Most dealers and resellers are awash with valuable data that could make a significant difference to their bottom line, if used correctly, writes Kevin Falconer, GM at Channel Data.

Customer lists, sales data, market research, even complaints lodged by irate users, can be put to work to boost business and manage customer relationships more effectively. Unfortunately, this data is rarely analysed or put to productive use because many small business owners often neglect to apply core marketing principles to attract and retain clients.
This is where data mining is able to play a significant role. Data mining – sometimes referred to as data or knowledge discovery – is the analysis of corporate data from a number of perspectives with the goal of refining it into information that can be used to increase turnover and cut operating costs.
Sophisticated data mining software is able to summarise and categorise data while finding correlations or patterns among many hundreds of fields in large relational databases.
Data mining is not a new science. It has been around for many years. However, new developments in computer processing power and disk storage solutions are sharpening the accuracy of data analysis and reducing the costs associated with these systems.
Today, even the smallest computer reseller can benefit greatly from data mining technology. The knowledge gained, if used effectively, is capable of putting the humblest ‘mom-and-pop’ store ahead of the big warehouse chains when it comes to customer service and retention.
Data mining consists of four key phases:
* The extraction, transformation and loading of data onto the system;
* The storage and management of data in a multi-dimensional database;
* The analysis of the data by application software; and
* Presentation of the data as knowledge in an easily understandable format.
To accommodate these requirements, all that is needed is access to the data storage repositories within the business. These contain operational or transactional data – such as sales, cost, stock holdings, payroll information and accounting – and non-operational data, such as industry sales figures, forecast data and macro economic data.
Once analysed and the patterns, associations or relationships defined, the data should be available as useful information. For example, the analysis of point-of-sale transaction data can yield important information on which products are selling fastest and when the most sales are logged.
As more and more data is mined and converted into knowledge historical patterns will crystallise and trend lines will become evident. Judicious application of this information can assist with the long-term management of customer relationships.
In many businesses the Pareto principle – the 80/20 rule – applies. This often prompts sales staff to emphasise service to the majority of customers (the 80%) who, while making the most demands, generate the least profit – while neglecting the needs of the small percentage (the 20%) of customers who are the most profitable.
It is therefore vitally important to be able to recognise profitable customers and identify the characteristics that define them. When changes in their purchasing behaviour are noted, this can be either present an opportunity to the business to follow the trend, of a threat that will need to be promptly addressed.
The acceptance of data mining as a technology is growing rapidly, evolving to the extent that businesses so equipped will be able to counter the marketing activities of opposition organisations with effective, targeted campaigns of their own. Accurate, up-to-date customer information will give sales staff many cross-sell, up-sell and repeat-sales opportunities.
Data mining, integrated into modern customer relationship management solutions are already enabling IT reseller businesses to grow the ‘lifetime value’ of their customers and so assume leadership positions in the IT industry – whether their target market is individual end users or corporate clients.