Research Data Design Inc. has applied concepts from Lean Manufacturing to its call-center operations. Through regular, hands-on interaction with performance data, employees companywide have developed an understanding of how they impact the bottom line and have become actively engaged in boosting corporate profitability.

BPM Magazine: Could you describe your management reporting process for me? What data's reported, who gets it, and when?

John Stepleton: Sure. We have three call centers; they're responsible for reporting their metrics each day. We produce a daily income statement, and we track our EBITDA performance on a daily basis. So we track what we did for the day prior and where we are month-to-date. And we compare that to our projected goals. We meet at 10:06 every day. In that meeting each location reports what revenue they created.

BPM: Where do you get the EBITDA numbers?

Stepleton: In a typical year we do about 1,400 projects, and each one of them has a unique price point because each project we do is a custom project for our clients. So when we get the contract, we track what the rate is per completed survey. We have a SQL database that tracks which project each completed call is for and, therefore, how much revenue it generates. We were really frustrated with the time-and-attendance packages that were out there, so we ended up writing our own time-and-attendance software that feeds into our SQL database. As our part-time staff come in and make calls, the amount that we pay them is logged in real time. So in the morning we report, by office, gross revenues; our variable costs, which is all of our variable labor, all of our incentives (we have a fairly sophisticated pay-for-performance package for our front-line employees); and all of our fixed expenses that we know about for the month. And then we produce the EBITDA.

BPM: How do you use the daily EBITDA to improve decision-making?

Stepleton: We also track labor as a percentage of revenue. For us, that is the one number that most affects our profitability. It shows whether we're efficient at what we do. We then can drill that down into some other metrics. We can look at revenue per hour on a specific job. And our systems are designed so that we can pull up a report and look at every project that was worked on the previous day and then drill down into the details. So we can pull up person-by-person productivity on a specific job at a specific location.

This morning I'm working on a customer project we've been doing for years. It has been highlighted as the job we want to focus on this month because it's the poorest-performing project of its type. So each day I'm able to go in and produce a report which shows how many surveys we've completed and some of the key metrics related to the research, but then I can look at employee-by-employee production. I can look at how many hours they worked, how many completed surveys, how much revenue they generated on that project that day, and what their relative revenue per hour is. For example, we had a range yesterday from $9.29 an hour to as high as $52 an hour. So we can understand what the $52-an-hour person's doing well and apply that to the person averaging $9 an hour in revenue. Having these systems in place has allowed us to be able to get to this level of detail in managing.

BPM: Does it also affect which projects you take on, how much you charge, that type of thing?

Stepleton: Absolutely. This year a big initiative for us was to tie our scheduling function to our pricing. Before, we would just charge a standard price; there was very little adjusting the price considering how full or empty we were. Having a really solid forecasting tool, now we can adjust our pricing. So another report that we look at daily represents what our pipeline looks like, how many completed hours we have month-to-date, how many hours are still on the schedule for the rest of the month, what our budgeted revenue target is, and then what the gap is. If we're not to our minimum budget level, that tells the sales department to be more aggressive about filling that time. We also then can predict what our future EBITDA is for the month, so we take into consideration this pipeline, what our production metrics are for the month, and then have a projected EBITDA.

BPM: What are some of your production metrics?

Stepleton: We look at utilization rates for our call centers. Our main productivity measure is revenue per hour. And labor as a percentage of revenue. Those are our three. We were taught through the MIT "Birthing of Giants" program that you need to have three or four key metrics to really manage. If you have more than that, there's too much to look at and it's too difficult to effect change.

BPM: What was your process for determining which metrics to focus on?

Stepleton: Very painful. It took us a year and a half to really understand which metrics to focus on. One of the challenges is to avoid picking metrics that don't have behaviors attached to them -- behaviors that we can affect.

In my first year in the "Birthing of Giants" program, the discussion was: You need to create a rhythm in your organization; meet daily. Rhythm is an important piece of it. Because it's one thing to have a number, but if you don't have a way to communicate that up and down the organization, it doesn't do much good. So we started to do the rhythm and we started to just pick different metrics. Initially we tended to pick things that were very broad and vague that didn't tie to specific behaviors, just top-line revenue or profit. But there are too many other things that can affect those numbers. When we tried revenue per hour of labor, we realized that metric would help us understand how we're pricing the work. And then because labor is our big variable cost, labor as a percentage of revenue helps us understand our efficiency. The metrics have changed from time to time, but I think the ones that we use now are really the ones that affect the success or failure of the business.

