By Liam O’Neill, Director bpmd
This is the first of two short pieces on AI from the perspective of the process professional.
This one is about the business impact of AI through the lens of process. The second piece will look at how process professionals can use AI in their own work, from maintaining process data to supporting transformation.
There are two broad types of process team
The first is focused on control. That means understanding the current way of working, documenting it, keeping a central and compliant view of it, and supporting audit, onboarding and governance. In simple and stable business environments, that can be useful.
Where these teams often fall short is mindset. The focus becomes keeping the status quo in good order rather than helping the business improve. Too much energy goes into improving the quality of the process data, and not enough into improving the quality of the process itself.
The second type of team is focused more on transformation. These teams still care about understanding the current state and making sure process knowledge is captured and reused, but their main job is to support change. They help translate business requirements into something delivery and IT teams can work from, ideally through process, so there is a clearer view of what needs to change and why.
AI affects these two types of team in different ways
If your team mainly exists to document and control the current way of working, AI is going to make life harder. Process change is coming off the back of new technology, and in a lot of smaller or more agile businesses it is already happening at pace. More work is being automated through AI.
Automation itself is old news. RPA has been around for years. Workflow tools have long taken some of the burden out of repetitive work. What has changed is that there is now a brain behind the mechanical hands. That increases the pace of change and expands the type of work that can be disrupted.
Some roles are already seeing heavy impact. Marketing teams are using AI in outbound, research and analysis. Research-heavy jobs are being reshaped. A lot of knowledge work is becoming faster to produce and easier to scale.
If your role is to control the way work happens, one of two things will follow. Either your organisation moves slowly on AI and you can keep up, or it embraces it and the pace of change outruns the way your team operates.
At the same time, AI governance is unlikely to sit with the process team. In most organisations, that responsibility will sit with data, enterprise architecture, digital, security or technical IT. It is a data, risk and technology issue first. Very few BPM teams are going to own that mandate.
There is still value in the assets that process teams already hold
A good process repository gives you a rich and structured view of how work happens across the organisation. That can become a useful training and knowledge asset but some people struggle to read process models comfortably. AI gives you another way to access that information.
You can take that structured process content, feed it into a proprietary knowledge base, and let people query it in plain language. They ask what the approval path is for an invoice, what happens when there is a product defect, or what the route is for handling an exception. The AI reads from the process repository and gives them a usable answer, pushing them to the process model if they want more detail/context.
That is useful when people are asking for process information. The question for process teams is how far that value holds when fewer humans are querying and more activity is happening through bots and automated agents. That is where teams that only act as process data stewards start to come under pressure.
The process teams that stay relevant will move further into transformation enablement
That is where process has real value in an AI-driven business.
Most organisations will be poor at spotting the right AI opportunities. Some use cases will be obvious. Others will be less obvious. Priority will often go to the loudest voice in the room rather than the biggest source of end-to-end pain.
This is where process comes in.
Through process mining, process modelling and analysis, you can create a clearer view of how work happens today and where the real issues sit. That gives you a stronger basis for identifying where AI may help and where it may just add more noise.
It also helps you challenge weak automation ideas
Take invoice matching. A business might want to use AI and intelligent document processing to handle exceptions and improve matching rates after failures in the ERP. That might help. The wider process still matters. If the root cause is poor upstream PO data, inconsistent purchasing behaviour or bad supplier inputs, then the exception handling problem is only a symptom. In that case, automating the back end may just hide a broken process.
That is where the process professional matters. Your role is to bring the wider process view and test whether the thing being automated is a sensible target for automation. You are there to make sure the business is solving a real process problem rather than applying AI because the tool is available.
That same role carries through into requirements and design
Process teams often hold valuable information about handoffs, decision logic, exceptions, controls and workarounds. That context helps technical teams understand what the business actually needs and how work really happens.
There is also a role during rollout and change
AI creates uncertainty for people. They worry about disruption, changes to roles and loss of control. Process is a good way to make that concrete and easier to understand.
Saying that AI will automate campaign management sounds threatening. Showing the future process is more useful. A bot supports campaign research. A human decides the campaign. A human checks the content before it goes live. Once you show the interaction between the AI and the human in a clear process view, it becomes easier for the business to understand what is changing and where the human role still sits.
Process also has a role after go-live
The technical teams will monitor how well the AI is performing in its own silo. They will look at things like request volumes, token usage and success rates. Process professionals need to keep looking at end-to-end process performance.
The real question is whether the AI is improving business outcomes.
Is it increasing lead volumes? Is it improving fulfilment? Is it reducing errors? Is it improving throughput or reducing cycle time? Is it helping the business hit the result it actually cares about?
That is the lens process teams need to hold. The job is to show whether the end-to-end process is performing better because of the AI, not just whether the bot is doing what it was designed to do.
The role of the process professional is shifting
A team that only takes a control view will struggle to keep pace with the rate of change. The governance of AI will sit elsewhere, and the current state will move too quickly to simply document and preserve.
A team that supports transformation has a stronger role to play. It can help identify the right opportunities, challenge the wrong ones, provide context for requirements, make the future state easier to understand, and track whether AI is improving end-to-end business performance.
Process still matters for the same reason it always has. Businesses still need to get work done. AI changes how some of that work gets done. Understanding the way work flows, where decisions sit, where handoffs happen and what outcomes the process is meant to deliver still matters.
That is the job of the process professional.
Your role is to help the business get work done properly and make sure change improves the process rather than just adding more technology.