The article that precedes this one, on the same blog, attempts to define what we mean by a business operating system. It is the definition of a category that has not yet had a settled name, and which separates the concept from that of management software, CRM, process automation, marketing agency. What the definition leaves open, however, is a question that comes back in almost every diagnostic conversation: where you enter from. That is, in practice, what the first thing is to do when a company decides to build such a system.
The answer is not univocal, and this article tries to explain why. No company tackles the operating system from scratch as an abstract project. Every company arrives there pushed by a very concrete symptom that the owner has under their eyes every day. The symptoms are different, and they group into four families. They are the four doors of entry to the system. The family from which you enter does not determine where you will arrive, but it does determine the order in which you build the parts.
Four doors, one system only
The first door is acquisition. The company isn't growing as it should. Campaigns on the usual channels yield less, contacts arriving are less qualified, the cost per acquired customer rises. The owner arrives with the request that seems natural, "we need more marketing", but soon discovers that the problem comes before marketing: weak positioning, unclear offer, sequence that doesn't close. The acquisition door is crossed by starting from the diagnosis of the entire funnel, not from the first new advert.
The second door is sales. The company receives enough contacts, but handles them in a dispersive way. Salespeople keep the history in their heads, quotes are written by hand each time, pipelines stop when one person is on holiday. The owner arrives with the request "we need a better CRM", and discovers that the problem isn't the CRM, it is the presiding over the commercial conversation. The sales door is crossed by starting from the mapping of channels, customer touchpoints, documents produced every day.
The third door is the back-office. The company's senior staff spend their days on repetitive work. Data extractions, document compilation, replies to questions that repeat, deadline management. The decisions that require their judgement wait. The owner arrives with the request "we need to hire", and discovers that the problem isn't of heads, it is of sorting work between what requires human competence and what is enough to oversee with a system. The back-office door is crossed by starting from the inventory of where the seniors' time is consumed.
The fourth door is content. The company has stopped communicating coherently. The blog starts and stops, social media is half-abandoned, product news comes out when it happens. The owner arrives with the request "we need a social agency", and discovers that the problem isn't who publishes, it is who keeps the company's voice alive. The content door is crossed by starting from extracting what the company knows, what it says only verbally, and what has never been put into a reusable form.
The four doors are different from each other in symptom, typical sector, owner's vocabulary. But they all lead to the same point: structured knowledge of the company, a system that makes it operational, ongoing oversight of the parts that live every day. The difference is not in the destination. It is in the module activated first, because it is the first where the symptom is acute.
Why the order matters, not the list
In a traditional commercial conversation, the supplier arrives with a catalogue. You need X, Y, Z. The supplier proposes X, Y, Z, possibly as a package. Buy it all and we'll switch it on the same day.
This logic does not apply to the business operating system, and understanding why helps to understand the difference between a service supplier and a partner who builds a system. The four doors are not catalogue modules. They are entry points to a single underlying structure, and the order of construction matters more than which parts are activated.
A concrete example. A mid-sized private clinic enters from the sales door: it has a new patient funnel that it manages badly. The sales system, however, requires clear knowledge of clinical procedures, doctors' specialisations, treatment packages. Without that structured knowledge, every automated reply to the patient is generic. So before building the sales system, the knowledge layer that feeds it is built. That is, part of what, in other clinics, would be activated through the content door. The doors talk to each other.
Another example. A trustee firm enters through the back-office door: the senior is saturated with repetitive work. The back-office system, however, requires a customer communication pipeline that today does not exist in a structured way. It thus turns out that part of the work is to build the sales door, even if the owner had not asked for it. The doors talk to each other, again.
For this reason, in a diagnostic conversation, you do not start from a pre-packaged bundle. You start from the symptom, and you understand which door to begin with, in what order to build the pieces, where they influence each other. It is slower work than a catalogue. It is also the only way, in many cases, not to build a system that works only halfway.
What it means, in practice, to cross a door
In order not to leave the discussion abstract, we describe how each of the four doors looks from the inside.
Acquisition. We start with an audit of current channels, published content, visible SEO, social presence, active campaigns if there are any. It is almost always discovered that the funnel has gaps before marketing: weak positioning, unclear offer, content that speaks to everyone and therefore to no one. We intervene at the level before ads, then we rebuild the acquisition system that reflects how your target informs itself today. Average time for the first cycle: four to six months.
Sales. We start with the mapping of the customer's actual journey, from the channels they arrive on to the decisions that require human judgement. We discover the gaps: leads lost between channels, follow-ups from memory, inconsistent quotes. We build the system that oversees every step: bot for off-hours replies, CRM with a structured pipeline, automatic generation of offers, intelligent reminders on deadlines. Average time for the first cycle: three to five months.
Back-office. We start with the inventory of where the seniors' time is consumed. Data extractions, document compilation, recurring replies, deadlines. We separate what requires human judgement from what only needs supervision. The mechanical activities are moved onto a system that produces under the team's control, not in its place. Hours per week are freed up, not minutes per day. Average time for the first cycle: two to four months.
Content. We start with the extraction of what the company knows and says verbally. Positioning, market, customers, real cases. An editorial identity and a plan that produces continuously is built: articles, posts, videos, communications. If the faces of the executive figures are needed in the videos, digital clones with explicit consent are evaluated, to produce editorial series without saturating the person's time. Average time for the first cycle: three to six months.
Whichever the door from which you enter, the system underneath is one only. The structured knowledge above, the customer interface layers below, the automations that oversee every day in the middle. The door you cross first determines where you see the first result, not how much of the system you have built at the end. The system is not bought in modules. It is built, one door at a time, in the order that makes sense for you.
The question often put to you, when you say you are looking around to build an operating system, is "but what do you do, exactly?". It is a correct question to which it is hard to give a short answer, because the right thing to do changes from company to company. The short answer that works best is this: it depends on which of the four doors you feel you have to cross first. If the conversation continues from there, it is almost always a good conversation.
If you recognise yourselves in one of the four doors, we talk about it in a forty-five-minute conversation. The diagnosis is free.