We still may add some new metrics. The pipeline number -- showing the gap in terms of our hours on the budget vs. hours sold -- is one we added just this week to our daily huddle. We may add a metric for a period of time until it becomes a nonissue and then take it away and add another one. So it's the 10th of the month, and I can tell you precisely, as of yesterday, what our hours gap is. We have 5,300 hours, which is about 19 percent of our total budgeted hours, that are missing. That drives behavior in the sales and account management functions, and since we report it daily, everyone can see the effect of the behaviors. The gap may have been 6,000 hours two days ago, and now it's 5,300 hours. As we start to book the work, the whole company can see that improvement.

BPM: Who's included in the daily huddle?

Stepleton: Our main huddle is all of the management staff, so it includes people from accounting, HR, IT, sales, account management, and our technical staff.

At first I didn't believe that it could work, but it actually does work. The format has changed significantly over the years. When we first started, we would go through each person reporting what they were working on that day. We then as a company would do our critical numbers, and then we would report what was stuck. We have developed a more sophisticated accountability today. On Mondays we each report what our top three initiatives are for the week, the three things that are going to help each individual reach their quarterly goals, which then roll up to the corporate goals. On our intranet, every full-time employee has a page which identifies their quarterly goals and their weekly top-threes. Then on Friday we track what people's progress is to attaining their weekly goals.

One of the beautiful things about the system is that it's completely transparent. Anyone in the system can look at anybody else's goals. They can go to my page and look at my goals; there's absolutely no hiding, from the top of the organization to the bottom. It's some pretty brutal accountability. It's not really designed as a punitive measure; it's really a culture of helping each other become successful. We know that we all have got to get this extra stuff done in order to improve the company, and we need each other to hold us accountable.

BPM: How have employees reacted to this?

Stepleton: We've been reporting the financial metrics for six years now; we started in 2000. The initial reaction was, "You didn't hire me to do this. I don't want to know this. This is too much information." It really freaked people out; they felt that it wasn't their problem. But the reality is that it is their problem, ultimately. It took probably six months to a year for people to get comfortable hearing that information. Because it's not always positive, and that makes people nervous.

Three and a half years ago, we went through a very difficult time as an organization, financially. We ended up having to do layoffs. If it wasn't for the fact that we were reporting this information, I think people would have been completely caught off-guard. But instead, people understood. They had seen the numbers for months. It was almost as if they said, "What took you so long?" Now, after six years of doing this, I think that the organizational culture has a dependency -- that's a scary word, but maybe it's the right one -- a dependency on the metrics. When we don't have the numbers, we feel uncomfortable as an organization.

BPM: What software do you use to support all this data analysis?

Stepleton: It's mostly homegrown. The advantage is that we can customize it as we see fit. When we want to make a tweak, it's a heck of a lot easier than working with an outside vendor. But developing our forecasting and performance management systems has been painful. Starting in 2000, we automated a tremendous amount, but we're now pulling back and wanting some functions to be more manual.

For example, we have a software system that gathers data on our costs, our EBITDA numbers, and labor as a percentage of revenue -- all the key things we want to know. But instead of printing out reports for our daily huddle, now we have a whiteboard where we literally put the numbers up with dry erase ink. Something about interacting with the data changes how you interpret it. And it definitely has had an impact.

I used to be enamored with the idea of dashboards, and we have some cool tools that don't get used anymore. When our Lean Manufacturing consultant came through, he said, "You've got to stop doing so much in an automated format. Files on computers don't get looked at. If you write it on a wall, people walk by it six times a day." So we use colored markers. If the numbers are above where they should be, they're green. If they're below, they're red. If they're on target, they're black. It's easy to glance at the board and see which office is performing well and which one's not performing well.

BPM: Would you advise that a company develop software in-house? Do you think it's worth it?

Stepleton: It can be. There's not a lot of software that's designed for a $10 million company. Given that fact, it was almost a requirement for us. Regardless of whether you're going to buy software or build it, though, you should first take time to do it manually -- more time than you think you need to. Walk through your process and gather the data manually.

BPM: In spreadsheets?

