Your company grew and is starting to lose control: four growth crises, with Mexican cases

You go on vacation for a week and come back to three problems nobody wanted to decide. Sales are going well, but there is never enough money. The same question reaches five different people on WhatsApp.

And every time you hire someone, everything seems to go slower, not faster.

If this sounds familiar, you are probably not doing anything wrong. Your company grew. What is happening to you has been studied for more than fifty years [1][2], and it has also happened to very large Mexican companies.

One clarification before we start: Hábil did not work with any of the companies that appear here and did not take part in these cases. We tell them using what the companies themselves reported publicly (their annual reports and their filings with the Mexican Stock Exchange or with the U.S. Securities and Exchange Commission, the SEC) and reputable press, each fact with its reference number. When something is our reading, we say so.

In 1 minute

• The risk: growing with the same way of working you started with. What worked with ten people can hold back fifty. Larry Greiner described it in 1972: each stage ends in a crisis that calls for a different way of organizing [1][2].
• The case: Cemex began unifying its systems in 2001. When it bought RMC in 2005, for about $6.5 billion, it already had the means to integrate it; in 2006 it reported the integration complete [14][15][16].
• The cost of not acting: Cemex warned that achieving about $360 million in annual savings depended, among other conditions, on standardizing its processes [15]. And, according to Greiner, many companies stall, fail or end up being sold during the transition [1].
• What you have to decide on Monday: which of the four crises you are in and which system is worth fixing before the next one. At the end there is a checklist to take to your meeting.

In this article

1. How common is it to grow "by gut feel" in Mexico?
2. What does Greiner's model say, in short?
3. What stage is your company in?
4. Why does everything still go through the owner?
5. Why do good people leave when they are not allowed to decide?
6. Why do you get three different numbers for the same question?
7. When do controls start getting in the way?
8. What breaks in your systems at each stage?
9. Which mistakes cost the most?
10. Who sits in each chair in your company?
11. What should you bring to Monday's meeting?
12. Key takeaways
13. What does Hábil do?
14. Frequently asked questions

How common is it to grow "by gut feel" in Mexico?

Even among SMEs, few are run on the numbers a larger company needs.

• Only 9 in 100 SMEs track ten or more indicators: 8.8%, according to the 2018 ENAPROCE survey by INEGI (Mexico's national statistics institute) and the Ministry of Economy. 28.2% track three to five [7].
• Two in three microbusinesses track no indicator at all to know how they are doing: 66.5%, according to the same survey [7].
• Almost half of all establishments keep no accounting records: 46.5%, according to the 2019 Economic Censuses [5].
• An establishment has, on average, a life expectancy of 7.8 years, according to the same census [5][6].
• 95 in 100 businesses are micro, with 10 people or fewer: 95.4% of the economic units in the private and state-owned sector in 2023, according to the 2024 Economic Censuses [3][4].

Growing without numbers works while the owner can be everywhere. When that is no longer possible, the first crisis arrives.

What does Greiner's model say, in short?

Each way of organizing works for a certain size, and stops working when the company outgrows it.

In two lines: in 1972, Professor Larry Greiner published in Harvard Business Review that companies grow through calm stages that end in a crisis; the journal republished it in 1998, with a commentary by him [1][2]. The practices that helped a company grow can stop working when its size changes [1][2].

Three points from the article that are useful to you:

• The stages last years, not months. In 1972, Greiner observed that companies that get through a crisis usually have four to eight years of continuous growth. In 1998 he widened the range: from 3 to 15 years per stage [1].
• Your industry sets the pace. In fast-growing markets the stages are shorter; in mature markets, longer [1].
• Age is not the only factor: size matters too. Problems tend to change as headcount and sales volume rise; a company that does not grow can keep the same practices for a long time [1].

Greiner describes five stages, and in 1998 he suggested that a sixth might emerge [1]. Here we take the four crises he describes in detail: that is our selection, not a rule of the model.

What stage is your company in?

