Winning hearts & minds in Cloud
Helping your people thrive on the Cloud
Date: March 2019
Authors: Tanushree Gupta, Finn Toner, Dee Yang
Contributors : Salem Amazit, Brice Challamel, Daisuke Chikura, James Cowe, Charlie De Courcy, Tom Grey, Travis Hahler, Jack Kwok, Sarah Masotti, Kaitlin Reimann, Nick Taylor, Paul Williams, Ayaka Yamada.
Prepared for: Partners
Document type: White Paper
About this document
This whitepaper provides guidance to help enterprise customers understand the desired end state for people to best consume and operate GCP.
It introduces the GCP Change Management Methodology which provides step-by -step guidance to help enterprise customers navigate change.
This whitepaper is intended to act as a conversation starter and thought provoker for organizations who want to navigate the transition to cloud - either working themselves, with GCP partners, PSO partners, or with PSO customer teams directly.
Cloud PSO customers, prospective customers, PSO partner organizations and anyone who is interested in landing successful change in the cloud era.
The reader should have a high-level understanding of the concepts described in Google Cloud Adoption Framework.
Executive Summary 3
1. The challenge of moving to the cloud 4
2. Going further with the Google Cloud Adoption Framework 6
3. The evolving employee experience 8
4. One size fits no one 10
4.1 Where do you want to go? 10
4.2 How do you plan to get there? 14
5. Ten things we’ve learned along the way 18
Find out more or get in touch 19
When moving to the cloud, many organizations concentrate their focus on the change in technology and overlook an area as complex and impactful: cultural change. Having your people ready to embrace the change, supported by the right processes, and equipping them with the right skills is as important as getting the technology right.
To realize the full value of cloud technologies, many organizations are rethinking their IT organizational structure. There are a variety of potential talent implications too—from adopting agile ways of working to hiring for more cloud-centric skills, as well as the reskilling and redeployment of current IT skills.
As one of the organizations that pioneered hyperscale infrastructure, which led to the creation of the cloud, Google has spent years nurturing its culture and workforce to best operate in the cloud.
We leverage this experience to help organizations ready their workforce for the change and in this whitepaper we aim to:
- Provide a framework for thinking about your organization’s cloud readiness (chapters 1 & 2).
- Explore how different elements of the framework evolve and mature (chapter 3).
- Share typical customer scenarios surrounding the journey to GCP (chapter 4).
- Explore common migrations approaches which customers consider (chapter 4).
- Articulate a set of principles based on best practices and lessons we’ve learned along the way which influence our change management work with our customers (chapter 5).
These are principles rather than hard and fast rules. There is no one-size-fits-all answer about the target state, and every organization will have unique considerations which will guide key decisions throughout the journey. For some, the journey to the cloud can be simple; for many, it will have complications. We believe that considering how to get everyone moving in the same direction will help make the radical impact offered by the cloud more tangible and more sustainable.
1. The challenge of moving to the cloud
Cloud computing is a fast-moving beast. The development of processing and storage technologies means that computing resources have become cheaper, faster, and more available than ever before. We now live in a consumption-driven world; to be more competitive, deliver better outcomes faster, and with greater reliability and efficiency, organizations need to find a new way to consume IT. At the same time IT resources (for example, storage, computing, etc.) are provided as general utilities which can be leased and released through the Internet in an on-demand fashion. The momentum gathering behind cloud adoption will continue to increase, as more organizations seek to use it to differentiate themselves from competitors. According to International Data Corporation (IDC) , in 2018 almost half of IT spend was cloud-based, with estimates to reach 60% of all infrastructure and 60-70% of all software, services, and technology spending by 2020. But the journey to cloud can be a complex transformation. Many organizations are finding greater costs and greater obstacles to the adoption of cloud than they anticipated. There is increasing recognition that successful digital transformation isn’t only about technology, it’s about people.Organizations will only realize the full potential of new technology if people use it, know how to make the most of it, and are empowered to do things differently than before. This recognition can be articulated by a range of questions you might hear asked in exasperation:
- Leadership: How can we get leadership visibly and meaningfully behind the journey to the cloud?
