Loading…
ES/EN
In a company of 20 to 50 people, every hour of repetitive work carries a disproportionate cost because there are few people to absorb it. Deciding what to automate first does not require a six-month consulting engagement — it requires a clear criterion: frequency times time times error cost. That calculation almost always points to the same four or five processes. Starting with one and doing it well changes how the entire team views what comes next.
In a 200-person company, a manual process that consumes four hours a week represents roughly 0.5% of total capacity. Someone handles it, does it reasonably well, and the business impact is marginal.
In a 30-person company, those same four hours a week represent nearly 1.5% of total capacity. But the real impact is higher still, because in small teams the people doing repetitive work are the same people who could be doing work that generates value. There is no separate pool of administrative resources distinct from strategic ones — the same people handle both.
This means automating inside a company of 20 to 50 people has a different return equation than automating in a large enterprise. The opportunity cost of manual work is proportionally higher, and the impact of freeing capacity is more visible and more immediate.
The challenge is that with a small team, there is also less slack for getting the first automation wrong. A half-finished implementation that gets abandoned has a real cost in time and in team trust.
To identify the first automation candidate, every repetitive process in the company can be evaluated across three variables:
Frequency: How often does this happen? Daily, weekly, monthly? A daily process with a small per-instance impact can be more valuable to automate than a monthly one with a bigger per-instance impact.
Time: How long does it take each time? Include the full time — not just the person executing the process, but whoever approves, reviews, sends the confirmation, and fields the questions when something is unclear.
Error cost: What happens when this process fails or is delayed? This includes direct costs (an invoice not sent, an order not processed) and indirect ones (a frustrated client, a demoralized employee, a regulatory penalty for late filing).
The process with the highest combined score across those three dimensions is almost always the right first candidate.
It is tempting to start with the most visible process or the one with the most strategic weight. But those processes tend to be the most complex, have the most exceptions, and generate the most internal resistance.
The first automation project has a goal beyond solving the process itself: it needs to demonstrate that this works, build team confidence, and establish the technical foundation for what follows. A simple process that gets automated well and delivers visible results in weeks is worth more than an ambitious project that takes months and generates skepticism.
A salesperson sends a proposal and has to manually remember to follow up at 2, 5, and 10 days. When there are more than three active proposals at once, follow-ups start getting dropped. The automation here is straightforward: a trigger based on the send date that creates a reminder or sends the follow-up message directly.
Requests that go to the manager on WhatsApp, who sometimes responds and sometimes does not. No record, no traceability, no clear rules about which amounts require approval from whom. A simple digital flow — a request form, amount-based routing rules, notification to the approver, confirmation to the requester — transforms this process without requiring complex software.
Every time a new client comes in, someone has to send the welcome documents, create the system access, schedule the kickoff call, and register the data in the CRM or the relevant spreadsheet. Each of those steps can be automated. The result is a more consistent client experience and a team that is not relying on memory to complete each step.
Weekly or monthly reports that someone prepares by copying numbers from one system to another. These are particularly strong automation candidates because preparation time is high, the process is perfectly repeatable, and the automation is directly verifiable — the report either arrives or it does not.
The payment reminder that someone has to send manually before each invoice comes due. In companies with more than 20 active clients on 30-day payment cycles, the volume of reminders easily exceeds what one person can manage without missing any. An automatic flow that triggers X days before the due date is one of the automations with the clearest return on investment.
A 28-person consulting firm identified that its client onboarding process took between six and eight hours per new client, distributed across three different people. Some of that time was necessary — scoping conversations, technical configuration. But a significant portion was pure administrative work: sending emails, creating folders, logging data in different places.
The frequency × time × error cost analysis put this process at the top of the list: it happened several times per month, consumed many hours, and omissions created a negative first impression at the most critical moment in the client relationship.
The automation covered the mechanical steps: automatic project folder creation, sending welcome documents, creating the support ticket in the help system, scheduling the kickoff call reminder. The parts that required human judgment stayed human.
The result was that administrative time per new client dropped from six hours to under one. The freed capacity was redirected to work clients could actually see and value.
Not every process is ready for automation. Before moving forward, it is worth checking:
If any of those conditions are not in place, the right investment is resolving them first — not jumping into automation.
Is your team evaluating what to automate first but not sure where to start? Schedule a diagnostic session and we will apply this framework to your actual processes. You will leave with a concrete first candidate ready to implement. Let's talk.
MORE IN THIS CATEGORY
What an Internal Ticketing System Is and When You Need One
What an internal ticketing system is, what problems it solves, and when it makes sense to implement one at a mid-size company. For operations leaders in LATAM.
How to Build a Weekly Operations Report That Generates Itself
How to eliminate the manual operations report with a system that collects, structures, and distributes data without human intervention. For companies in LATAM.
How to Integrate WhatsApp Business with Your Operations System
How to connect WhatsApp Business API to your CRM, ticketing, and order systems. What the API enables, what it costs, and where to start