Locate your crisis by the signals, not by the org chart: the phrase your team repeats most tells you where you are.

Greiner does not set how many people correspond to each stage. For a size reference, Mexico does have official cut-offs, published by the Ministry of Economy in the Official Gazette of the Federation in 2009 [28]:

The headcount cut-offs match the ones INEGI uses in the ENAPROCE [7][28].

The next table crosses those sizes with the four crises. It is a guide and it is our reading: your company may get there sooner or later, depending on its industry and its pace of growth [1].

And here is what you can count on Monday to confirm it. These are guide indicators and our reading, with no official threshold: what matters is whether the figure rises month after month.

A detail Greiner stresses: each part of the company can be at a different stage [1]. For example, your plant can be in crisis 3 while sales is still in crisis 1.

Why does everything still go through the owner?

The owner who does everything was the strength of the small company and is the bottleneck of the one that grew.

What it looks like: the owner sells, collects, hires, signs checks and puts out fires. While the company is small, that works. As it grows, everything stops when the owner is not there.

Greiner describes it this way: the founders find themselves carrying management responsibilities they do not want, and at that stage new ways of keeping the books are needed to get control of the money [1].

What resolves it: someone else takes charge of making the company run every day, so the owner can decide where it is going. For Greiner, the first critical decision is to find a strong manager the founders accept [1].

A Mexican example: Bimbo. It started in 1945 as a small bakery with four products [11]. Today it presents itself as the largest bakery in the world [10]. How the leadership was split, in its own reports:

The chief executive changed from one report to the next [10][11]. Bimbo was already a multinational: it does not illustrate the crisis of a small company, but the principle of separating whoever sets the direction from whoever runs the operation.

What your system has to withstand: information that stops living in the owner's head. If the only person who knows how much each customer owes is you, nobody else can decide. Every sale, invoice and collection, in a single place.

Why do good people leave when they are not allowed to decide?

When department heads know more than the leadership and cannot decide, the company loses its best people.

What it looks like: there are department heads now, but everything still waits for approval from the top. According to Greiner, the people lower down already know the market and the machines better than their bosses, and they get caught between following procedure and taking the initiative [1]. If nobody lets them decide, they become disenchanted and leave [1].

What resolves it: delegating decisions, not just tasks, with clear limits. For example: "up to 5,000 pesos, the department decides; above that, the leadership team." Greiner warns that it is not easy: a director who succeeded by giving orders finds it hard to let go [1].

What your system has to withstand: each area being able to decide with data without asking anyone for the report. That calls for role-based permissions (who sees what, who approves what and up to what amount) and for the system to keep a record of who decided. Delegating without that record is delegating blind.

Why do you get three different numbers for the same question?

When each area calls things by different names, nobody can add up the company; unifying before acquiring is what makes faster integration possible.

What it looks like: each area, branch or acquired company does things its own way. Each one has its own Excel file, its own system and its own names for the same thing. The director asks for one number and gets three.

According to Greiner, managers with freedom prefer to run their own business without coordinating plans, money, technology or people with the rest of the company; each one sees only their own part, and the top feels it has lost control [1].

What resolves it: coordination. Common rules and tools so that everyone pulls in the same direction, without taking away their day-to-day decisions [1].

What did Cemex do before its largest acquisition?

In a doctoral thesis on the company, a Cemex vice president of operations recalled the origin of the project: one of the regional vice presidents said that he «no podía manejar seis países en centro y Sudamérica donde todo mundo le llama diferente a las cosas…» [could not manage six countries in Central and South America where everyone calls things by different names…] [17].

Cemex's filings with the SEC document the sequence, with dates and figures:

What matters for you is the order: first a common language and common systems; then the big acquisition. Cemex itself tied the savings to standardizing processes [15] and said those platforms let it integrate its acquisitions faster [16].

The story does not end there. In 2007 Cemex bought the Australian company Rinker for more than $15 billion, including debt [18][31]. In March 2009 two rating agencies stripped it of its investment grade; most of its debt came from financing its acquisitions, Rinker included, and its investments [29].