- Collaboration: How do we break down the silos between the business, technology, and other supporting functions?
- Capabilities: What new skills and capabilities do we need to make the most of cloud?
- Talent: Where do we find these new skills, and how do we get them to join and stay with us?
- Engagement: How do we win the hearts and minds of those who are skeptical of the cloud?
And there are several great reasons why answering these questions successfully is critical for successful transformation:
- Higher performance: Organizations where leaders set the direction, but give their teams the autonomy to decide the route, have higher levels of trust, cooperation—and service delivery performance.
- Increased engagement: Increased levels of inter-team ‘bridging’ is an important link to engagement; employees who experience a collaborative culture like this are 1.8 times more likely to recommend their team as a great place to work.5
- Minimize resistance: People tend to be confronted with more information than they are able to process, and have a tendency to stick to the status quo; communication strategies which channel ‘dolphins not whales’ are more effective at breaking this down.
The Google Cloud Adoption Framework provides our perspective on what drives successful cloud adoption and we build on it here to articulate the principles for successful people-centric GCP change management.
2. Going further with the Google Cloud Adoption Framework
The Google Cloud Adoption Framework is our perspective on successful cloud adoption, based on our experience from working with customers. It provides structure to the rubric of people, process, and technology that you can work with, providing an assessment of where you are in your journey to the cloud, with actionable next steps to progress. The framework focuses on the intersections of these three rubrics as this is where there is often friction and complexity and hence the model leans in here. We’re going to ‘zoom in’ on the People part—which breaks into a list of five epics required to successfully transform people and culture— and look at how they evolve over time and with maturity through the transformation. (See Table 1 for a summary of the epics.)
To kick start the change, you need two things right from the start. One - top-down sponsorship from leaders who are committed to the journey, and two - generate a cross-functional and psychologically safe environment for those involved through great teamwork.
But to build early momentum for early adopters, you need to go further. Transparent and open communications which reinforce collaborative and blameless behaviors with a culture of experimentation will set the right tone within the organization, and win hearts and minds. And this is further empowered by the alignment of people operations to your goal - that is, ensuring that the right structure, roles, and skills are in place to make change a long-term success.
Table 1: Summary of the GCAF people-related epics
|Sponsorship||Leaders continuously and meaningfully demonstrate support for the cloud adoption strategy, giving early adopters a mandate for change|
|Teamwork||Teams live and breathe collaborative and trusting behaviors, so cloud technology can be used in the optimal manner|
|Communication||Transparent communication is used to reinforce and nurture a culture where failures are shared openly, and treated as opportunities for improvement|
|Behavior||There are clear expectations and support structures for people, who continually seek to improve how they work as a team and empathize with their audience / user|
|People Operations||The organizational ‘muscle’—like skill development, role clarity, and performance measures —is toned to help cloud adopters fulfill their new tasks and duties|
3. The evolving employee experience
As your journey to cloud progresses, the experience of your people—how it looks, sounds, and feels to work in your organization will evolve. The Google Cloud Adoption Framework defines the key characteristics of different levels of adoption as described below (Table 2); this provides the basis for an articulation of the changing employee experience (Table 3).
Table 2: The Google Cloud Adoption Framework maturity definition
|Maturity Stage||Organizational State|
|Tactical||Individual workloads are in place, but no coherent plan encompassing all of them with a strategy for building out to the future|
|Strategic||A broader vision governs individual workloads, which are designed and developed with an eye to future needs and scale|
|Transformational||With cloud operations functioning smoothly, you have turned your attention to integrating the data and insights gathered from cloud|
Different organizations often have different starting points, and so there are no two identical target states. However, there are some common expectations at each evolutionary step in maturity.
