White Papers
Installed or implemented? Why manufacturing systems succeed or fail on the factory floor
19 November 2026

Armand Janse van Rensburg, Operations Director, Lancea Konsult
Drawing on implementation experience of more than 100 factories, from fast-moving consumer goods to metal fabrication and pharmaceuticals
Abstract
Most manufacturing system projects end with the software installed. Far fewer end with a factory that runs better because of it. This paper asks why, and argues that the answer lies less in the software than in where the implementation starts: from the system's functions, or from the operation's real requirements. It draws on my career moving between running factory operations and advising them, on more than 100 factories from fast-moving consumer goods to metal fabrication and pharmaceuticals, and on research into manufacturing system implementations.
The paper makes five points. Much of a factory's real requirement is never said out loud, so it never reaches the requirement workshop. Systems are built to theory and generic models, while each plant's edge lies in practice built up over years, which is why some tailoring is always needed. A requirement is worth challenging, because it often comes from what a customer has read or has always done. A factory serves the whole business, and a few well-chosen numbers, usually two or three, connect the floor to sales and profit. And the right system is the one that fits the operation, not the one with the most features.
The implication is that an implementation needs operational experience working alongside systems knowledge. The paper closes with four things an owner or operations manager can do on Monday.
If you're choosing or about to implement a manufacturing system, sections 3 and 5 are the place to start. If you're already running one and getting less from it than you expected, start with sections 4 and 6.
1. Installed is not the same as working
I'll start with a generalisation about my own profession, and I know it is a generalisation. As consultants, a lot of the time we try to fit a factory to a system's functionality. Every system has its limitations, so what happens is that a specific operational challenge or requirement gets fitted into the system's packaged box. If you start from the other side, take the operational requirement first and match the system to it, you actually start to solve the operational challenge. In my experience that gap, between tackling an operation from the system's point of view and tackling it from the operation's, is quite a big one.
It is also the gap that decides whether a system is used. Mapping the operation's reality back to what the system can do is where a solution is either adopted or falls flat once the factory is running.
Research points the same way. A 2024 study followed three Belgian manufacturers that deployed the same manufacturing execution system. The outcomes were similar in some ways and quite different in others, and the researchers traced the differences to the decisions each company made while selecting, designing, implementing and using the system, and to how the work was divided between employees, management and developers.¹ A 2025 review of 272 papers on ERP implementation found the top three success factors to be top management support, the organisational fit of the system, and user training,² none of which is a software feature.
This paper is about that starting point, and about why it matters that the person finding out what a factory needs has worked in one.
2. Where I'm looking from
I studied industrial engineering at the University of Pretoria, and my career has moved between running operations and advising them. I started in manufacturing planning (inventory, production scheduling, sales and operations planning, and then procurement), with time on the floor making sure the schedule was kept. I then worked in supply chain consulting, went back into operations to run supply chain and business improvement at Jumbo Brands, and joined Lancea Konsult. Since then I've probably been to more than 100 different factories, from fast-moving consumer goods to metal fabrication to pharmaceuticals. Each move back into consulting came with a clearer reference for what the operational requirements really are.
As an industrial engineer, a profession concerned with "the design, improvement and installation of integrated systems",³ I tend to start with how the operation could run better, and treat the system as one of the means to get there. A systems specialist starts from what the software can do and how to configure it well, and every implementation needs that too. What I've found valuable is bringing the two together: the operational experience, and an understanding of what systems can do. The rest of this paper is written from that point of view.
3. The real requirement is mostly unspoken
Every operation runs on things people do without naming them. To the people who run it, it is common sense, the way normal working life goes, so nobody spells it out for an outsider. Walking a floor with a customer, or sitting with them through their business processes, I find these in the way they approach their planning, track their efficiencies, or review their production schedule and throughput. They are key to how the plant runs, but they don't come up in meetings. A requirements workshop records what is said; the real requirement includes what isn't.
