Automation Readiness Framework

Assessing the Dev-Ops readiness of an organisation is not really new. The entire concept of building automations exists in various industry verticals, and builds one of the foundations for Industry 4.0. With the advent of Agentic AI Systems, suddenly automations and automated (agentic) workflows are in the spotlight again. 

In my professional life I have helped to roll out different types of automations from OPC UA based connection of field devices to the MES/ EPR layer to engineering design automations. Throughout these years I started summarizing some of the lessons into a concise step wise guide to assess "Dev-Ops Readiness" or better put the readiness of a corporate to implement automations. Find the rough steps below:

Step 1: Assessment

There are many ways to assess the readiness of your organisation. Here you'll see two tools that help you quickly iron out where possible areas of interest are. While doing the assessment make sure that there is clear strategic buy in, since these types of projects tend to be cross-cutting.

See in the attached Table below three important tools:

1. (Modified) CAMS Evaluation to rank specific domains and dimensions in your organisation on a scale from 1 to 5. Everytime you find there is a low score (for instance 1) you have clearly defined actions

2. Automation Stack Evaluation to understand which layers already exist in your organisation. This stack can serve as a foundation to roll out automation at scale

3. Value Stream Mapping, a tool from the LEAN toolset, can help you to understand flows of various kind through your organisation. Flow of information, products or others.

Step 2: Plan

After identifying your gaps through the assessment it's time to turn those findings into actionable work packages. You can start by clustering your findings into themes. In my experience, you'll typically find patterns around some core areas like: 

1. technical debt (those legacy systems that everyone knows need addressing - can be hard can be easy to address), 

2. process fragmentation (where different departments have evolved their own ways of doing things - very hard to change) 

3. capability gaps (where people simply don't have the skills yet - education, education, education - C in CAMS stands for Culture after all).

The planning phase is about Alignment. Each gap you identified touches different parts of the organization, and each stakeholder group has their own priorities and constraints. 

Step 3: Build the foundation

This might be the most important and at the same time the hardest part. Building foundations is unsexy work - there's no immediate gratification, no flashy demos to show executives. But skip this step, and your automation efforts will be built on sand.

Foundation building in automation is like preparing soil before planting. You're creating the conditions where automation can take root and flourish. This means establishing the basic hygiene around data governance, API standards, security frameworks, and - critically - the cultural practices that will sustain automation efforts long-term.

In practical terms, start with your lowest-hanging fruit from the assessment. If data quality scored a 1, begin there. If your API ecosystem is non-existent, that's your starting point. The goal isn't perfection - it's establishing minimum viable standards that can support your first automation pilots.

Step 4: Pilot (PoC)

Now comes the fun part - your first real automation implementation. Choose your pilot carefully. You want something with clear business value, manageable technical complexity, and stakeholders who are genuinely excited about the outcome. You may find the business value through your value stream map, where when done right you should uncover waste (in shape of resources, materials, time, etc.).

The pilot phase is your opportunity to stress-test everything you built in the foundation phase. Your data governance policies, API standards, security frameworks - they all get their first real workout here. Expect gaps to emerge, and budget time to address them.

Choose a use case where success is visible and measurable. A 30% reduction in processing time that everyone can see is worth more than a 50% improvement buried in backend systems.

Step 5: Scale

Scaling is where many automation initiatives stumble. What worked for a single use case suddenly becomes complex when applied across multiple departments, processes, or systems. Don't customize everything. Every department will have legitimate reasons why their process is different, why they need special handling.

1. Look for common automation patterns that can be templatized and reused.

2. Establish clear criteria for automation prioritization. Not every process should be automated, and not every automation request should be accommodated immediately. 

Most critically, scale your organizational capabilities alongside your technical capabilities. This means training programs, communities of practice, and clear career paths for people working on automation. Technology scales exponentially, but organizations scale linearly - plan accordingly.

Step 6: Optimize

Optimization is the phase where automation moves from project to practice. You're no longer implementing automation - you're optimizing an automated organization.

This phase focuses on three key areas: 

1. performance optimization (making your automations faster, more reliable, more efficient), 

2. capability optimization (expanding what your automation platform can do), and 

3. organizational optimization (making automation a natural part of how your organization operates).