The People Maturity Framework (Table 3) gives guidance on ‘what good looks like’ as a goal, as maturity evolves. For each of the epics, you can see the desired state evolve, with the cloud adoption increasing across the organization. The People Maturity Framework can be used to understand where you are today, and highlight where there may be gaps from where you need to be tomorrow.
If there’s a specific area of interest, or a specific gap you would like to discuss on with us, we would love to hear from you.
Table 3: People maturity framework
|Sponsorship||Sponsorship limited to senior management from one line of business or a teamCloud projects are self-funded primarily via cost-savingsAdhoc governance||Sponsorship extends to the C-LevelDedicated cloud budgets to drive the cloud initiative at a strategic levelCentralized governance overlooking all the cloud-related initiatives||Comprehensive C-level sponsorship which consistently set the tone for the cloud-first policyError budget for experimentationCloud-first policy empowered by decentralized governance|
|Teamwork||Cloud adoption progress is driven by individual contributors with a personal interest in cloud and collaboration among teams is difficult||Cloud adoption is driven by a dedicated cross-functional team of advocates, often referred to as Cloud Center of Excellence (CoE), working across project boundaries and collaboration outside the CoE is difficult||The CoE (Cloud Center of Excellence) becomes a leading light for new ways of working - operating in a transparent, collaborative, and autonomous environment|
|People Operations||IT organization is largely organized in a siloed structure, with multiple handover pointsNew cloud skills introduced, e.g. cloud security, virtual machine provisioning, procurement, budgetingThere is no standard way of measuring success||IT organization pivots towards a more empowering, agile set-upNew cloud skills introduced, e.g. serverless, microservice, Containers and Kubernetes, automationClearly defined objectives and KPIs supporting the adoption of cloud||IT organization consists of small and autonomous teams without the traditional role separationsNew cloud skills introduced, spilling from IT into the business, e.g. data scientist, image recognition, TensorflowError budgets are accepted|
|Behavior||Collaboration with parts of the IT org is difficultKnowledge is ‘owned’ rather than ‘shared’ freely IT as the ‘cost center”||Psychological safety and trust are established; clear balance of empathy and accountability IT as the ‘Enabler’||Error budgets and blameless postmortems are well recognizedGreat tech is freely available for people who need itIT as the ‘Invisible Innovator’|
|Communication||There are frequent communications top-down within each team. Key information and data is shared within the team, but not cross-functional||Gradually more consistent and transparent communication is conducted across teams. Key information and data is shared with more people||Consistent and transparent communication is conducted across teams. Everyone understands the ‘why’ and has access to key information and data|
4. One size fits no one
While it can be comforting to look for best practice, and tempting to seek a universal ‘correct’ answer—it is important to remember that there can be no one-size-fits-all answer on how your organization should design its future people state and manage its transformation journey. Every organization will have its own unique considerations which will determine their own unique journeys to cloud. For example, budget constraints, legacy technology solutions, organization setup, regulatory requirements—to name a few.
The answer to two questions will influence the change impact on your people, and their future experience:
- Where do you want to go? (aka - what’s the cloud vision?)
- How do you plan to get there? (aka - what’s the migration approach?)
4.1 Where do you want to go?
The below graph demonstrates the three broad change outcomes organizations tend to consider and the relative importance of each of the levers to get there. As a transformation journey progresses—with its impact at scale and duration—the supporting structures of people operations gain relative importance. In contrast, as tactical changes are landing, there is greater importance on individual behaviors and teamwork.
Graph 1: The impact on people at each maturity stage
4.1.1 A typical tale of Company A: “I want to get moving to the cloud quickly”
There is no single ‘Company A’—but there are many organizations who are in a scenario like this. Their priority is to quickly take advantage of the cloud’s scalability. For example, this could be moving a physical data center to a virtual one. This is what’s sometimes known as a ‘lift and shift’—where the infrastructure is delivered as-a-service, with minimal change to how the business interacts with IT.
This change requires leadership clarity around the vision which for the moment, is reasonably conservative. Company A doesn’t want to pull up any trees just yet while moving to the cloud. So the communication will probably be cascaded from the top, and sponsorship must be focused on establishing trust in the cloud; this will put Company A in a position to accelerate further migration, and build momentum over time.