If you have run a schedule or chased a shortage yourself, you quickly recognise the workaround when you see it: how a supervisor decides which job goes first when two are late, which numbers the production meeting really looks at, which material is always counted by hand because nobody trusts the system figure. It is the language of the domain, and you learn it in the trenches, not from theory or from systems knowledge alone.
Research on the shop floor recognises the same problem. A 2018 study in the International Journal of Production Research found that tacit knowledge was not being transferred between production workers, and that the paper checklists generated from MES and ERP systems left workers with either too much information or not enough of the right information.⁴ Practitioner research also gives familiar advice: Panorama Consulting's ERP reports recommend defining business requirements and aligning software with operations before choosing a system, and name business process work and organisational change management as the two most important success factors.⁵ ⁶ Yet that advice only works if the requirements you define include the unspoken ones.
The question to ask: has the person designing your solution walked your floor on a normal shift, and what did they see that wasn't on the requirement list?
4. Systems are built to theory; plants run on practice
Systems are generally built to theory: to what everyone expects to happen in principle, which is often not what happens on the shop floor in practice. Even the international standard for connecting business systems to the factory floor, ISA-95, treats them as separate worlds. Business planning and logistics (where the ERP sits) and manufacturing operations are different levels. Schedules and work go down from one to the other, and actual production performance has to come back up.⁷ So the plan and the floor are expected to differ, and someone has to reconcile them. When nobody does, the floor finds its own way, and that is usually where the spreadsheets and side systems come from.
Practice also differs from plant to plant. A food production plant is quite different from a metal fabricator, which is very different again from a pharmaceutical supplier, yet the ERP and finite scheduling systems available are reasonably generic, with different tweaks. Even inside one industry, each customer manages its operation in its own way: how they create their jobs, how they approach scheduling, how they track the shop floor, how they report their production KPIs. In the factories I've worked in, each one has a unique element that gives it its edge in how it brings a product to market, and that drives the efficiency and throughput that make the factory, the engine of the business, generate revenue and profit.
Standard systems are written towards best practice, which is a good starting point but rarely where a manufacturer's competitive position comes from. That usually comes from the way of operating it has built up over many years, and one of the most damaging things an implementation can do is make a business give that up to fit a standard solution. (We made the case for building beyond best practice in an earlier paper, Should I customise my ERP?.) The harder part is telling which of a plant's differences are its edge and which are only constraints, and that is difficult to do from a system's point of view alone.
It is also why, in my experience, some tailoring is always needed. Even where a packaged solution covers most of the ground, there are always elements to tailor for each customer: sometimes within the capabilities of the existing system, sometimes by building out functionality around it, and often by streamlining a business process along the way. I have yet to see an implementation where none was needed. The tailoring belongs around the things that give the plant its edge, with the standard left to handle what is only habit.
The question to ask: what in your current way of working will the solution keep, and why?
5. Challenge the requirement, not the customer
A fair question is whether a consultant should ever lead a customer back to best practice, and whether that oversteps. The way we typically approach it is to challenge the requirement. Every request gets the same question: is this the best thing for the business? Whether the answer turns out to be best practice or something particular to that business depends on the requirement. A lot of the time a customer has one thing in mind because they've read that it is best practice, or because it is the way they've always done it, and practically it may not make sense: because of the data, because of system constraints, or because of the time it would take operationally to keep it going.
One great example is finite scheduling. In environments with very complex bills of materials, long run times and complex job hierarchies, scheduling every resource finitely doesn't always make sense: the effort of keeping it accurate outweighs what it gives back. So we might direct the customer to a buffer, theory of constraints approach instead, scheduling the constraint areas rather than every resource. This is established practice, described in the ASCM (Association for Supply Chain Management) Supply Chain Dictionary as drum-buffer-rope: identify the constraint, schedule it, protect it with a time buffer, and tie the release of material to the pace of the constraint.⁸ Research on the method notes that the other resources usually don't need their own detailed schedule, because each operation is paced by how its order's buffer is consumed.⁹ Knowing that the option exists, and recognising the plant where it fits, is the kind of judgement that comes from experience and having scheduled a factory yourself.
