power storage. 

3.     As an Electricity North West employee, I want to enhance business operations. 

4.     As a customer, I want my smart meter to offer services and integrate with other smart devices. 

5.     As an Electricity North West employee, I want an integrated system of partners. 

6.     As a customer, I want to interact with my DNO via an omnichannel experience. 

7.     As an Electricity North West employee, I want to make data-driven decisions. 

8.     As a customer, I want to receive Communication updates from my DNO consistently. 

9.     As a customer, I want to interact and receive services from my DNO. 

From these, DXC constructed three prototypes, which allowed customers to experience the conceptual network of 2027. Each prototype represented the same parties but from different viewpoints. 

·       Prototype 1 established the customer domain and provided an understanding from this perspective. 

·       Prototype 2 acted as an intermediary communication platform between customers and DNOs. It demonstrated the potential for automated, robotic responses to customer enquiries managed via service desk knowledge repositories and learning machine technologies. 

·       Prototype 3 provided a view towards the evolution of network visualisation and real-time event/network management. 

Prototype 1 – smart energy hub

To construct a view of the customer service demands for 2027, a realistic model of the domestic energy system must be created to understand the activities that the energy network will be supporting. The analysis of current consumption patterns shows a close relationship between the daily work/life-cycle for consumers. Peak demand times coincide with the standard patterns of life, morning time demands for heat, food preparation and showering, the mid-morning to late afternoon reduction in demand, and then the peak demand of evening time activities. 

Prototype 1 modelled different house types; each house type is constructed from zones which mirrored the primary activity being conducted in that zone. Associated appliances supporting those activities provided the bottom level view of consumption metrics. 

Prototype 1 visually demonstrated to customers how they consume electricity including the monetary value of electricity used by specific appliances and areas of the home, in real time, see Figure 8.1. It also showed when and how much is available in terms of generation and storage. This could allow customers to self-manage their electricity and sets a picture where independence from the grid can be achieved with relatively low cost or impact to the householders.

The objective of Prototype 1 was to provide a basic visual, interactive tool for customers to gauge reaction to this vision for 2027. This prototype allowed us to investigate perceptions around whether the extensive adoption of smart home technologies is regarded as credible and explore customers’ views about whether the changing demand and generation will materialise to facilitate more manageable energy networks. 

Prototype 1 also provided a conceptual data of energy consumption in the home that can potentially be shared with the DSO.

Prototype 2 – chatbot

Customer service must be empowered to move from being reactive to proactive. The ability to predict network events and how they will affect individual customers can be gained from the data generated in Prototype 1. 

The automation of customer service through the use of chatbots, AI and the transformation of knowledge into consumable data services, offers DNOs a scalable and customisable solution for service development. The DNO can be represented by a constructed personality that consistently and accurately serves the organisation’s policy towards customer service. 

Where prototype 1 acted to demonstrate the level at which granular data can be acquired and critically, could be made available to the network operator; prototype 2 demonstrated how the fully automated service desk will capitalise on the level of customer profiling, facilitated by engaging with device services and home system insights supplied by the smart energy hub.

Prototype 3 – control centre hub

Prototype 3 demonstrated a platform that the DNO might use to visualise event-driven network communications. This is capable of integrating multiple data sources and real-time event mapping, through on-boarding ‘edge of the network devices’ to extend the traditional grid boundaries and produce a geo-visual data representation of system events. 

System events are categorised and associated with the extended entity profile of the network. Access to real time, accurate data provides the network operator with infinite possibilities for mapping network events, conditions and external factors that might influence or impact on the network or the customer, for example IOT, smart devices, mobility team devices, weather services, network status, event timelines and network maintenance and employee tracking and coordination. 

Prototype 3 provided a vision for centralised communication and control for network management. It exhibited the in-depth data analysis that could be possible as smart devices begin to share their data profile to the energy network. Prototype 3 is the tool which allows a DSO to capitalise on the level of data now being supplied for analysis, pattern and fault detection. 

This prototype leveraged the data generated from the implementation of prototype 1 and 2 and could provide every employee with real time visibility of all network events, or access to authorised levels of information.

Phase 2: exploratory research with customers

This research found that customers currently expect access to multiple communication channels to contact DNOs. These include pull channels – such as the Internet, social media and radio – and direct contact methods – such as telephone, text/SMS, direct messaging and email. There are some notable differences in preference for communication channels from different customer types. This was linked to age, socio demographic and geographic factors; however, the availability of a range of methods was considered to be particularly significant for interactions during supply interruptions when it might not be possible to use mobile phones, landlines or computers.

Direct face to face communication was generally not expected; with the exception of a visible site presence during prolonged outages or to provide technical advice, concerning for example complex network connections.

When interacting with a DNO, customers expected to receive a consistent level of service on all channels, and for customer services records to be integrated across these channels. This ensures any information a customer has provided or received is available to contact centre agents, irrespective of which channels have previously been used.

