Does an ERP implementation fail because the software is badly configured? Rarely. It fails because the warehouse manager in Lousada kept using the radio instead of the terminal for six months, and nobody confronted him. It fails because the CEO promised "no loss of production" and the production team, afraid of looking incompetent, hid the problems. It fails because nobody calculated how much an hour of organisational resistance costs — and that cost, silent, devours the return on investment.
Change management is not a motivational workshop. It is the architecture of power.
The lie everyone believes: "The ERP will solve this"
An operations director at a textile factory in the Vale do Ave called us to an urgent meeting a few years ago. Implementation stalled. Finance would not validate data. Purchasing was in a panic. Production refused to log hours in real time.
We thought: technical failure, lack of training, wrong configuration. We were wrong.
The problem was deeper. The production director had told the team that the ERP was "one more thing to fill in" — an extra task, not an instrument that would simplify the work. The accountant, for 20 years, had validated invoices by hand, line by line. Now the system was asking her to trust an algorithm. The head of purchasing feared that data transparency would make him "dispensable" — if the CEO could see all the lead times in real time, why would he still consult him?
None of these problems was technical. They were all political.
We see it repeat itself: companies buy a MULTI ERP or a QAD Adaptive expecting the software to solve management. The software does not solve. The software exposes. And when reality is exposed — inefficiencies, errors, lack of control — people resist. Not because the software is bad. Because the truth is uncomfortable.
The ERP does not fail for lack of features. It fails because people realise they are going to become more exposed.
The invisible cost of resistance
There is a metric that nobody measures: how much, in person-hours, does the time spent on workarounds cost? An operator who should be logging production in the system but keeps using a sheet of paper. A warehouse manager who validates picking in Excel because he "doesn't trust" the interface. A salesperson who insists on sending orders by email, in parallel with KORA B2B.
Multiply those hours by 50 employees, over six months, and then by an average hourly cost of €15 — and you get a frightening number. And nobody reported it, because it does not appear in any column of the project.
Resistance is not a communication problem. It is an incentives problem. If the operator's performance evaluation system is still "production hours logged", he does not care whether the log is on paper or in the system. If the warehouse manager's bonus is still "picking without errors", he has no reason to trust that the KORA Inventory picking system will improve things — it may make them worse, if the system is badly calibrated.
We implemented projects where we thought change management was 70% communication and 30% technical training. We were wrong. It is 30% communication, 30% training, and 40% redesign of processes and incentives. If you do not change the process — and, more importantly, if you do not change who wins and who loses with that change — resistance is rational. It is not ignorance. It is legitimate defence.
The political map of the implementation
Every company has a geography of power. The CEO wants "visibility". The CFO wants "cost control". The production director wants "less bureaucracy". The warehouse manager wants "to carry on as before". These objectives are not compatible — at least, not to begin with.
A successful ERP implementation begins by drawing that map. Not in "stakeholder alignment" meetings. In private, honest conversations with each of the key players. What do you gain from this? What do you lose? What do you fear? And then: how do we redesign the system so that you gain more than you lose?
A real example. A tool distribution chain in Paços de Ferreira needed to implement integrated warehouse management. The logistics manager was afraid of one thing: that the system would make him "invisible" — if everything was automatic, what were 20 years of experience for? The solution was not to train him better on the software. It was to redesign his role: instead of "executing picking", he became "optimising picking routes and resolving exceptions". He gained responsibility, he did not lose it. The resistance disappeared.
This takes time. Time that projects do not have, because the CEO wants "go-live in 14 weeks". But a rushed implementation that fails in month eight is far more expensive than a careful 20-week implementation.
What nobody says about training
Companies invest 5% of the project budget in training. They should invest 20%. And not in the training they do — which is "here is the button, you click here". But in the training that costs: acknowledging that people are afraid, that the change is real, and that it is no shame to take time to adapt.
An operator at a weaving mill in Guimarães, with 30 years on the machine, does not need a 200-page manual. He needs someone who sits beside him, for a week, and says: "You knew how to do this on paper. Now we are going to do it in the system. It will be strange. We will make mistakes. That is normal." He needs permission to be slow, for a while.
Technical training is necessary. But emotional training is what makes the difference. And it is what nobody budgets for.
There is a second problem: the "power user" who trains everyone. Often it is someone the CEO chose because they are "quick to learn", not because they are good at teaching. That person absorbs everything, becomes indispensable — and then leaves the company, or becomes overloaded. They never created redundancy. They never documented the knowledge. They never delegated.
The go-live date is an illusion
An ERP project does not end at go-live. It ends at "stabilisation" — and stabilisation can take 12 to 18 months. But the budgets, the timelines, the CEO's attention, all disappear at the go-live date. The system is left orphaned, and the problems that arise afterwards are "post-implementation problems", which nobody had foreseen.
This is especially serious in sectors such as footwear or textiles, where there is seasonality. You implement in January, everything goes well with low demand. July arrives, and the system collapses because nobody tested it with real volume. Or you implement in October, everything works during the high season, and in January — when demand falls — nobody understands why the system "now doesn't do what it did before" (because the real data changed, and the configuration was made for a different scenario).