Experience also lets someone who has run a plant do more than record a request and configure it: they can advise, from having seen what works and what doesn't across many operations, on how to approach the scheduling or the performance tracking in the first place.
The question to ask: when did your implementer last push back on something you asked for?
6. A factory serves the whole business
In every business there is a key person, often more than one, who bridges what happens in the factory with the rest of the business: making sure the schedule and the execution happen in a specific way, and balancing that against what sales and the wider business need. Working closely with those people is where the exact requirements become clear. One gap I often see is a focus on a single part: on scheduling alone, or factory efficiency alone, or on sales or procurement from the other side. At the end of the day the solution has to serve the entire value chain: the touch points to sales, to procurement, and to generating revenue and profit from what the factory executes. The planning discipline says the same: the APICS Dictionary defines sales and operations planning as bringing "all the plans for the business (sales, marketing, development, manufacturing, sourcing, and financial) into one integrated set of plans".¹⁰ In practice I cover it department by department: what would add value in each, operations in particular, and what that translates into for the business as a whole.
At Jumbo Brands, the business ran quite a basic ERP system, with no complex scheduling or shop floor data collection. What made the biggest difference was not a better system, but rather performance tracking: a few KPIs that were key to the business, tracked very specifically and in a very basic way, with daily reports that showed operational efficiency and how it translated into profit from a sales perspective. The link to sales was the important part, because it showed whether the efficiency achieved on the floor was actually generating money.
That is where I would tell most operations managers to start: identify the elements in your operation that really make the difference to how the business runs. It is not a list of 10 or 20 indicators. It is probably two or three specific ones that are critical to the success of the business. Focus on those, use a system to give you visibility of them, and you will start to see the difference as you drive improvement. The people on the floor need to understand those few numbers too. A 2024 study of shop floor management at a Slovak manufacturer found that operators answered "I don't know" in 58.33% of responses when asked whether the plant's KPIs suited their work (12 respondents), while team leaders and managers were largely satisfied; the researchers concluded that operators could not see the KPIs' connection to specific problems.¹¹ Kaplan and Norton built a balanced scorecard around a "handful" of critical measures, to keep managers out of information overload.¹² What the floor adds is the choice of which handful, and that choice is different in every plant.
The question to ask: who in your business decides what runs next, and were they in the room when the system was designed?
7. Fit for purpose, not the most features
The Jumbo Brands example also taught me that you can have the best possible system and still not apply it operationally. Sometimes more basic solutions are the most relevant ones. A fit-for-purpose way of tracking the data, with visibility of it and the means to measure it, is what drives efficiency.
Does that mean the system matters less? No. Tracking a KPI depends on having the data behind it, and being able to see it. Without that data you can't drive improvement or change, and the data has to come from somewhere and live somewhere. We believe that place is a well-designed system: one with the features your requirements call for, and the visibility to show you the data when you need it. But how much system? Enough to give visibility of the things that matter most to your operation, at a level of detail it can keep up, but not so much that keeping the system going becomes a job in itself.
8. Honest limits
None of this is a licence to impose a way of working on anyone. Operational experience is useful because it helps you tell the difference between a habit and an edge, and that cuts both ways. Sometimes the right answer is the standard one; where the standard is good enough, standard is right. Sometimes the customer's way of working is exactly what makes them competitive, and the system should protect it. And sometimes the request itself is what needs to change.
There is also a limit on what experience can see from outside. The people on your floor know things about your operation that no consultant will learn in a few site visits. An important part of the job is to draw that knowledge out and connect it to the system.
9. Conclusion: why experience on the floor matters
A consultant who has worked in operations sees the requirements nobody mentions, because they have worked under them. They know where theory and practice part ways in a plant, and which differences are worth protecting. They can challenge a request with the credibility of having done the job, and they keep the factory connected to sales and profit. Combined with sound systems knowledge, that is what turns an installation into an implementation.
If you are an owner or operations manager reading this, here are four things to do on Monday:
- Walk your floor on a normal shift, and write down what people do that no procedure mentions.
- Take your current list of system requirements and ask of each one: is this best for the business, or something we read about or have always done?
