Details Matter: Little Data Trumps Big Data

Have you ever gone into a Starbucks during a busy time and noticed how the staff shifts roles to handle variation in customer volume? When you arrived, there may have been three people in line served by one cashier, two baristas, and a fourth staff member restocking the milk and sugar. When two more customers walk in, the person stocking moves to the second cash register to take orders. Add two more customers and one of the baristas begins taking orders, writing them on cups for her colleague at the espresso machine.

The number of team members in the store is static even as the customer demand rises, yet the speed with which customers experience the first stage of service stays relatively consistent, because the staff members use status data to adjust their processes real-time.

Such fluid adjustment is only possible because the in-store Starbucks team has been trained to observe their processes real-time and make workforce allocation decisions according to defined triggers. Starbucks leaders are never caught off guard about the customer experience because they actively collect—visually, in their case—and use real-time status data to perform more consistently.

Status data reflects how far along a work team is on completing work-in-process (WIP) and whether work queues are shrinking or growing. In my first post in this three-post series about little data trumping big data, I wrote about the importance of accurate data on the customer demand a department, business unit, or entire organization has on an hourly, daily, weekly, or monthly basis. Status data, in contrast, reflects the progress made on addressing that demand through a disciplined process.

Status data is important—especially when it’s visually displayed—because it tells team members when they need to reallocate resources or the work itself to meet customer demand in the requisite time. It also tells leaders how teams are performing and whether priorities, maximum queue levels, or the way work is organized need to shift. Real-time adjustments may be necessary to accommodate process slowing, unexpected demand, or unfilled capacity.

For example, teams may handle work and allocate resources one way when there is one customer to serve, and another way when there are 20. Teams need clear status data in order to make those adjustments, and not all organizations take steps to collect and display it.

Case in point: an engineering client we worked with had no system for tracking how many active projects they had, how far along the project teams were on any of them, and what delivery commitments the firm had made to its customers. As a result, leaders often learned they were late delivering product only when customers began to complain. The engineers would then tax their work teams and budgets by requiring overtime and paying premiums for expedited testing. The added pressure resulted in costly errors, higher turnover and more delays due to poorer output.

A more proactive, lower cost method is to track and visually display progress in real-time. Status data should include the stage various work products are in. Having these details leads to lower costs, faster delivery, and higher quality work, since everyone is clear about work status and has the ability to make adjustments and have proactive conversations based on that knowledge.

It can be done. A medical device manufacturing client of ours experienced a “whip” effect with its sales volume, whereby the company processed 25% more orders on the last week of the month than on weeks 1-3. The last week of the quarter had three times the order volume of any of the previous months, and the last week of the fiscal year was higher still. In response, we showed the team how to collect and visually display its demand data on incoming orders, and match that with corresponding data, updated daily, on the team’s status fulfilling that demand. We then helped design a more fluid work system that adjusted based on the volume of work-in-process.

In non-quarter-end weeks, the client’s regular team could process orders within the required time and quality standards.  When orders began ramping up in the weeks leading to quarter-end, the company was able to use status data to redeploy support staff from another department who’d been trained to process orders. At year-end—when status data began to show a growing queue, the organization added well-trained temp staff to the full-time and support resources.

These adjustments are possible when an organization is acutely aware of real-time work status—that and a fresh cup of coffee allow leaders sleep more easily and team members to perform at their best.