Most businesses already have enough software. There is a CRM, an inbox, a spreadsheet, a project tool, a calendar, an accounting system and usually a handful of smaller tools that somebody added because one specific problem needed fixing. None of those tools is necessarily bad. The problem is what happens between them.
A lead arrives in one place. A decision is made somewhere else. The work is handed over in a chat. A deadline lives in a calendar. The reason a deal was lost may never be written down at all. People compensate for the gaps with memory, meetings and more messages. It works until the company gets busy, somebody is absent or the volume grows beyond what one person can keep in their head.
I keep coming back to a simple question: where does the business itself live? Not the invoices. Not the contacts. Not the tasks. I mean the actual operating logic of the company. Where can you see what came in, what matters now, what is blocked, who has capacity, what happened next and whether the whole process produced money or just activity?
In many companies there is no single answer. The business lives partly in software and partly in people’s heads. That makes the company harder to run than it needs to be. It also makes automation much less useful, because automating a broken or invisible process usually gives you a faster broken process.
That is the idea behind a business operating system. I do not mean a giant piece of software that tries to replace every specialist tool. I mean a layer that connects the important parts of the operation and keeps the company moving from one state to the next. A new opportunity should become a qualified lead. A lead can become a customer. A customer creates work. Work needs capacity, people, deadlines and follow-up. The result needs to be measured. The next decision should be easier because the system remembers what happened.
Mandorka has moved in that direction. It started much closer to the workforce problem. That was useful because labour is one of the hardest constraints in a real business: you can sell work and still fail if you cannot deliver it. But hiring is only one part of a much larger chain. A company first needs demand. It needs to decide which work is worth taking. It needs to know whether it has the capacity to deliver. Then it needs to coordinate execution, follow up with the customer, measure the outcome and learn from it.
Once you look at the whole chain, the pieces stop looking like separate products. Leads, customers, workers, partners, tasks, availability, follow-ups and outcomes are different views of the same operating reality. The value comes from connecting them. If demand goes up, capacity should become visible. If capacity is missing, the system should know what kind of person or partner is needed. If a job is completed, the result should feed back into future pricing, sales and planning.
This is also where AI becomes more interesting to me. I am not interested in adding a chat box to every screen. AI becomes useful when it has context, boundaries and a real job to do. It can help classify incoming demand, prepare a next action, spot missing information, summarize what changed or suggest where a process is leaking value. But the company still needs a clear operating model underneath it.
The same principle applies to small companies and larger ones. The scale changes, but the questions are surprisingly similar. What do we want? What do we have? What can we execute? What happened? What should we do differently next time? A good operating system should make those questions easier to answer without turning the company into a bureaucracy.
I like businesses that become clearer as they grow. That does not happen by accident. It happens when the important flows are visible, decisions leave a trace and the system is designed around outcomes rather than activity. That is the direction I am building toward with Mandorka.