A premium vehicle retailer in San José grew revenue by more than 35% and operational efficiency by 21% by documenting its processes, centralizing knowledge and measuring how its sales team performed.
The work began in January 2024, and it did not start with technology. It started by writing down how the company actually worked — the part that gets skipped because it is invisible and unbillable. This case covers what we found, the order we fixed it in, and why the result was not a market effect: that year, the Costa Rican vehicle market moved in the opposite direction.
What was holding growth back?
The company was selling, but every sale depended on whoever happened to be there that day. There was no searchable knowledge base, so staff spent time working out how each procedure ran instead of running it.
The initial diagnosis found three brakes, and none of them was demand or product:
- Business processes with no central home: nothing was documented in a single searchable place, so each person operated their own version of the procedure.
- Sales hiring and onboarding without structure: no consistent induction framework and no agreed performance measure, so a new rep's output depended on who trained them.
- Technology decisions without a return criterion: software was judged on features rather than on what it would give back.
From the outside all three looked the same: an operation that held today's volume and would break under tomorrow's.
When the procedure lives in one person's head, the operation doesn't scale: it just replicates the bottleneck.
What does undocumented knowledge actually cost?
It costs measurable productivity, not just friction. The academic evidence is consistent: how a firm documents, measures and manages its operation explains a large share of why it out-produces its competitors.
The study Management as a Technology?, by Nicholas Bloom, Raffaella Sadun and John Van Reenen for the National Bureau of Economic Research, measured management practices across more than 30 countries and found that structured practices — tracking what happens inside, setting targets, reviewing outcomes — account for between 20% and 50% of the total factor productivity gap between firms.
The same research programme, summarised by the London School of Economics, attributes between a quarter and a third of productivity gaps — across countries and within them — to management. Put the other way: a company that does not write down how it works competes at a disadvantage that never shows up on a financial statement.
That is the real cost of undocumented knowledge. It is not that people work badly: it is that everyone reinvents the same procedure, and that repetition is not billed on an invoice — it is billed on margin.
The gap also compounds. Every month a process stays unwritten, another exception gets added to somebody's memory rather than to a document, and the cost of writing it down later goes up. Companies rarely decide against documenting their operation; they postpone it until the person who knows it leaves.
What came first: documenting or buying software?
Documenting. Buying software before knowing how the work is done only automates the mess and makes it more expensive to undo.
Workflows were redesigned on top of the written processes, and only then was the infrastructure built to sustain them:
- A local intranet centralizing procedures, with an internal file and knowledge management system.
- Direct management of sales hiring and training.
- Advisory on software purchases, judged on scalability and return rather than trend.
- Agile frameworks embedded in daily operations, not delivered as separate training.
- Marketing and product design teams aligned around a single version of the message.
- A new line of business opened, backed by prior market research.
- Performance dashboards so leadership decides on data rather than impressions.
The order matters more than the list. It is the same criterion we apply in business consulting: make the operation visible first, decide what to automate second. Inverting that order is the most common cause of an expensive system nobody uses.
How do you structure hiring for a sales team?
With a written induction framework and a defined performance measure agreed before the first hire. Without them, each rep's output depends on the luck of who trained them.
Three concrete interventions in this case:
- A role profile derived from the documented commercial process, not from a generic description copied off the market.
- An induction with fixed content held in the intranet, so ramp-up does not depend on a supervisor's calendar.
- Agreed, visible performance indicators, reviewed on the leadership dashboards at the same cadence every month.
That set is what turns a hire into an onboarding. The difference shows in month two.
What were the results?
Revenue grew by more than 35% and operational efficiency improved by 21%. Neither came from a single action: they came from ordering the operation and staying through execution until the change stuck inside the company.
| Dimension | People-dependent operation | Documented operation |
|---|---|---|
| Procedures | In the memory of whoever was in that day | In a searchable intranet, with one version in force |
| Ramping a new rep | Depends on who trains them and on their calendar | Fixed induction content, independent of the person |
| Software purchases | Judged on features and on trend | Judged on scalability and on return |
| Leadership decisions | On impressions and one-off reports | On performance dashboards reviewed at a fixed cadence |
| Growth ceiling | Whatever the key people can absorb | Whatever the process can absorb — and that can be widened |
The 21% efficiency figure is the one that usually goes unnoticed, and it is the one holding up the other: without it, 35% more revenue would have demanded 35% more structure. In practice that efficiency shows up in concrete places: time a rep spent working out a procedure went back into selling, and leadership stopped rebuilding by hand the information a dashboard now shows.
It is worth naming what did not happen, because it is the part that usually gets assumed. Headcount was not doubled, the sales floor was not replaced, and no enterprise platform was rolled out to hold the growth. The same team, working from written procedures and against visible targets, absorbed a volume it could not absorb the year before.
Was the result the market or the operation?
The operation. Costa Rica's automotive market was not pushing: it came off a growth cycle and moved into normalization.
According to figures from the vehicle and machinery importers association (AIVEMA) published in January 2026, 73,123 new vehicles entered Costa Rica in 2025, 4.82% fewer than the 76,828 of 2024. The association itself describes the drop as a normalization of demand after a significant growth cycle.
It was the first decline after four consecutive years of increases. And it was not only a volume decline: the composition of the market moved as well. In 2025 gasoline vehicles accounted for 53% of imported units, diesel for 28%, electric for 16% and hybrids for 3%, per the same source.
That matters to a premium vehicle retailer more than it looks. A 16% electric share means the product knowledge the sales floor held three years ago no longer covers the conversation, and that a new rep's induction has to keep pace with the market. That is only possible if the induction is written down somewhere: if it lives in a veteran's experience, it updates whenever that person has time.
Growing revenue more than 35% inside a contracting market is not a tailwind. It is selling better against the same available demand — which is exactly what changes when the process stops depending on people.
Where does a company start if it wants the same?
By writing down the process that generates revenue today, not the one that ought to exist. The sequence followed in this case is repeatable and needs nothing bought up front:
- Write the commercial process exactly as it runs today, shortcuts and exceptions included. Documenting the ideal version serves an audit, not an operation.
- Mark the points in that process that depend on a single person. Those are the ones setting the ceiling, and there are usually two or three, not twenty.
- Define how each of those points is measured before touching it: with no baseline there is no way to know whether the intervention worked.
- Centralize what is written in one place the whole team can search, with a single version in force and someone responsible for keeping it current.
- Only then evaluate tools, on scalability and return, against a process you already understand.
- Review the indicators at a fixed cadence. A dashboard looked at when there is a problem is a report, not a management system.
The order is not a methodological preference. It is what separates an intervention that stays inside the company from one that unwinds when the consultant leaves.
What can your company take from this case?
If your operation depends on specific people rather than documented processes, the growth ceiling is already set. The first step is not buying software: it is writing down how you work today and measuring how well it performs.
At ZENITAR we work this way because the result is measured in the client's operation and not in a report: in this case, 35% more revenue and 21% more efficiency at a Costa Rican company, with execution sustained until the change stayed inside.
If you are seeing the same symptom, two reads that follow from here: what to check first when the company stops growing and how to build a growth plan that actually gets executed. And if the bottleneck is already a systems problem, the work comes in through software and web.