The impact on the current IT operating model is limited to the infrastructure layer and how it is maintained and consumed. There will be increased need for ‘infrastructure as a service’ capability, and the skills to support that rather than maintaining and managing a physical data center.
The level of change experienced by the rest of the IT organization is reasonably small (see Table 4) and the business probably won’t know when the change happens.
Table 4: Indicative capability impact for companies who want to move to the cloud quickly
|Infrastructure Architecture||↑||Increased need for skills to manage the virtual resources on the shared infrastructure (e.g. IaaS, or Infrastructure as Code)|
|Hardware Maintenance||↙||Reduced demand; you don’t own the hardware, and you don’t need to manage or work with the hardware vendors that fix it|
|Hardware Procurement||↙||Reduced demand; less or no more data center hardware to buy|
|Capacity Management||↙||Infrastructure resource planning is likely to be subsumed into the customer contracts and pricing administration role.|
|Security||↗||Potential increase in demand, to manage access and identity in a cloud environment (shared infrastructure)|
|Finance / Procurement||↗||New skills required to manage a different budgeting process (Opex focused) and chargeback models|
4.1.2 The ballad of Company B: “I want to be a cloud-first company”
For every Company A who wants to dip their toes into the cloud water, there is a ‘Company B’ who wants to use the cloud opportunity to transform the way they do things, becoming what we call a cloud-first organization. From 100-year-old retailers to automobile giants, becoming cloud-first changes the conversation from IT cost-reduction, to supporting and fuelling business growth.
Cloud-first organizations can quickly build and deliver services in response to customer needs; they automate more and more IT delivery, becoming lean, service-aligned organizations which focus relentlessly on improvements, resilience, and reliability.
This sort of change is known broadly as Site Reliability Engineering.
Having a vision like this implies a greater fundamental change in the way the IT organization operates. Delivery changes to become more agile, team setup changes to become more collaborative, and IT changes how it interacts with the business to become more customer-focused. And all of this has an impact on the way people get things done.
New skills will be required—like container management using Kubernetes (see Table 5) - which may mean upskilling current employees, or hiring additional people with these skill sets.
Table 5: Indicative capability impact for companies who want to be cloud-first
|Microservice Architecture||↑||Increased ability to segment applications into microservices, which significantly increases the overall agility and maintainability of applications|
|Site Reliability Engineering||↑||Increased demand on those responsible for the availability, latency, performance, efficiency, change management, monitoring, emergency response, and capacity planning of their service(s)|
|DevOps||↑||Greater demand on the ability to build and run applications in an agile fashion and set up structured communications among teams to solve a problem together rather than traditional build vs. run|
|Automation||↑||Enabling auto-scaling, infrastructure as a service, continuous integration and delivery, automated monitoring and reporting, to eliminate human errors and save time|
|Containers and Kubernetes Engine||↑||Pack each part (e.g. application, process, etc.) in its own container, facilitating transparency and resource isolation. Kubernetes Engine provides the foundation of a cloud-native application|
|Serverless Computing||↑||Build and run applications and services without being constrained by traditional server concerns|
|AI & Machine Learning||↑||Skills to leverage AI and machine learning technologies to optimize applications, provide insight in hours, and support innovation|
|Agile Project Management||↑||Agile delivery methodology would be in high demand until it becomes a core competency across the organization, to maximise agility and accelerate business velocity|
|Waterfall Project Management||↙||With the increased rate of cloud adoption, there would be less need for waterfall delivery approach|
|Incident Management||↙||Traditional platform and hardware support moves to the provider; in-house capability more focused on joint incident detection|
Cultivating DevOps-esque behavior is relatively easy in an organization of 20 or 30 people. But if you need to scale this behavior to 200, or 2000, then a more formal scaling mechanism is required. In contrast to the top-down communication in Company A, those leading the change for Company B must spend more time building the lightweight—but strong—structures which recognize and encourage the right behaviors.
