If you are searching for the root cause of a failed ERP project in the software itself, you are most likely looking in the wrong place. The pattern across companies that end up disappointed after go-live is remarkably consistent: the software selected was not wrong, the configuration was not fundamentally flawed, and the budget was often not even overrun. The failure lives somewhere quieter — in the project manager consultant’s inability to fully own the bridge role between the business and the system. A consultant with solid technical knowledge but without a genuine understanding of how the company actually operates will, almost inevitably, push the organisation to reshape its own workflows around the software. At that point, it is not the problem being resolved. It is the company’s operational logic being dismantled.Consider an anonymised case from a mid-sized construction and real estate company based in Ankara — 272 employees, residential projects running simultaneously across four cities, and accounting, procurement, and site operations units that had worked in functional silos for years. The general manager made the ERP decision in late 2013, partly driven by the newly enforced e-Invoice and e-Ledger obligations under Turkish Revenue Administration requirements. The software selection was finalised within three months. The consulting firm deployed a technically capable team: servers were set up, modules were configured, users went through training. Seven months after go-live, a significant share of users had quietly returned to their Excel files and email chains. The project was technically complete. Operationally, nothing had changed. The consultant’s project plan had been built to install the system — not to understand the company.The project manager consultant’s function extends well beyond technical integration. The questions that must be answered in the first three weeks are these: Which unit makes which decision, and whose approval does it actually require? How does a given process work in practice, versus how it appears on an org chart? Where does user resistance come from — technical unfamiliarity, natural resistance to habit change, or a genuine signal from the field that the process and the software do not fit? Without documenting the answers to these questions, every configuration decision becomes a liability that resurfaces in later phases. In the Ankara construction company’s case, procurement authorisation had in practice been delegated from the general manager to individual project managers — but this real distribution of authority had never been asked about. The system was configured around centralised approval. The result: every purchase request stalled in the queue, and project managers fell back on email.The requirements analysis phase is the most underestimated and most time-efficient segment of any ERP project. An experienced project manager consultant does not move through it in two or three meetings. She extracts real workflows, exceptions, and informal practices from each unit — and that requires fieldwork. The answer to ‘who actually uses this system and how’ does not live in the meeting room. It lives in the warehouse, on the construction site, and on the accountant’s screen. Had this analysis been conducted properly in the Ankara case, the distributed procurement authority would have been visible within the first week, and configuration would have reflected it. The cost: roughly two additional weeks of analysis. The benefit: avoiding seven months of operational chaos. Viewed through a TCO (total cost of ownership) lens, the value of those two weeks is beyond argument.Another critical responsibility of the project manager consultant is grounding management expectations before a single module goes live. In mid-sized Turkish companies, ERP decisions typically originate from the general manager’s desk, and a set of expectations accumulates rapidly around the decision: once the system is live, reports will flow in real time, inventory problems will disappear, monthly financial closes will accelerate. Some of these outcomes do materialise — but none of them happen automatically and none of them happen on go-live day. If the consultant does not manage this expectation explicitly and in writing before the project begins, the pressure of ‘the system is not delivering what we expected’ arrives reliably by the third month. The project is then declared a failure by management even when it is a technical success. Owning this communication is not peripheral to the consultant’s role. It is central. Some consultants avoid the conversation because it is uncomfortable. Not having it leads to a far heavier reckoning later.The question of user habits also deserves candour. In the Turkish SME environment, training is typically delivered once, organised by module, and left for users to connect to their daily routines on their own. A project manager consultant should design training not as a one-off technical session but as a workflow integration process. In practice, this means conducting field observations after the first week, watching where users get stuck, and running at least two on-site coaching sessions within the first thirty days of live operation. In the Ankara company, the majority of the accounting team’s difficulties with the e-Ledger module were not technical. The module’s working logic differed from the sequential period-close routine the accountants had used for years, and that difference had never been explained. One hour of side-by-side working would have prevented two months of support tickets.For any manager evaluating a project manager consultant before signing an ERP contract, three practical decision criteria apply. First: how much time does the consultant allocate to requirements analysis? If ‘we understand your needs’ is declared after two meetings, that is a clear warning sign. Second: does the consultant formalise management expectations in writing, or does every request receive a reflexive ‘yes, we can do that’? Third: is post-go-live support structured with defined check-ins and field observation, or is it reduced to ‘call us if something breaks’? None of these answers appear in a software demonstration. Yet they are precisely the variables that determine whether the project delivers operational value or only technical completion. Choosing the right ERP matters. But the quality of the person embedding that software into the organisation is a more decisive variable than any feature in the product catalogue.
This article was originally published in Turkish by Gökhan MERCANOĞLU on February 1, 2014. The English edition has been reviewed and edited by the author.