Stepleton: It could be a spreadsheet; it could be literally a blank sheet of paper. Our walls are covered with butcher paper. So sometimes we just throw it up on the wall and look at it. Sometimes we put it in a spreadsheet, or a combination of them. And we wallow in it, so to speak, for a period of time to make sure that we understand the data itself, how we're going to use the data, how it needs to be represented.

What we've run into a lot is, we come up with an idea, we walk into the IT department and tell them what we want, they design it, and then a week later we realize that it isn't exactly what we wanted to see. So we end up throwing out all that work they did and starting over. I wish someone had told me at first: Slow down, do it manually for a period of time, work out the bugs, and then create your reports.

I also strongly advise anyone who's shopping for performance reporting software, or developing it themselves, to maintain some manual interaction with the data, deliberately. Automate things that are time-consuming that you'll save labor hours by inputting. I cannot imagine inputting all of our time sheets manually; that's ridiculous. But the process of taking a number out of a system and putting it on a whiteboard and using that as your report out for the day has a human effect that you can't reproduce with a system.

BPM: Meaning that the person writing the numbers on the whiteboard will understand them better?

Stepleton: Yes. We have a spin wheel on our huddle board, and each department is on the spin wheel. Each week a different department has to put the numbers on the board. Can you imagine someone from HR having to put all these metrics up? Now they have to look at those and go, "Why is that number out of whack? That doesn't look right." It's really pretty amazing. It's such a simple concept, but it has a huge impact on what you do with the numbers.

BPM: What results have you seen from getting more people involved in the numbers?

Stepleton: A couple of things. One is that, like I mentioned before, when change happens, people understand why. Before, management would go in and make changes, using data to make their decisions, but the middle management and front-line employees didn't have the luxury of the hours of discussing why and what our strategy was. So it really creates transparency in the organization.

The second thing is it helps employees feel like they can effect change. I think people want to feel like they have control, at least the people that we have here now. It took a while to root out people who just didn't want that. And that's OK. There are plenty of places they can go work where they can just come in and do their thing. But I think now people feel like they do have control because if they make a change, they can see the positive effect that it has on the metrics.

BPM: What is an example of specific actions that individuals have taken that they can see tangible results from?

Stepleton: There are so many. I'll pick a recent one. We do tracking studies that involve surveys every day, all year long. Because of the bidding environment, we tend to bid them extremely competitively. So our margins are, at times, razor thin. We've started to identify the lowest-performing tracking project each month and use that as our example, or poster project. It becomes a mission to fix it. In one particular project, we were running at about $20 revenue per hour, which is well below what our target is in terms of being profitable. Our goal was to increase productivity in 30 days by 30 percent. Well, that forced people to go into the metrics of the job and understand why it's been performing poorly, and they discovered right off the bat that one simple thing causing some severe problems was the way that they staffed the project on a daily basis. It had to do with how many hours an employee would work on the project. As you start to dig into these data points that are in our systems, you can see that those who work less than two hours on this project aren't productive. That has to do with how difficult it is to find the person we're looking for. In this case, we're looking for a very difficult-to-find group. If an employee is not on the project for at least two hours, statistically they have a lower probability of getting someone to do the survey. So the first thing we did was reduce the number of people working on the project and increase the length of the shift. That increased productivity by 25 percent in one day.

BPM: Wow.

Stepleton: Yeah. It's crazy. And it's just a result of someone looking at the data and saying, "That looks funny."

Now we have a process for identifying what the productivity of a specific job is, which defines what the minimum shift length is. Looking at our bad processes has been painful. A lot of employees really started to get down on this process because as you look at all your systems and every function and every operation, you realize how little is standardized and stable. It's a daunting task. But I think we're past that phase and now to the point where you just fix and standardize processes on a daily basis, and eventually it will be a much different organization.

BPM: Do you have any advice for someone just embarking on a performance-management improvement project?

Stepleton: I would encourage people to engage others who are doing things well. None of our ideas are our ideas. We steal everything from other people; it's amazing what people are willing to share. And readers can call me. I'd be happy to share our experiences. I've given tours to some of my local peers, and I've gone to their businesses. I think as entrepreneurs we don't do enough of that. Sometimes it can be difficult to engage others because everyone's protective, but the reality is, even if my competitor knew what I was doing, I know how hard it is to implement. So my advice would be, look for people whom you hear of that are doing well, and go visit them. Get on a plane, get in the car, and go visit them. See what they do and take pieces of what they do and integrate it into your business.