fact-based management Clean E Early in our business, we were asked by a leading regional hospital to conduct a study of the cleaning quality in their facility. While there were many subjective opin-ions about the condition of the facility, there was a notable absence of any hard data. It soon became clear that agreement about cleaning success was determined by the level of agreement about what “clean” means. Our first step was to figure out how big a problem cleaning was and to understand how many of the complaints documented were actually cleaning related. Based on a study of complaints, it turned out that 22 percent of all complaints reviewed were caused by maintenance defi-ciencies for painting and repair. Of all the remaining complaints, the con-dition of private and public restrooms was the largest concern. We went on to interview building man-agement, occupants and staff about their specific concerns in restrooms. While we reached some worthwhile con-clusions during the course of this project, we were struck by the overwhelming absence of any common criteria or condition voiced by respondents as a basis for their opinion about cleanliness. In short, nearly all of the information avail-able was anecdotal. There had to be a better way of defining the conditions and attributes that was more fact-based. Defining By: Vince Elliott Nine customer-driven attributes can essentially define clean. clean or dirty to building occupants, rather than any bacteria count or other scientific measure. We used both interview and complaint record reviews — two years worth of infor-mation — as a basis for collecting a data-base that was item and condition specific. In all, we identified 80 unique conditions from hundreds of comments. For example, comments about dusty chairs, dusty sofas or dusty benches were consolidated into one of the 80 conditions as “dusty furniture.” Using spreadsheets, we recorded this list in a table format on a large clipboard with items and surfaces on one tab and condi-tions and attributes on the other. Industry professionals recorded the appearance problems they found on each item and surface in each room. The result was a massive data set of every condition found on every item in the building inspected. The body of work provided a research-based definition of “clean” and, in the research, we learned a few key lessons. must be objectively provable by visible confirmation. This minimizes the subjective feeling that something is good or bad. We learned that rating scales, like one to 10 or poor to excellent, do not lend themselves to objective measurement or management. Thus, the definition of “clean” becomes a visual test on the provable attributes or con-ditions present or absent on items subject to cleaning. In short, we learned that, in order to improve quality, you must manage it; in order to manage it, you must measure it; and, in order to measure it, you must define it by customer-driven attributes. What we found, using Pareto analysis of the more than 4,600 respondents, was a clear picture of the attributes associated with each item found in the various types of space in an office environment. The resulting frequency distribution showed that nine cleaning conditions accounted for 94.6 percent of all conditions found by the inspection team. Nonetheless, the other 71 conditions rep-resented only 5.4 percent of all conditions identified by anyone. Given the broad participation in the study, these findings seem to establish a valid, objective definition of the attributes that determine if an item, room or building is “clean,” as defined by building occupants, managers and cleaners. CM Vincent F. Elliott is the founder, president and chief executive officer (CEO) of Elliott Affiliates Ltd. of Hunt Valley, Maryland. For more information, visit www.ealtd.com. He is widely recognized as the leading authority in the design and utilization of best practice, performance-driven techniques for janitorial outsourcing and ongoing management. www.cmmonline.com Things We Learned Since that initial study, we have repeated this survey of client occupants in a nominal group setting up to and including this year. While there have been changes in the technology that allows information to be collected, analyzed and shared, one thing we learned is that the descriptive attributes must connect customer expectations to process activities. Further, we learned that any attributes agreed upon must be easily understood and identified by everyone connected by the cleaning system. We learned that any attributes adopted Early Methodology With this challenge in mind, we set out to conduct a study of the conditions and attributes that might be used to describe something as “clean” or “dirty.” The system of complaints was based on appearances — that is, something looked 45
CMM – Archives Aug 2011: 45
