Let's be honest. There was champagne on the go-live day. The consultant took a photo with the team. Someone printed a «successful implementation» certificate. And general management breathed a sigh of relief: Finally, SAP is live.
Three years later, that same company calls the same consultant for the same requirement that they already paid for twice. The «key user» they trained quit. The fixed assets module was never used. And month-end closing is still a four-day odyssey because the only person who knows how to run the process is on vacation.
That's not a successful implementation. It's a well-disguised dependency.
This is the uncomfortable truth no one tells you in a proposal presentation: SAP post go-live dependency It's not an accident. In many cases, it's the predictable outcome of a system designed to perpetuate it.

The Most Expensive Myth of Your SAP Implementation
Every SAP project includes, at some point, the following promise: «We are going to empower your key users to be autonomous post go-live.»It's a nice phrase. Reassuring. And in most cases, it's pure theater.
The «key user» model — or key user— was born with a reasonable logic: instead of training everyone, you concentrate knowledge in 5 or 10 strategic people who then replicate downwards. It sounds efficient. The reality is that it works more like a funnel than a multiplier.
Think about it coldly. Your key users received between 40 and 80 hours of training. They learned the basic operational flows of their area. But they didn't learn the data architecture, the parameterization logic, the configuration criteria, or how to connect the modules to each other. The consultant knows that. And the consultant doesn't document it, because their business isn't about you being independent.
Therefore, consequently, what your «key users» actually know is how to operate the system within the lanes that the consultant set up. If the lane fails, the truck stops. And you call the consultant.
The Nobody Does Knowledge Arithmetic
Let's look at the numbers no general manager wants to see in a board meeting.
If your company has 5 key SAP users, and the average turnover in middle management in Latin America is between 20% and 30% annually, you have this:
- Year 1 post-go-live: probably 1 key user already left.
- Year 2: maybe 2 more. One was promoted and no longer operates the system. The other resigned for a better offer.
- Year 3: You have 2 left, if you're lucky. And one of them was on the original project, so he's the irreplaceable «expert.».
Is there documentation of the knowledge that has left? In 80%% of cases, no. Is there a succession plan for SAP knowledge? Almost never. Does your company have a specific onboarding process for SAP when someone new arrives? Rarely.
That being said, the most painful statistic: according to McKinsey, 70% of ERP implementations fail to meet their original objectives. And Gartner projects that by 2027, more than 70% of recent ERP implementations will have failed to deliver on promised business value. The number one cause, in both studies, is not technical. It's human: adoption, change, knowledge.
The system works. People don't know how to use it.
The Perverse Incentives Nobody Names in the Meeting Room
Here comes the most uncomfortable truth of all. Prepare yourself.
Your SAP consultant makes money from your doubts. Every support ticket is income. A «configuration adjustment» request adds another billable hour. And when you call because «the system isn't doing something it should be doing for me,» the meter runs.
Now, what's the financial incentive for that same consultant to teach you to solve that yourself? None. Quite the opposite: if they teach you too well, you lose the need to call them.
I'm not saying all consultants are dishonest. There are excellent professionals in this industry. What I'm saying is that the Business model is structurally designed for SAP dependency post go-live. Consulting firms don't live off completed projects. They live off support contracts, never-ending phases 2 and 3, and optimizations that always need to be done.
And the company — your company — accepts that model because no one wants to admit they paid millions for a system that operationally depends on an outside vendor. That admission is too uncomfortable for the board.