- Name the numbers that decide whether your factory makes money, usually two or three, and check that the people on the floor can say what they are.
- Ask whoever implements your system how they will find out what you aren't telling them.
10. How we work at Lancea Konsult
At Lancea we do business consulting, business process work and system implementation, and we bring the operational side with it. Our objective is to connect with a customer at a business level first, understanding what brings them value before a system is considered, and then to see the systems as a support layer to that, not the other way around. Our implementation method reflects this: its design stage begins with a business process review and a gap analysis, and identifies the solution requirements and best practices before a detailed design is drawn up.¹³
SYSPRO's manufacturing operations management struggled to land in South Africa for many years. When Lancea took it on, we approached it from operations, and SYSPRO now positions manufacturing operations as strategic. Its partner directory lists Lancea Konsult as a "specialist implementation partner for SYSPRO Manufacturing Operations Management (MOM)",¹⁴ and our implementations now reach the UK, the USA and Australia. Lancea is a SYSPRO Elite Solutions Partner, with 100+ SYSPRO deployments.
If you'd like a second view on where your operation and your system don't quite meet, from people who have worked on a factory floor, we'd be glad to have that conversation: contact us.
References
¹ De Cremer, S., Van Hootegem, G. and Brasseur, C. (2024). The same but different: Understanding variation and similarity in the outcomes of a similar technology. A comparative case study on the deployment of manufacturing execution systems in three Belgian SMEs. Information and Organization, 34(4), 100539.
² Leyh, C., Lorenz, A., Faruga, M. J. and Koller, L. (2025). ERP System Implementation Projects: An Update on Critical Success Factors. In Information Technology for Management: Intelligent Alignment of IT with Business and Society, Springer, pp. 38 to 57.
³ Institute of Industrial and Systems Engineers. Industrial and Systems Engineering Body of Knowledge: What is Industrial and Systems Engineering? IISE.
⁴ Hannola, L., Richter, A., Richter, S. and Stocker, A. (2018). Empowering production workers with digitally facilitated knowledge processes: a conceptual framework. International Journal of Production Research, 56(14), 4729 to 4743.
⁵ Panorama Consulting Solutions (2010). 2010 ERP Report. Panorama Consulting Solutions.
⁶ Panorama Consulting Solutions (2016). 2016 Report on ERP Systems and Enterprise Software. Panorama Consulting Solutions.
⁷ International Society of Automation. ISA-95 Standard: Enterprise-Control System Integration. ISA; also published as IEC 62264.
⁸ ASCM (2024). ASCM Supply Chain Dictionary, 18th edition. Entries: drum-buffer-rope (DBR), drum schedule, buffer, rope, five focusing steps. Association for Supply Chain Management.
⁹ Lizarralde-Aiastui, A., Apaolaza-Perez de Eulate, U. and Mediavilla-Guisasola, M. (2020). A Strategic Approach for Bottleneck Identification in Make-To-Order Environments: A Drum-Buffer-Rope Action Research Based Case Study. Journal of Industrial Engineering and Management, 13(1), 18 to 37.
¹⁰ APICS Dictionary, 14th edition, definition of sales and operations planning, as quoted in ASCM. Introduction to S&OP. ASCM.
¹¹ Mesarosova, J., Lestýnska Škurková, K. and Vizváry, Z. (2024). Digitalization of Shop Floor Management at Selected Segment in the Industrial Enterprise in the Context of Industry 4.0. MM Science Journal, November 2024. DOI 10.17973/MMSJ.2024_11_2024024.
¹² Kaplan, R. S. and Norton, D. P. (1992). The Balanced Scorecard: Measures That Drive Performance. Harvard Business Review, 70(1), 71 to 79.
¹³ Lancea Konsult. Implementation approach. lanceakonsult.com.
¹⁴ SYSPRO. Lancea Konsult, SYSPRO ERP Partner, South Africa. SYSPRO partner directory.
Running SYSPRO and want a second opinion?
Talk to the Lancea Konsult team about your environment.