In August 2009 it refinanced about $15 billion with some 75 banks and private creditors [19][30]. Our reading: the order let it integrate quickly; it did not protect it from the debt of the next acquisition.

How did Bimbo add companies and a single system at the same time?

While it kept acquiring (five companies in 2024 and several more in 2025, such as Wickbold in Brazil and Karamolegos in Romania), Bimbo kept moving its operations to a single cloud-based management system [10][11]:

• 38% of its operations at the end of 2021 [8];
• 70% at the end of 2022 [9];
• 90% deployment in 2025; that report no longer says "of its operations," so the figure may not measure the same thing [11].

Bimbo does not say that one was meant to serve the other; that connection is our reading.

What your system has to withstand: a single source of truth. "Customer," "sale" and "order" have to mean the same thing across the whole company, and each area's systems have to talk to one another.

At scale, the same Cemex: at the end of 2025 it received about 60% of its orders through Cemex Go, its digital platform for customers [20][21]. Cemex reports that this platform brings together its online store, its sales team and its service centers to give customers the same treatment through any channel, that it covers everything from order to payment, and that, with Cemex Go Link, its customers' systems connect directly to its own [20][21].

When do controls start getting in the way?

The controls that resolved the previous crisis become the next one when they cost more than they deliver.

What it looks like: to resolve the previous crisis, reports, forms, committees and approvals were put in place. It worked… until filling in the form became more important than serving the customer.

Greiner calls it the red-tape crisis: the systems and programs put in place to coordinate start to cost more than they deliver, and distrust grows between headquarters and the field [1].

What resolves it: simplifying. For Greiner, the way out is to work in teams that solve problems quickly, with simpler formal controls, combined into multipurpose systems, and real-time information built into day-to-day decisions [1].

What your system has to withstand: approvals and reports that come out of the operation's own records. If a system asks people to enter the same data three times, the system is part of the red tape.

An aside: splitting up and simplifying the group. This is not a diagnosis of these companies.

Three Mexican groups reorganized the structure of the group, not the operation. We do not present them as cases of any of the four crises: their sources talk about strategy and portfolio.

• Herdez, in 2025. It spun off its stores (Nutrisa, Moyo, Cielito Querido Café and Chilim Balam) into Grupo Nutrisa, which has been listed on the Mexican Stock Exchange since September 18, 2025 [12][13]. The reason it gave: that each business would operate «con mayor autonomía estratégica y acceso a capital propio» [with greater strategic autonomy and access to its own capital] [12].
• Alfa, from 2020 to 2025. At one point it had four businesses: lightweight auto parts (Nemak), technology and communications (Axtel), petrochemicals (Alpek) and food (Sigma) [25]. From 2020 it handed the first three over to its shareholders, one by one, until it became a single food business [25][26]. The process ended on April 7, 2025 [26][27].
• FEMSA, since 2023. On February 15, 2023 it announced a plan to focus on its core businesses, sell its stake in Heineken and seek alternatives for other businesses it did not consider strategic [22][23]. Its chairman said those actions would significantly simplify its structure [22]. In July 2025 it closed the sale of part of its logistics business [24].

What breaks in your systems at each stage?

Each stage calls for different data; the technology decision is best made one stage before you need it.

Greiner wrote in 1998 that each stage needs information technology in a different way: a functional structure calls for revenue and costs by area; a decentralized one, the profit of each unit [1]. And control shifts from market results to costs, then to the profit of each unit and after that to plans and investment [1].

Translated into what breaks and what is worth deciding beforehand. The last two columns are our reading:

Which mistakes cost the most?

The most expensive mistakes tend to be about diagnosis and sequence, not technology.

