three best practices for scaling up

Committing to the cloud means relying on this technology to be more than just a form of on-demand computing resource.

The Cloud is not limited to virtualized IT infrastructures: it is largely a lever to strengthen your capacity for change and adapt in time to the market requirements imposed by the company’s businesses.

We are far away, here, from changing infrastructure. This is why it is futile and even counter-productive to copy the patterns of traditional computing. If you are in charge, CIOs need to get involved, federate all departments of the company and progress according to five main stages:

1. The start -up stage, where the company formalizes its needs, evaluates its current situation, launches the benchmark of its suppliers and above all determines its ambition and the expected benefits

2. Next, experiment, with a definition of priority use cases. Most of our customers are there now.

4. Acceleration and generalization are the two stages that correspond to popular scaling up. This is the hardest moment, where you move from Testing and Learn to broader deployment-to enterprise-wide transformation, in effect.

5. Then comes stabilization, with specific pullbacks established based on success indicators.

Many companies (or CIOs) are currently responding to steps 3 and 4, So what are the recipes for success for successful scaling?

1. We need to rethink the IT Department

This is certainly not easy to achieve, but the DSI must agree to change itself, to allow all business departments to be “cloud ready”.

To expect a different result in the cloud, you need to completely change the game. It will play out in several levels:

Culture, first

Thanks to the Cloud, new levers will be able to be activated to speed up time to market. But to take advantage of this, DSI teams must change.

Traditionally, IT departments have attached the notion of responsibility to their respective perimeter leading them to limit and divide their perimeter of responsibility, they now have to register with co-responsibility. Don’t forget that it’s rare to see organizations clouding 100% of their applications, so we’re talking about a hybrid or dual IT cloud that further reinforces this need for collaboration.

Organization, then.

DevOps is no longer an option. Full DevOps is a must for cloud scaling. For some use cases, we’ll even go with some or NoOps models where an application is fully managed by a team, both in terms of development and support.

In this context of complete re -foundation, you may feel discouraged. Work on mastering your IS thanks to two levers:

Evangelize, culture, measure

Interpret your change in the cloud through extensive engagement with stakeholders, both on the expected benefits and on the impacts during the work.

Develop a training and evangelization plan that is worthy of the name, identifying beneficiary priorities and in favor of validating training that contributes to hard work and recognition.

Set up a strategy for identifying common goals and key outcomes to be achieved to collectively and clearly measure the level of change accomplished: draw inspiration from methods such as the OKR approach (goals and key outcomes)

These courses are inherently focused on architecture and knowledge of cloud services, but they are not exclusive. The cloud encourages you to explore new territories, such as FinOps or Green IT (affecting financial profiles or metrology engineers) and strengthen powerful territories such as security (Security by Design) and management without forget the new areas that are rapidly evolving: AI, no code / low code platforms, data in the cloud, edge computing …

Maintaining skills

Strengthening and stabilizing the skills to keep your job is a key factor of success

Not long ago, up to 70% of our clients ’CIOs were external. We noticed that they are gradually re-investing in people with re-skills in cloud expertise (architects, FinOps, CloudOps) and they are also recruiting.

One last point about people: DSI partners should also thrive and question their alignment with your own changes. This is particularly noticeable in sectors such as aeronautics, banking and transportation where requirements are evolving upwards.

2. Assemble a strong team. Codename: CCoE

Your cloud center of excellence (CCoE) doesn’t exist yet … but it will be the key success factor for scaling up.

Set up this cross-functional, autonomous entity as soon as possible (thankfully, even before the experimentation stage), which will integrate innovation and be a point of contact for all departments of the company .

The CCoE should remain a human-sized group (between 3 and 8 people) and may only have 1 or 2 referents initially, especially since you will not be transferring 100% of your IS at once, but start with 5 -10% of applications.

However, keep a few important things in mind:

  • Ensure management support with sponsorship at the executive committee level of the company,
  • Federate the DSI and the trades with the pilot trade of the first migrations and acculturation workshops by highlighting use cases. The cloud is the gateway that brings them together.
  • Make sure you have cloud poly-skills in this CCoE with architecture, DevOps, security, FinOps, data and change management representatives for training and acculturation.

Please note that this team is not intended to replace operational teams: CCoE is there to set the framework, not to do it, even if it is made up of cloud specialists.

Its function is also to promote transformation, rationalization and standardization, so as not to re-create what we see in our traditional IS.

In particular, it will ensure that ambitions are met, by promoting the consumption of new cloud services without compromising security. It therefore ensures a balance between innovation, performance, safety, costs and the eco-responsible commitment of the company.

Large companies such as Crédit Agricole, Engie or SNCF have already initiated these changes in small teams. In the context of large groups with multiple CIOs, these teams can sometimes be split at the level of each CIO with a CCoE referent at the level of a Group CIO.

3. Hold the breath to stay on course

Remember that change has a beginning and an end. You need to take care of your mountain and build a roadmap, with breathing stages just like in the dive.

Also don’t forget that Comex has 2-3 years visibility on strategic plans …

With this in mind, our advice is to set a first basic two-year goal for your move to the cloud, with intermediate steps that will allow you to step back and adapt the procedure as needed (see OKR approach on top). You will prevent your teams from running out of steam and you will execute your project with controlled turnover. Also create KPIs that everyone can understand (number of cloud applications, evolution of TTM, evolution of costs vs. usage, carbon impact).

One last tip: Be transparent with sponsors about the quality of your change. Startups are often less than expected and it is necessary to implement operational efficiency strategies, based on efficiency measures to find the right rhythm. Agile approach and Lean management provide an important source of inspiration for continuous improvement. Learn how to move forward without blinkers. From these first steps, we notice that having a consulting partner who is capable of opening your horizon to the maximum by providing relevant feedback and shaking you up some habits too anchored in the culture of your CIO and in the wider part of your company, proves to be a valuable asset.

Leave a Comment