The solution is obvious: set aside 20-30% of the budget for the first 12 months post-go-live. Keep the project team available. Define "post-implementation health" KPIs, not just "go-live on time and on budget".
The metric that matters: how much is the time you saved worth?
A CEO of a garment company in Vila Nova de Famalicão asked us: "What is the ROI?" The honest answer is: it depends. It depends on how much time you save on manual tasks. It depends on how many errors you stop making. It depends on how many faster decisions you manage to take.
But there is one metric that few calculate: how much time does your IT team spend "fixing things"? In companies without an ERP, it is common for a technician to spend 30% of the time on patches, backups, data recovery, synchronisation between systems. In a well-implemented ERP, that drops to 5-10%. That difference is pure saving.
Or, looking at the other side: how much time does the operational team spend "looking for data"? An accountant who spends two hours a day looking for invoices in email, in scattered files, in two different systems. A centralised ERP, with integrated document management, reduces that to 20 minutes. Multiplied by 250 working days a year, and by a cost of €20/hour — that is €1,400 of value per person, per year. In a company with 50 people, that is €70,000.
But that calculation requires you to measure the current state. And many companies do not measure.
The Portuguese context: the silent urgency
In 2025, only 53.7% of companies in Portugal with 10 or more people used integrated business management software. This means that close to half of the Portuguese business fabric still does not have an ERP — it continues to operate with spreadsheets, parallel systems, fragmented information. This is not a choice. It is, often, fear.
Fear of losing control. Fear that the software will expose inefficiencies that the CEO does not want to see. Fear that an implementation will fail and cost money that is not there. And, honestly, that fear is justified — because most of the implementations we see fail do not fail because of the software, but because of change management.
The RRP sets aside €650 million to support the digital transition of Portuguese companies. But the money is there. What is lacking is confidence that, this time, the implementation will work. And confidence is built with honesty: about what can go wrong, about the real time it takes, about the fact that change is uncomfortable, but is possible.
The inconvenient truth
Change management fails because companies hire consultancies to do it, and consultancies have an incentive to say "everything is fine". If a consultancy says "this implementation is going to fail because there is an unresolved political conflict between the CEO and the CFO", it is dismissed. If it says "let's do a 'shared vision' workshop", it is paid for two more months.
We, as a software vendor, have a different incentive — we want the client to succeed, because success means reference, contract renewal, and recommendation. But even so, it is tempting to say "the software solves this" when the truth is "the software exposes this, and then you have to deal with what you see".
The difference between an implementation that works and one that fails is not the software. It is the courage to say, in month three, "this is not going well, and we need to change strategy" — instead of pretending everything is fine until month eight, when the project collapses.
If you are going to implement an ERP — whether MULTI, QAD Adaptive, or any other — start by accepting this: the software is the easy part. The change is the hard part. And the change is not solved with features, with training, or with communication. It is solved with a redesign of power, of processes, of incentives, and with time.
Everything else is illusion.
Frequently asked questions
Why is change management more important than the technical configuration of the ERP?
Because a well-configured ERP merely exposes the reality of the organisation. If people resist, it is not a technical failure — it is because the change affects power, responsibilities and job security. Resistance is rational. Without a redesign of processes and incentives, the software solves nothing.
How do you identify who will resist the ERP implementation?
Map the company's geography of power. Who gains and who loses with data transparency? The warehouse manager who feared becoming "dispensable" if the CEO could see everything in real time. The accountant who trusted her manual judgement. Private, honest conversations reveal the real fears — they do not appear in formal meetings.
What is the real cost of resistance in an ERP implementation?
Nobody measures it, but it is enormous. An operator who keeps using paper instead of the system, multiplied by 50 employees over six months, at €15/hour — results in tens of thousands of euros lost. That "invisible" cost devours the return on investment because it does not appear in any column of the project.
How should training in an ERP implementation really be budgeted?
Companies invest 5% of the budget in training. They should invest 20%. But not in "click here". In emotional training: acknowledging fears, allowing initial slowness, sitting beside the employee for a week. An operator with 30 years of experience needs permission to be slow, not 200-page manuals.
What is the "power user" and why is it a risk?
It is the person quick to learn whom the CEO chooses to train everyone. They become indispensable, absorb all the knowledge, never document or delegate. When they leave the company or become overloaded, there is no redundancy. It is a single point of failure that compromises the sustainability of the implementation.
How long does it really take to stabilise an ERP implementation?
Go-live is not the end — it is just the beginning. True stabilisation takes 12 to 18 months. But the budgets, timelines and CEO's attention disappear after go-live. That support gap is where many projects fail silently, months after launch.
How do you redesign an employee's role to eliminate resistance?
Instead of training better on the software, change what the person does. A logistics manager afraid of becoming "invisible" if everything were automatic became an "optimiser of routes and resolver of exceptions". He gained responsibility, he did not lose it. Resistance disappears when the person comes out ahead with the change.