Communication should feel much more ‘bottom-up’, and ‘middle-out’ than top-down.
Both the scenarios described here are inspired by real customers of Google Cloud. And both the desired end-states are feasible and valid —driven by respective sets of business priorities, implementation complexity, and a myriad of other factors. Having examined different end destinations, we’re now going to explore the second question—how do you plan to get there?
4.2 How do you plan to get there?
Our natural tendency is to characterize change as a linear process—a straightforward progression from A to B to C. However, linearity is one of the pervasive myths about change; in fact, there are many different paths to successful organizational change. Based on our migration experience, and successful collaborations with our partners, we’ve identified five common change paths (see Graph 2) which organizations tend to take.
Graph 2: Five common change paths
4.2.1 Option A - ‘Move’ first and then ‘Change’
The enterprise which selects this path is taking a reasonably conservative approach to adopting cloud to modernize applications and drive growth. Change is relatively slow but tends to gather momentum over time.
This path will typically start with a ‘lift and shift’ approach for selected applications, minimizing changes to ways of working (see Customer A - section 4.1.1). Once migrated, the applications will be re-architected and optimized in the cloud. At this tipping point, new ways of working become more feasible for more people; this will require increased change management to realize, sustain, and celebrate the benefits of moving to the cloud as they’re unlocked over time.
4.2.2 Option B - ‘Change’ before ‘Move’
An enterprise on this journey is taking a more aggressive approach to reach its end goal. This path will typically start with re-architecting applications first to make them more cloud-ready, before migrating into the cloud.
This application modernization requires new skills (see Table 5) and ways of working (for example, Agile, DevOps/SRE) if it is to be successful. For example, while cloud can enable increased responsiveness to customers needs, this cannot be achieved through waterfall delivery; rather, a DevOps-led approach supported by automation needs to be in place.
Cloud can also make experimentation easier, cheaper, and less risky. However, this won’t happen in a culture which does not reward the behaviors of innovation and learning, testing, and iterating in order to ‘fail well’.
Taking this sort of approach requires more change management up-front, with a more immediate effort to launch and manage the change. Central to this is developing a clear vision that helps all stakeholders understand the direction of travel, why this change is important, and what to do differently to be successful.
4.2.3 Option C - Just ‘Move’
Building cloud-native applications is trendy, but it doesn’t mean that it is applicable for all scenarios. For some use cases, it’s sufficient to leverage cloud just to modernize the infrastructure layer. For example:
- Disaster Recovery - with a robust and scalable infrastructure layer, organizations can consolidate their disparate disaster recovery systems into one virtualized environment for disaster recovery
- Data Center - decommission the on-premises data center and move towards a virtual environment
- Big Data - manage, store, and analyze big data like structured data (that is, databases) and unstructured data (that is, social media, images, web, email, Internet of Things (IoT) sensors)
There is typically less change to the existing application workflow and ways of working, hence the complexity of change is low. The change focus is on those who are operating and consuming the infrastructure layer (Table 4).
4.2.4 Option D: ‘Invent’ in Greenfield
With a Greenfield strategy, a whole new infrastructure and application is built in the cloud. This approach really only applies when the organization needs to develop new products or offerings (for example, new e-commerce platform, digital banking, etc.). This path requires agility, access to a diverse development skill set (Table 5) and generous organizational tailwinds to make this Greenfield vision a scalable reality. Typically, this is the approach we see when customers start up a Cloud Center of Excellence—building a small team from a diverse set of professional backgrounds, who can experiment and document the new ways of working, scaling and advocating best practices. This option can potentially bring an interesting tension between how much of the organization is ‘new’ and how it interacts with the ‘old’ part. This can bring technical pain, governance pain and even procurement pain (for example, paying on consumption vs. traditional standards). Tension like this can lead to the feeling of a ‘two-tier IT’ where some people get the agility of the cloud, while some don’t. That can be painful, but it also provides a very visible incentive for teams to migrate to the cloud, and more agile ways of working.
