In this challenging climate, Ian Tomlinson of EPoS and e-commerce solution provider Cybertill argues that retailers should pay more attention to setting and using meaningful key performance indicators (KPIs).

As with so many things in life, the answer to every problem is closer at hand than you think. As retailers feel the pain of an economic slowdown and flagging consumer confidence, they tend to look outside their business for remedies and strategies. However, they would do better to look inside the business, measuring its performance and planning its future using intelligence that is readily available.

Effective business management relies on good information. It goes without saying that retailers need to monitor their exposure to risk, their cash flow and their profitability. However, the most valuable information often does not sit in the P & L accounts; it derives from measuring critical indicators. These indicators are as individual to the business as its DNA.

Too many managers monitor financial performance and expect to be able to make changes based upon what they find. Invariably, they are shutting the door after the horse has bolted. Financial data records the result of strategies and tactics, but leaves little room for manoeuvre after the event.

A more reliable approach is to define simple indicators that proactively take the temperature of a business. For an online retailer, indicators could be based upon visitor traffic by time of day and duration, highlighting the pages visited and the products that provoke enquiries and generate sales. A stores-based retailer could measure footfall, by location, by time of day and by length of visit.

In our business, selling e-commerce and EPoS solutions to retailers of all sizes, you could argue that 'sales made' is a key performance indicator. Whilst this is clearly of critical importance, I look elsewhere for my business intelligence.

A plasma display in my office shows me in real time how our telemarketing consultants are performing: the number of calls each makes, the total time being spent on the 'phone each day, the number of appointments made and, from another set of KPIs, the demonstrations that result from those calls. Throughout the day, I watch this data coming across the screen and am in time to spot trends, ask questions, sound alarm bells and make changes. Waiting for sales results at the end of the month could just be too late.

Every business is affected by scores of internal and external factors that enhance or diminish its performance. The sheer volume of information available can be distracting so it's important to focus on just a few key indicators which reflect performance, are measurable, are comparable against benchmarks such as a previous year's data or a competitor's results, and which indicate a course of action.

So how do you identify your KPIs? Whilst best practice suggests that you ask yourself three questions: what drives your sales figures, your costs and your cash-flow, you should always come back to what drives your own business. Whilst there are retail constants such as sales per square foot of store space, accept that drivers can vary enormously even within the same sector, and that they will need to change with time.

Analyse what would enable you to outperform your competitors and consider having these elements as KPIs. Understand the costs of each step in your supply chain. Investigate the effect that training and staff turnover have on your sales, and measure productivity by sales person against experience and length of service. Good stock management means having the right stock available at the right time in the right location. It also enables you to release cash by 'turning' stock. Set KPIs around product defect/return ratios and calculate how returns are eroding your bottom line.

When it comes to presenting KPIs, make sure that your chosen method highlights trends. You can do this by simple means as I do with my telemarketing activity screen, or you can invest in business intelligence software. These systems are particularly good at crunching data and providing 'traffic light' performance displays and exception alerts.