In this four-part series, we offer practical advice as to how organisations can successfully implement Cloud by running through the detail of our eight-layer Cloud operating model.
In the second instalment of this series we discussed the third and fourth layers of our eight-layer Cloud operating model: ‘Products & Services' and 'Governance & Performance.'
In this instalment, we address the fifth and sixth layers: ‘Partners & Locations’ and 'Process.’ We will provide an overview of the challenges faced in these areas and explain why they are important factors to consider when designing and implementing your Cloud operating model.
Understanding your resource landscape
Achieving a competitive advantage involves harnessing the skills, capabilities, and offerings of partners, suppliers, and adjacent companies. Fundamentally, organisations need to establish their partner strategy, identifying whether the transformation will be delivered in-house or via partners. Our experience has shown that transformations led by in-house teams are rarely successful, lacking the skills, capabilities, and traction needed to get off the ground in a meaningful way.
Consider your delivery options
It is important to consider that while one supplier may not provide the necessary variety in services, too many can result in additional costs and management overheads. The delivery partner(s) should also work alongside the organisation in a collaborative manner, rather than as separate organisations.
Decide on a commercial model
Organisations should then consider the commercial model that they use with their suppliers. Risk reward models, where a financial incentive is offered to a partner based on a defined risk, encourages a shared commitment to delivering a transformation. However, an organisation’s commercial model should be a collaborative decision, with the delivery partner creating a roadmap that reflects the organisation’s budget and timeline.
Location, location, location
Operating models should be designed with an understanding of the key geographies impacted as this can affect key decisions, such as hosting locations, requirements for data sovereignty, and the service delivery location. Global organisations will need to use a Follow the Sun model to support their cloud implementation, although it is important to note that strategic delivery centres need not be the same as the hosting locations.
Hiring locations for new employees should also be considered as remote locations are likely to be a detractor for new talent. City central locations or organisations offering working from home benefits are more likely to attract the best talent.
Understanding the business and technology processes across the business
The elastic nature of Cloud allows for the scaling of compute resources to achieve efficiencies and cost savings. To achieve these benefits, organisations need to adopt new ways of working, such as streamlining processes and increasing automation. Technologies, such as machine learning combined with robotic process automation (RPA), are being employed to empower people and automate low value activities. Combined with re-designed, lean processes, an organisation can transform manual operating processes into ones that are dynamically responsive to the demands of the business.
Some examples of Cloud-focused process automation include:
Automated alerts and analysis
An organisation can automate analysis and alerts around usage and costs by defining thresholds and the automated actions for when thresholds are exceeded.
Change management processes
You can automate change implementation and testing to provide confidence to stakeholders on the repeatability of simple activities. For example, an organisation can automate server builds by implementing automatic checks that are predefined by an admin. Once they have passed these checks, they will be automatically deployed to production.
Another perspective is moving towards self-service capabilities. For example, if a user wants access to an application, this could bring up a portal where the user can request access with a single click of a button. The system then automatically seeks approval from the right person, which only requires a single click from that person. Once the process has completed, the original user gains access with minimal manual intervention.
System patching has been the bane of many Operations teams’ lives due to manual application shutdowns, backups, and patching processes. Leveraging cloud-based automation to do this for your ops team in a reliable, repeatable way eases the burden, increases security, and leaves more room for engineers to innovate and work on other issues.
All of these examples are underpinned by policies that enable automated processes to run. They are based on pre-approved thresholds that can be scaled up or down as and when required. Managing these scaling decisions automatically frees up costly time and resources, helping to achieve cloud cost efficiency.
Simply adapting legacy data centre processes is not a feasible solution due to the magnitude of change across technology, desired skillsets, and ways of working. By doing so, organisations could potentially face the following issues:
- Keeping legacy processes and governance will stifle the organisation’s ability to transform quickly, defeating the purpose of moving to cloud;
- Financial processes would not match up to the change due to the switch from CAPEX to OPEX investment.
When considering processes, it is important to examine the main areas with a view of making processes leaner and less expensive. This includes:
There is limited value in having a cloud platform which you can deploy to in minutes if processes take days or weeks to action.
Cloud is constantly evolving and so your maintenance processes must similarly adapt to embrace the ever changing cloud landscape.
The notion of budgeting up front for technology investment is somewhat resolved with cloud. To maintain control, you will need to establish new processes to capture your cloud spend. For example, establishing thresholds to highlight when you are close to depleting budgets.
Signals and KPIs
You need to establish KPIs to determine the success of the processes. You should also rely on signals from monitoring to ensure the processes are implemented. This is particularly important in a risk or heavily regulated environment.
Having policies that support the safe automation of processes increases control and decreases risk.
When looking at the processes, it is important to consider whether you are likely to have global standard processes or local and specific processes. You will need to identify and manage these processes whilst ensuring they are catalogued.
What to expect in part four
In the final section, we address layers seven and eight of DMW's operating model: ‘Technology and Data’ and ‘Financial.' In the meantime, you can download our latest whitepaper on Hybrid Cloud by clicking the button below.