In terms of content, its presentation and delivery, customers stated that they wanted concise, clear, relevant information about the issue or process in which they were interested, such as a supply interruption or a new connection. When exploring perceived acceptance of new innovative platforms, there was no appetite for information to be presented by augmented reality ie a hologram. Participants also did not want to be offered surplus material such as live video footage of activities related to the enquiry or process. If DNOs invest in technologies that can deliver enhanced, real time information that could be pushed to individuals through smart home hubs or other devices, the overarching view of customers is that they should be able to control the type and frequency of such notifications.

The ECP considered scheduling to be very important in situations where a DNO needed to visit them. Expectations were compared with those now routinely provided in other sectors. The minimum expectation was that, with the exception of emergencies, visits should be planned in advance and any changes or delay should be clearly communicated, in a timely manner via the customer’s preferred platform.

When data about a customer’s consumption is collected by smart meters or other devices, they expect to control whether it is shared with their DNO and understand the extent to which it should be aggregated, for example at local level, to enable the network operator to effectively manage the network as opposed to property level consumption data. DNOs should be explicit and transparent about the data they collect and how they intend to use it.

Customers were also clear that they wanted to control how such data is used (eg turn off appliances automatically, to optimise energy usage).

The detailed outputs from the customer research is contained in the Avatar ECP Final Report[1]. 

Virtual worker trial

Codebase8 were appointed to integrate a virtual worker to cleanse the data on the PSR to support the welfare process.

The virtual worker was tested to review their performance and assess the potential outputs and likely support resource needed to manage any exceptions raised when running in a live environment. This phase took longer than expected and the virtual workers were much slower to process records than envisaged. 

More testing, fixes and performance refinement were needed before the virtual workers could be moved to the live environment to ensure that the defined business rules worked as expected and no data was incorrectly deleted.

Following further rounds of UAT the virtual worker continued to be unsuccessful and we were unable to successfully pass the required performance and exception levels. It was then agreed to stop further development and not to move to live operation as it presented too much risk and required more intervention than expected. Therefore, this element of the project was abandoned.

Lessons Learnt

1.1       Phase 1: current trends and expert thinking
Consulting with specialist organisations was extremely valuable and helped to develop and shape the concepts and techniques which were explored with customers throughout the project.

This phase of research represented the first time that Electricity North West innovation projects had facilitated colleague engagement workshops, specifically to elicit insight from colleagues across the wider business. Engaging with colleagues proved to be extremely effective and the learning gained from their knowledge and experience highlighted common themes. This helped inform the development and refinement of ideas and shaped the conceptual solutions that were taken forward, based on real customer experience.

1.2       Phase 2: exploratory research with customers
Bringing forward the prototype development and introducing these to customers at the panel sessions instigated more meaningful discussions rather than just asking customers to come up with ideas.

1.3       Virtual worker trial
The procurement process for services to provide new technologies, designed to integrate with existing secure systems, which hold extensive personal and sensitive customer data can be protracted and introduce time constraints in delivery. It is important to start engagement on these types of contracts as early as possible.

Some automated worker solutions are more suited to our on-premise architecture than others and the levels of security required for sensitive data in a regulated arena proved to be a significant constraint. This was not obvious at the design stage of the project and only became apparent when we requested the same level of security as any other third-party access to our applications and started UAT. 

Whilst we were advised that the speed of the virtual worker would be slower, the magnitude was not clear and during UAT we observed that the virtual workers were not fast enough to be cost effective. We would, however, consider this type of product in future when our applications are more cloud based and we would select a product more suited to of the specific application whilst delivering the required security.

1.4       Challenges arising from different methodological approaches
We were aware that DXC favoured an iterative, agile approach and believed this would deliver the best outcome, allowing prototypes to be developed and modified in response to ongoing feedback and recommendations. However, the extent and flexibility of the internal resource required to accommodate this methodology was not apparent to us.

We had to manage a number of changes in the methodology to achieve the optimum output and accommodate the non-prescriptive agile approach. This was a challenging phase of the project which was exacerbated by the short-term fixed contract appointment of the DXC delivery team, which resulted in key members being replaced at critical stages of the development process. 

A series of meetings were held and a compromise agreed to ensure further slippage in prototype development was avoided. A number of measures were instigated to ensure the direction of the project was appropriately managed going forward. This was facilitated via weekly meetings where ongoing prototype development was communicated and interdependencies, requiring Electricity North West input, were identified and mapped. This allowed us to schedule our time accordingly.

This represented a divergence from DXC’s proposed methodology but ensured sufficient time was incorporated into the programme to provide an Electricity North West resource, at key stages of prototype development. This allowed conceptual solutions to be appropriately evaluated and ensured that the final prototypes were appropriate to present to customers.

This project highlighted the benefit of entering into a bilateral agreement with the organisation ultimately responsible for the delivery of goods or services, because of the challenges that can arise from a subcontracted arrangement.