1. Going back to running everything from the top. Greiner observed that some leadership teams, when they feel they are losing control, try to return to centralization [1]. Consequence: according to him, it usually fails, because the company already operates at too broad a scale [1].
2. Not knowing what stage you are in. If the leadership does not know the stage, it may not see that the time to change has come, or it may impose the wrong solution [1]. Consequence: you fix the wrong crisis.
3. Skipping stages. Greiner warns that, out of impatience, it is tempting to skip one; but each stage leaves lessons the next one needs [1]. Consequence (our example): committees before knowing who does what, which can add red tape without giving control.
4. Delegating without data or an audit trail. Our reading. Consequence: the decision is delegated but the information is not; the leadership goes back to reviewing everything and the delegation collapses.
5. Acquiring or opening before unifying. Our reading, with Cemex as a contrast [14][15]. Consequence: each new company or branch adds its own system, and the consolidated figure can take weeks.
6. Automating the red tape instead of removing it. Our reading. Consequence: the useless form gets filled in faster, but it is still useless.

Who sits in each chair in your company?

Each crisis forces a function to change hands; naming them is a cheap way to prepare.

It is a tool of ours, not part of Greiner's model.

At first, the owner sits in all four. Each crisis forces a chair to change hands or gain its own authority.

You do not need to hire four people at once: you need to put a name on each chair, even if one person sits in two.

What should you bring to Monday's meeting?

Twelve yes-or-no questions, three per crisis, help you locate yours and where to start.

Answer them with your leadership team. The list is a tool of ours, not part of Greiner's model.

How to read it: the crisis with the most "no" answers is yours. If there is a tie, start with the lowest-numbered one: each stage leaves lessons the next one needs [1].

And three steps for after the meeting:

1. Put a name on the four chairs. If yours appears on more than two, you already know where to start.
2. Write down the two processes that hurt the most, with their decisions: what happens if the customer does not pay, if the material does not arrive, if the manager does not approve.
3. Decide now on the system for the next stage: the last column of the table "What breaks in your systems at each stage?" tells you which one.

Key takeaways

Three questions for your next leadership committee:

1. Which of these four phrases does our team repeat most: "I can't handle everything anymore," "let me decide," "I no longer know what's going on" or "there's more paperwork than work"?
2. If tomorrow we bought a company or opened ten branches, would our systems be able to add them up from the first month?
3. What technology decision are we putting off that is going to hurt us at the next stage?

What does Hábil do?

Hábil builds and connects the systems a company needs when it no longer fits in a spreadsheet.

In terms of what this article covers, that comes down to three things:

• What lives today in spreadsheets and emails, moved into a system with permissions, history and reports, so that information stops depending on a single person.
• Approval workflows: who requests, who reviews, who approves and in what order, with the audit trail of each decision in plain sight. Delegating without delegating blind.
• Your ERP and your other systems exchanging data, so you stop entering the same thing twice.

And if part of the red tape of crisis 4 is checking by hand what your suppliers invoice you, Cuadre continuously checks your CFDI invoices and notifies you on WhatsApp when a new one arrives.

Before writing a single line, we review with you whether something already exists that solves it. If it gets built, the code and the documentation stay in your repository, in your name.

See Build your custom system and the article Build, configure or buy.

If you recognized yourself in any of these crises, let's start with the one that hurts the most. Request your first conversation, at no cost. In it we review whether it makes sense to configure, to use a product or to build, and what you would receive in each case. Write to us on WhatsApp.

Frequently asked questions

Does every company go through the four crises?

Not necessarily, nor in the same order or at the same size. Greiner presents the stages as a pattern, not as a calendar, and makes clear that his model is not a recipe [1].

Does this apply to an SME or only to companies like Bimbo?

It can be useful for an SME too. We use large cases because they publish their decisions; in an SME, the same crisis shows up as an owner who signs everything or as three versions of the same report.

Did Hábil work with Bimbo, Herdez, Cemex, FEMSA or Alfa?

No. This article is based only on public sources: their annual reports, their filings with the Mexican Stock Exchange (BMV) and the SEC, and the press cited in the references.

Sources consulted on October 22, 2026. Quotations from Spanish-language documents are kept in the original, with a translation in brackets.