The ERP project is done. Stock levels appear on screen, accounting closures run on time, purchase approvals no longer go through the fax machine. Management is satisfied. Then someone asks: ‘When are we going to sort out the customer side?’ At that moment, despite two years spent inside the company, you feel like you are starting a completely new project from zero. My argument is this: ERP tells you what happens inside a company, but CRM has to tell you what happens to the company’s customers — and most companies have never written that down, never measured it, and in some cases never even thought about it. Swapping the software is the easy part. Understanding the customer process in the field is the actual job.An ERP program looks inward. Where is the stock, where is the money, which order is at which stage — those are ERP questions. A CRM program looks outward. Who is the customer, what did they buy, when did they call, was their complaint resolved, when will they buy again. This distinction sounds simple but means something quite different on the ground. I saw this gap very clearly at a wholesale automotive spare parts distributor in Ankara, a company with 284 employees covering seven distribution regions across Turkey. We had finished the ERP installation; the stock module was working well. Then we started a preliminary analysis for CRM. ‘How do you keep track of who your customers are?’ we asked. Answer: ‘Accounting checks the card.’ ‘Where do you record complaints?’ Answer: ‘The service manager takes notes.’ ‘Which dealers have not placed an order in six months?’ Answer: silence. The ERP knew everything — just not about the customers. That is where the real project begins.The worst thing a consultant can do at that moment is sit down at a computer and start configuring software. First, you have to go into the field. At the Ankara company, I spent three days with the sales team. I talked to dealers who came in the morning, listened to phone traffic in the afternoon, and reviewed service forms in the evening. What I found was this: the sales team did not keep records because they ‘knew’ every customer personally. ‘Ahmet Bey knows us, he will call when he needs something’ — that was the operating logic. For a small shop this might work. For a wholesaler with 312 active dealer accounts spread across the country, it is not sustainable. Staff change. Ahmet Bey retires. The new employee does not know that customer. The customer relationship is stored in a person, not in a system. The real problem with the CRM project was not choosing the software — it was making the team recognise this habit for what it was.Done properly, a needs analysis for CRM takes at least as long as the software installation itself. When I say this, managers are usually surprised. ‘You installed the program, is it not finished?’ No, it is not. Before touching the software, you need answers to questions like these: How many different employees interact with a single customer? Do orders, complaints, and collections each pass through different people? Do those people talk to each other? If a complaint is taken by phone and resolved verbally, does it ever get written down? At the Ankara company, the sales manager and the finance manager had two entirely different views of the same customer. The person the sales manager called ‘our best customer’ was sitting at the top of the finance manager’s ‘slowest payer’ list. CRM has to bring those two views onto one screen — but first you have to understand where each side is keeping its data and why.On the software selection question, the options available in Turkey in 2004 were limited. Larger corporations were looking at expensive foreign packages like Siebel, but that budget was not realistic for a mid-sized wholesale distributor. Local software firms had started adding customer tracking modules as extensions to their ERP programs, and some companies were beginning to look at the CRM solution Microsoft had recently brought to market. With the Ankara company, we chose a different starting point: before installing any separate CRM tool, we first reorganised the customer card and order history structure that already existed inside their ERP program. The reason was straightforward. Moving to a new system with old, messy customer data is like painting over rotten wood — the problem stays hidden for a while and then reappears. Cleaning the data took six weeks. Nobody had planned for those six weeks, but they could not be skipped.The gap between what management expects and what the field actually looks like is the thing that puts the most pressure on CRM projects. The manager says ‘install the program, I want a report on my screen every morning.’ The person on the floor says ‘I already know everything, why would I write it down?’ Until that tension is resolved, no software holds. The solution we used at the Ankara company was simple: a single one-page form to be filled in after every sales contact. The form worked on paper just as well as on a computer screen. Once a week, someone entered these forms into the system. Why such a basic approach? Because the team had not yet made software part of their daily routine. Forcing it creates resistance; gradual introduction builds acceptance. Six months in, sixty-eight percent of the sales team was filling in the form regularly — a solid enough base to move to the next phase.Moving from ERP consulting into CRM work is less about technical knowledge and more about patience and observation. Learning a software package takes a few weeks. Understanding a customer process takes months. I still remember that silence in the Ankara company when nobody could answer the question about which dealers had not ordered in six months. The data to answer it was already there — in the accounting cards, in the sales records. Nobody had ever asked the question. The real value of a CRM project does not start on the day the software goes live. It starts the moment that question is asked for the first time and can actually be answered. For any company that has finished its ERP and is thinking about CRM, the first step is not to look at software. It is to ask honestly: do we know the last time we lost a customer, and do we know why?
This article was originally published in Turkish by Gökhan MERCANOĞLU on January 20, 2004. The English edition has been reviewed and edited by the author.