4.2.5 Option E: ‘Invent’ in Brownfield
A Brownfield strategy, on the other hand, is to invent a new application in the cloud environment, while there is an existing legacy application which will be retired. While this redundancy can be comforting—especially for mission-critical applications, this path can be dogged by the financial implication of dual running and is often the most expensive choice.
4.2.6 Lots of choices; is there an answer?
The pace of change, and the change management required, is driven by the end destination, and the choice about how you want to get there. These are fundamental choices, but we think there are some things to keep in mind which can help make these decisions.
5. Ten things we’ve learned along the way
The table stakes in your journey to the cloud are knowing where you want to go (aka your cloud vision), and having a sense of how you want to get there (aka your migration approach). The Google Cloud Adoption framework provides a structured way of thinking about these, and the People Maturity Framework articulates the key focus areas for helping you embrace the change. On your journey there will be many specific nuances and unique constraints, though there are some typical scenarios and migration approaches which can be employed to shape your change approach. Based on lessons we’ve learned along our own journey, and from our work with customers, there are also a set of considerations which we think are important to make your journey a successful one.
- Focus on the people, and all else will follow. It is worth investing time in painting a clear picture for people of the future, and getting their input. Nobody ever said ‘They focused too much on understanding how people feel.’
- Measure, measure, measure - and make sure you know if you’ve been successful. Be up front about the metrics you will use to measure yourself along the way. Set success goals and communicate them; make sure your feet know there is a fire somewhere nearby.
- Be clear about the capabilities you will need in the future - and where you’ll get them. Product or Service Driven Agile Development, Site Reliability Engineering, and Cloud Enabling Service capabilities are the basics for scalable, agile cloud enablement. Make sure you’re confident these are available—either internally, or through your partner(s).
- Fast is better than slow. But finding your balance between central control and agility will be hard. Agility is achieved through increasingly decentralizing capabilities, balanced with a central foundation. There isn’t a ‘universal truth’ for hitting this balance and there will be disagreements about it which you will need to navigate.
- Sweat the basics; people will look to you for guidance. More people will be interested than you think. Map out who they are, what they’ve got to gain and establish champions to promote and support the cloud transformation.
- Ensure there is a ‘non-tech’ learning plan available. Not everyone will feel comfortable with the future picture of collaboration, innovation, and agility. Give people the opportunity to develop and practice these news skills, before they need to use them for real.
- Start thinking about the needed tech skills now. Whether you're choosing to hire in skills, or build it up in-house, there will be some cloud, AI, and Machine Learning skills which will be in short supply. Know how you're going to fill those gaps.
- Appeal to self interest. Any classical economist will tell you that a shortage drives up prices. Salaries for cloud skills are on the rise, and job adverts for cloud skills are on the rise. People will want to ride this wave.
- Things won’t be perfect the first time - fact. The journey itself is an opportunity to role model the future. Give people the opportunity to reflect and share their suggestions about how to improve; retrospectives can be your best friend.
- Share what you learn along the way, both positive and negative. You will learn a lot. This is a pot of valuable insight which will benefit others. Share with generosity and scale the lessons you may have had to learn the hard way.
These ten statements aren’t a manifesto; they’re not rules; you might disagree with them—and we would love to hear about that! They are reflections about what we think is important when moving to the cloud based on working with our customers and partners, and from what happens in Google. We’ve distilled, tinkered with, and baked them into an approach; if you’re interested, please get in touch.
Find out more or get in touch
- We are teaming up with partners to pilot our change management methodology for GCP. If you’re interested in learning more, get in touch with your Google PEM.
- If you want to contribute to the conversation, keep an eye out on the Cloud Connect Community where we will be sharing more about sustainable change and how to build the skills to lead change in the cloud era.
- And if you would like to find out more about the overarching Cloud Adoption Framework, you can download the whitepaper here: https://cloud.google.com/adoption-framework/