The Single Point of Failure That Has a Name and a Last Name
Let me ask you a direct question: how many people in your company can, today, execute a complete month-end close in SAP from start to finish, without external help?
Who knows how to parameterize an account determination rule? How many understand why a purchase document doesn't flow to the inventory module? Can anyone diagnose why a cost report is showing inconsistent values?
The honest answer in 80% of the companies I know: one person. Two at most.
I call that profile the single point of failure with first and last name. It concentrates all the operational knowledge of the system. It cannot take a vacation without the team panicking. And if it decides to leave tomorrow, it will take all the critical know-how of the entire operation with it.
That's not a human asset. It's a critical vulnerability.
Interestingly, that profile often knows they have that power. And in some cases—not all, but many—they use it. Not maliciously, but unconsciously: they don't document because documenting would reduce their perceived value. They don't train others because that would reduce their position as a key player. And the company, without realizing it, has created a hostage of knowledge within its own payroll.
The Cemetery of Unused Modules
Now add another layer to this problem. Not only is knowledge concentrated in a few people. But, in addition, only a fraction of what was paid for is used.
In the SAP Business One implementations I've audited or assisted with, the pattern repeats with a regularity that no longer surprises me: the company activates Finance, Sales, and some Inventory. And that's where it stops. The Production module is «something we'll look at in phase 2.» Fixed Assets is «configured later.» CRM is «we're not ready.» Projects are «when we have time.».
Phase 2 rarely arrives.
The SAP license management industry has a name for this: Shelfware. Purchased software that is never used, but whose annual maintenance is still being paid. In SAP, annual license maintenance runs around 20% of the original license value. Which means that every year you pay 20% of something you never activated.
According to industry data, companies that bought 100 ERP modules 10 years ago use, on average, only 40. The rest is buried budget. Negative ROI sustained year after year, with no one performing an autopsy.
The reason is simple: modules that no one knows about are not activated because activating them implies knowledge that doesn't exist internally. And the consultant who could activate them charges for doing so. So the cycle closes: post go-live SAP dependency freezes the capacity for evolution.
What's the Real Value of the Go-Live You Celebrated
Let's do some more awkward math. Not the project kind. The Year 3 kind.
SAP Business One implementation in a medium-sized company in Latin America: between 30,000 and 80,000 dollars, depending on complexity. Add to that the licenses, infrastructure, training costs, and the time of the internal team. You easily exceed 100,000 dollars in initial investment.
Now, what happens next? The consultant's annual support: between $3,000 and $15,000 per year, depending on the contract. Additional configuration requirements: $2,000 to $8,000 annually, on average. Retraining when new staff comes in: more billed hours. Total accumulated in 3 years: potentially more than the original implementation. And that doesn't even count the internal time the company's «SAP expert» spends solving problems that were theoretically already resolved at go-live.
However, few companies do this calculation. Because acknowledging it would force difficult decisions. And it's more comfortable to keep calling the consultant.

The Way Out: It's Not a Project, It's a Decision
What I'm describing is not a condemnation. It's a diagnosis. And every correct diagnosis opens the way to treatment.
The output of post-go-live dependency doesn't start with technology. It starts with a strategic decision that the general manager or the board must make: We want to be autonomous in SAP, or we'll continue to be captive clients.
That sounds obvious, but it isn't. Because autonomy in SAP has a transition cost that many companies are unwilling to pay in the short term. It requires documenting processes, structured training (not reactive), building redundant knowledge (more than one person per critical process), and negotiating with the consultant from a different position: not «we need you to fix this for us,» but «we need to understand how this works so we can operate it ourselves.».
Concrete actions to escape the trap
Audit the knowledge, not just the system. Before auditing which modules are active, ask yourself how many people in the company truly know how to use them. That's the real X-ray of your SAP dependency post go-live.
2. Require operational documentation as a deliverable from the consultant. Not just technical manuals. Step-by-step procedures for critical processes: month-end closing, inventory management, billing, purchasing. If the consultant doesn't deliver them, it's a sign they're not interested in your autonomy.
Design a succession plan for SAP. For each critical process, there must be at least two people in the company who know how to execute it without external support. Treat SAP knowledge as an asset that requires redundancy, just like a server or a production process.
The next level of autonomy
4. Review paid modules vs. active modules. Take stock. What did you buy? Are you actually using it? How much are you paying in maintenance for something you never activated? That analysis can reveal opportunities for savings or, conversely, for activating features that are already yours and that you are not taking advantage of.
Renegotiate the support contract. Not from the position of a dependent client, but from the position of a company that wants to build internal capacity. The best consultants understand that difference. Those who don't understand it are probably not the consultants you need long-term.
If you want to dive deeper into how dirty data within SAP amplifies this dependency problem, this article develops it without filters. And if you recognize the pattern of power exercised over the ERP from the directory, The tyranny of the standard also has its anatomy here.
The Question You Should Ask Yourself This Week
I'm not asking if SAP is well implemented in your company. That doesn't matter so much anymore.
If your lead consultant tells you tomorrow at 8 AM that they have an emergency and cannot attend to you for the next 30 days, what happens to your operation?
Can your team close the month? Do you have the ability to issue invoices without external support? Does anyone know how to process payments? Can you generate the reports the board requires on the first Monday of the month?
If the honest answer causes discomfort, you've just identified your real operational risk. It's not a technology problem. It's a problem of sovereignty over your own system.
SAP dependence after go-live is not a death sentence. It's a choice. Most companies choose it without realizing they're doing so. The first step to getting out is recognizing you're in.
The go-live was three years ago. The question is whether your company has matured enough to operate on its own.
This article is part of the series Uncomfortable Truths That SAP Reveals, published on hectorpincheira.com. If it makes you uncomfortable to read, that's a good sign. It means that something you're describing is happening in your company.




