The company that spends most on CRM is often the one that gains least from it. I saw this paradox clearly at a dealer meeting for a building materials manufacturer: colourful dashboards on the head office screen, and on the ground, customer records that practically no field representative had filled in properly. The software vendor called the project a success; the manufacturer’s management could not understand why the system was not working. The real issue was straightforward: dealers and field staff were not entering data because they knew the system produced no value for them personally. Deploying CRM across a dealer network is not a software project — it is a process design challenge first and foremost, and every implementation that misses this distinction stalls somewhere between the kickoff presentation and live operation.The building materials sector in Turkey carries a distinctive distribution structure. Between the manufacturer and the end user — a contractor, a subcontractor, or an individual homeowner — there are usually two or three intermediary layers: the regional dealer, the local hardware supplier, and sometimes a specialist applicator firm. Consider a construction chemicals manufacturer based in Eskişehir, operating with 280 employees and 34 active dealers spread across Turkey — a scale entirely typical of Turkish manufacturing SMEs. The company deployed a CRM package to centralise dealer sales data, track repeat customers, and support better seasonal stock planning. Technical installation was completed in eight weeks; training sessions were held. Four months later, only 7 of the 34 dealers were actively entering data into the system. Head office responded by saying ‘the field needs more training.’Are dealers actually resisting, or is something else happening? Asking the question properly reveals that both sides have a point. Management is right: without dealer-level sales data, production planning and pricing decisions become guesswork. Knowing which product categories are turning quickly in which region, and which customer profiles are reordering, is essential for central decision-making. Yet dealers are equally right: a dealer processing 40 to 60 transactions daily, handling most of the turnover by phone order, running accounting on a separate package, and receiving nothing in return for CRM data entry sees the whole exercise as pointless administrative burden. Threatening to link bonus tiers to entry compliance creates a short burst of activity but produces no lasting behavioural change. Data discipline is built through value loops, not through pressure.In Turkey’s building materials sector, this tension has a very specific dimension: the relationship between a dealer and a contractor is usually personal and informal. The contractor knows the dealer; pricing is discussed on the phone; a proforma invoice may or may not be requested. Under these conditions, CRM becomes meaningful only when the dealer can look at a customer history and say ‘this contractor bought four tonnes of insulation material last spring — they will probably come back this winter’ or ‘they had a late payment last season, I should reach out early this time.’ In other words, CRM creates value when it helps the dealer manage their own customers better. When it is designed purely as a data repository for head office reporting, it remains empty.The most significant change in the second phase of the Eskişehir project was this: an ‘open orders’ summary and a ‘customer movements in the past 90 days’ panel were added to each dealer’s CRM home screen. Dealers now saw something relevant to their own business the moment they logged in. A second adjustment was sending field sales representatives a weekly regional report generated from the data they themselves had entered — creating a visible loop between input and output. Once that loop was in place, the active data entry rate climbed to 57 per cent within four months. Not perfect, but a functional starting point. I tested a comparable process adjustment at roughly the same time across a network of ceramic tile dealers operating in the Izmir region. There, the critical difference was that the dealer profiles were younger and more comfortable with computers: the entry rate reached 72 per cent within 14 weeks. Different conditions produce genuinely different results, and applying a single prescription to every network is the most common mistake I see on these projects.There is one clear condition under which this approach fails entirely. When a dealer network is highly fragmented — consisting of small five-or-six-person depot operations — the cost side of CRM quickly outweighs the process benefit. In that type of network, accounting runs on a separate system, orders arrive by phone, and the customer list lives in the owner’s head or in a personal notebook. There is no integration point, training capacity is minimal, and the dealer already knows every customer by name. Installing CRM in this environment does not solve a problem; it introduces a new one. The right question is not ‘which CRM product fits our sector’ but rather: does the average scale and operational maturity of our dealer network actually have the capacity to absorb the discipline that CRM requires?If you are planning to deploy CRM across a building materials dealer and end-user network, answer three questions before touching the software shortlist. First: what will head office do with the data it receives, and how will any benefit flow back visibly to the dealer? Second: when a dealer logs in, what will they see that is useful for running their own business that day? Third: which three or four core processes — order taking, customer history, payment tracking — are currently operating without CRM, and exactly where will the system connect to them? Without satisfying answers to all three, moving straight to software selection starts a cycle that disappoints both the dealer network and the central team. In my experience, the difference in outcomes between a project that spends eight weeks on process analysis before selecting software and one that goes directly to installation is not marginal — it is the difference between a system people use and one people avoid.
This article was originally published in Turkish by Gökhan MERCANOĞLU on January 23, 2007. The English edition has been reviewed and edited by the author.