
Who should own the description
What describing a process actually gives you
What a description consists of
Which processes to describe first, and which to leave alone
How processes are described internationally
Which methodology fits which situation
Using flowcharts properly
A simple BPMN diagram, and analysing what it shows
The description models available
Templates
The software worth using
A worked example using online tools
The actual output of the exercise
Implementing the described process rather than filing it
Evaluating how it performs afterwards
Adjusting it once reality disagrees
Describing an onboarding process end to end
Describing a hiring process
Describing a training and a performance process
Describing an exit process, and what each one exposes
Forming the project team and choosing the processes
Preparing the concept and defending it to leadership
Launching the project
Reporting, closing and monitoring afterwards
What a process is, and the characteristics it has
The different types of process
What improvement means as opposed to change
What improvement is worth in practice
The core values behind improvement work
What triggers it, and what waste actually looks like
Standardisation, and why it precedes automation
Process maturity, the five phases, and where to start
What mapping gives you, and how it differs from modelling
Stakeholder, relationship and flow maps
Cross-functional and value stream maps
What a map consists of, and the stages of making one
The improvement framework
How improvement connects to strategy rather than to complaints
The phases from planning to implementation
What process architecture is and what it buys you
Process-oriented architecture
What modelling is, and how to manage it
Where to start when nothing is documented
What problem solving means here
5S, five whys and 8D
Activity network and affinity diagrams, benchmarking
Checklists, cause-and-effect and flow diagrams
Linear regression and the nominal group technique
The prioritisation matrix
SIPOC and SWOT applied to a process
DRIVE, ICOR and the problem-solving funnel
What a culture of improvement consists of
The phases of changing one
How to change behaviour rather than announce a principle
The role of managers, and the role of the support function
The classification of systems
Where to find providers, and an overview of the market
How these systems evolved, and what that implies
What to automate first, and writing selection criteria
The types of applicant tracking system
Automating pipeline analytics and sourcing
Selection criteria for one
Negotiating with providers, and getting a real discount
Whether to implement the practice first or buy the system first
An overview of goal-tracking solutions
Continuous feedback functions
Moving from forms to a proper system
The basics of online learning delivery
The types of system available
An overview of the market
Starting online training, and integrating the systems with each other
What collaboration tooling is actually for
An overview of the systems
Organising teamwork inside them
Getting people to actually use them
How chatbots evolved into something useful
An overview of solutions
What you can build yourself
Low-budget platforms, and where AI fits in
What the current capabilities actually are
An overview of existing solutions
Applying it across each stage of a process
Systems with AI built in rather than bolted on
The fundamentals, without the vendor framing
The current state of the technology and where it is heading
Which robots already exist and which are close
The three laws of robotics, and whether they mean anything operationally
How these models work, in the detail that changes how you use them
Practical use across an operational process
Integrating them into existing systems
Building your own custom assistants
Automating repeated queries and internal question load
First-line support and initial diagnosis
Generating personalised material at scale
Using the tools for internal communication
Using AI for data analysis and reporting
Analysing survey results at speed
Forecasting demand and need
Building strategy on the output rather than on the summary
Improving the text people actually read
Automating screening and pre-screening
Assessing fit with AI, and what that assessment is worth
Improving quality while reducing bias rather than encoding it
Personalised programmes at scale
Monitoring and analysing effectiveness
Integrating AI into learning platforms
Building courses on AI platforms
This course contains the use of artificial intelligence.
Automate a broken process and you get the same mess, faster and more expensively, with a contract attached to it.
The order everyone gets wrong
Somebody decides a function is inefficient, a vendor is chosen, a system is implemented, and eighteen months later people are still doing the work in a spreadsheet next to it. I have bought that system. The reason it failed had nothing to do with the software. Nobody had described the process, so nobody knew where it broke, so what got automated was the version of the process that exists in the head of whoever ran the demo.
What this course actually covers
Five blocks, and the sequence is the point. First, describing a process in BPMN until the break point is visible to everyone in the room, with worked examples end to end. Second, improvement: waste, standardisation, process maturity, the five types of process map, architecture, and the tool set from five whys and 8D through SIPOC and the prioritisation matrix. Third, choosing systems: the classification, selection criteria, negotiating with vendors, and integrating tools so data stops being retyped. Fourth, AI applied to real work — analysis, screening, learning, performance, plus implementation, security and the ethical limits. Fifth, eight lessons on agents: what separates an agent from a chatbot, what they do in each area, and how to build your own with a goal, memory, actions and integrations.
A note on the examples
Twenty-five of the thirty-nine lessons use people-operations examples: applicant systems, learning platforms, onboarding agents, retention forecasting. That is deliberate — this is the back-office function with the densest set of ready-made tools, so it shows the full chain from description to agent better than anything else. The sequence transfers to any process. The specific vendors do not, and given how fast this field moves, treat every tool name as an example rather than a recommendation.
Who is teaching this
I am Mike Pritula. I built the people system at Preply as it became a unicorn, including choosing and implementing most of its stack, and I have worked at Wargaming, iDeals and Alfa-Bank. More than 1.6 million students have enrolled in my courses across 185 countries, and over 150,000 specialists have gone through my programmes. I hold PHRi and SHRM-CP certifications and represent HRCI in more than ten countries.
What is included
Lifetime access to all 39 lessons
Active instructor support in the Q&A section
A Udemy Certificate of Completion
Working material: BPMN mapping, the five map types, 5S, five whys, 8D, SIPOC, selection criteria, agent architecture
A final project: your own working agent for one real task
Where to start
Take the process you most want to automate and try to draw it in ten boxes. If you cannot, you have found the reason the software would not have helped. Enrol now and start today.