Picture a sales rep sitting down at the start of the day with a stack of customer cards on the desk. Who gets a phone call today? Who gets a visit? Who gets a fax? Without a clear system, the answer usually comes down to habit or gut feeling. The rep calls whoever comes to mind first, visits whoever is on the usual route, and the rest wait. Large buyers sometimes go weeks without contact while small but steady customers are quietly forgotten. CRM (customer relationship management) software addresses this problem directly through a method called customer segmentation.
Segmentation means dividing customers into groups based on defined criteria. Three main criteria drive the process: value, behavior, and potential. Value captures how much a customer has purchased over time. Behavior covers how often they order, whether they pay on time, and whether they tend to raise complaints. Potential is a judgment about whether that customer could realistically buy more in the future. A CRM program pulls these three data points together and assigns each customer to a group automatically. When the sales rep opens the program in the morning, the list already shows who belongs in the top tier and who sits further down.
Doing this by hand is theoretically possible but practically exhausting. Sorting through hundreds of customer records in a spreadsheet takes hours, and every rep tends to apply different standards. One person weighs only the last six months of revenue; another goes by personal rapport. CRM software standardizes the process so that every customer is evaluated against the same rules and every team member sees the same result. When a rep is on leave, the colleague covering that territory can look at the screen and immediately understand which customers need priority attention.
Once the groups are defined, the real work begins: designing a different service model for each one. Top-tier customers, those with high purchase volume and consistent order history, typically warrant a phone call every week and a face-to-face visit at least once a month. They get first access to new products and receive individually prepared price offers. Mid-tier customers are contacted monthly, with relevant offers sent by fax. Lower-tier customers receive periodic group communications and occasional check-in calls. Before segmentation, a sales team tends to spread its time more or less evenly across all accounts. After segmentation, time spent on top-tier customers increases sharply while effort wasted on low-value accounts falls away.
A practical example helps make this concrete. A mid-size wholesale food distributor had a sales team following the same weekly route, visiting small corner shops and larger retail accounts with equal frequency. After running a segmentation exercise inside the CRM program, the twenty accounts generating the highest revenue were placed in a separate list. One dedicated rep was assigned to this group, visit frequency went up, and the remaining accounts shifted to phone and fax follow-up. Within a few months, order volume from those twenty accounts grew noticeably. The team was doing more with the same number of people.
There are real difficulties in making this work. Feeding accurate and complete customer data into the CRM program takes sustained effort. If records are incomplete or out of date, the segmentation produces misleading results. Some sales reps also resist the change; a person who has worked by instinct for years finds it strange to have a program tell them which customer deserves attention today. Another practical constraint in this period is that most CRM software arrives in a foreign language, and options with a Turkish-language interface and local support are limited. Installation and training almost always require a local authorized reseller, and that cost needs to be factored in from the start. Buying the software without budgeting for proper setup rarely leads anywhere useful.
If you want your sales team to use its time better, start by asking how solid your customer data actually is. How many of your accounts have a complete purchase history from the last two years on record? Are payment patterns being tracked? If the data is reasonably clean, setting up segmentation in a CRM program is a manageable project. If the data is scattered across notebooks and loose files, organizing it comes first, and that step can take longer than selecting the software itself. Before opening any program, sit down with your team and work through one question together: how much time should each type of customer actually receive? Agreeing on that answer before touching the software is the most important step of all.
This article was originally written in Turkish by Gökhan MERCANOĞLU on February 25, 2002 and has been automatically translated into English and other languages using machine translation.