You cannot centralize a process without centralizing information

In my article "Why did I fail to implement a new process or system?” discourse on the points of attention on which to act to mitigate the risks of an implementation, as well as the main activities to be carried out for the success of the project. After recent conversations with executives from large companies regarding the implementation of new systems and activities related to logistics (eg: control towers, implementation of systems such as TMS, fleet routing and scheduling), I bring in today's post another front of reflection: the centralization of processes.

I say emphatically thatyou cannot centralize a process without the centralization of information”. Centralize a process with the aim of increasing control and reducing head count it can be a big shot in the foot, and even make the operation even more inefficient, if the processes are not well defined and controlled systemically.

A commonly made mistake is the selection of new software to be implemented (eg order scheduling/routing) without a complete understanding and definition of the associated processes. And why could this be a mistake? Because companies sometimes end up defining their processes based on software purchased from the market, the so-called “off-the-shelf software”. They are not always suited to the company's reality and, consequently, processes can be built far from the best practices for the market in question, without contemplating the activities that would allow a more efficient control of the operation, or including functions that stifle the operation simply because " the system is like that”. The process must come first and then the system enters.

This is not to say that I am not in favor of purchasing an off-the-shelf system, on the contrary, they tend to be less costly and more suitable in the long run when compared to systems built in-house. Specialized companies are constantly investing in research to develop functionalities and apply constant improvements based on the lessons learned from their customer-users. In this way, they promote recurring upgrades of their systems, in a short response time, and at a cost that is much more competitive than the cost of development by internal IT teams.

On the other hand, building processes from off-the-shelf software applies very well in the case of smaller and simpler support operations (eg: reimbursement of expenses, registration systems, etc.), or in micro-enterprises, in which many areas still are being built and the implantation of a system can work as a kick-start to a series of important activities.

You may still be wondering what I mean by “information centralization”. The centralization of information is a great process in itself. Within this process, the first step is the design of all the processes that you want to implement in the light of best practices and adapted to the company's reality, in a simple way, but guaranteeing operational efficiency. The second step is the implementation of these already validated processes with all stakeholders and future managers of the operation. This implementation needs to have a pilot phase, where these processes are tested and adjusted, as questions arise and exception cases that were not contemplated in the initial design. After adjustment and roll out After this implementation, indicators must be monitored to measure the effectiveness and efficiency of the new process.

The next stage is the selection of which system will be adopted to make the process more productive and controlled, which must be started in parallel with the process implementation pilot, when you already have a clear vision of all the functionalities that will be necessary. Note that at this point the definition of the system becomes much easier and more accurate, as the process has already been built, tested and validated. The system then has its turn and all implementation stages (pilot and roll out) will be executed again, but more quickly, as the processes are already standardized. Some customization, even if small, may have to be done, either in the process to adhere to the contracted system or in the system to include features necessary for the reality of your company. In any case, these adjustments will be carried out in a planned manner, as they are already known at the time of software selection.

After all the standardization of processes and successful conclusion of the system implementation, we have what I call “centralization of information”. For the example of deploying programming and scripting software, at this point the system already has all the necessary information so that the programming activity can be done remotely (e.g. information about orders, such as customer location, agreed delivery time, products and volumes requested; and software parameterization with restrictions for these customers, profile of the fleet able to deliver, urban circulation restrictions, etc.) , as well as the entire team is trained to perform a standardized process based on the company's central guidelines. There is no longer any dependency on knowledge that was previously exclusively with the team of logistics analysts at each of the facilities. Imagine the routing being carried out in a centralized control tower, but with the need for the programmer to contact the logistics analyst of each of the facilities to ask which route to take or which customers to group on each trip. It would be very unproductive, right? This is an example of what can happen when information is not centralized.

With the centralization of information, the signal is green for the next phase: centralization of processes. Running a red light or trying to divert the path to the shortest route, due to pressure for cost reduction, comes with a high probability of crash, in this case, of not achieving the expected results or returning to the initial traffic light. I've heard reports of companies that are decentralizing operations that had already been centralized. The problem may not be in the centralization itself, but in the fact that it was carried out without following the steps mentioned above.: (1) process design and validation; (2) implementation of processes; (3) systems implementation; (4) control and centralization of information; (5) process centralization.

Ensure your processes are robust and standardized in all its facilities, and that relevant informations for the execution of the order cycle are not associated with an employee but with a system unified with remote access, so you can proceed to the centralization step.