Goals that your people understand

You're at work, it's the middle of the morning, and suddenly everyone is called into the largest room you have, for a presentation. The CEO and senior management have just finished their new strategic plan, and they are going to share it with the rest of the company.

Do you get anything out of the next hour? Or do you just spend the time trying to look awake?

I've been in too many presentations where the latter was true. The CEO starts off by saying, "We have two main goals over the next three years: to become the number two supplier of refrangulated widgets, and to reach a market capitalization of twelve gazillion dollars. Here is how we are going to do it ...." Then the rest of the speech might as well be in Babylonian, for all that I under­stand it. I'm sure the CEO is following advice he read somewhere, that he should make all employees "partners" in the company's "strategic thinking." But because the message is in terms I don't know, the only thing I get from the meeting is that I am now an hour behind on the day's work.

It doesn't have to be like this.

What does a better way look like?

There is another way to roll out corporate goals, one that makes them meaningful to every employee. It takes a little more work up front, but this is the kind of work that the management team is paid for in the first place—so it's fair to ask them to do it. Also, if they run into problems in the preliminary setup, that's a key indicator that there are problems with the strategy itself. So it is worth the effort.

The method is called Hoshin Kanri (Japanese: 方針管理, "policy management"), and it draws a straight line between the company's long-term goals and the work I have to do tomorrow. This helps me understand the company's goals, because I can see the effect they have on my job in particular. But it also helps me see how my job fits into the big picture.

Ironically, I saw the method used long before I learned it had a name. I just thought of it as "The way That Company does goals," and I wondered "Why doesn't everyone do this?" But of course I couldn't ask my next employer, "Why don't we do goals just like that other company I used to work for, that you've never heard of?" It was a relief to learn the name.

How do you do it?

The whole process unfolds in several steps. Some people use a special matrix to organize their work, but I won't do that here. The logic is the important part, and you can organize it however you choose.

Define your strategic goals

First, you have to define your long-term strategic goals. Where do you want your company to be in five years? Be careful not to define too many goals, but focus on the handful that matter.

You may also identify your most important operating imperatives at this point. But again, be careful not to cloud the picture with too much noise. (For the distinction between strategic goals and operating imperatives, see the discussion in this post, under "What is a strategy, anyway?".)

How are you going to get there?

"A goal without a plan
is just a wish."

Next, plan out very concretely what you have to do to reach those goals. A familiar aphorism attributed to Antoine de Saint-Exupéry says that "A goal without a plan is just a wish." So define the actions you will have to take, and the milestones that will prove you are on track.

In the first instance, this means defining annual goals as progress towards your long-term targets. But it also means spelling out what your goals will look like, concretely, when you have achieved them, and then identifying what it will take to cross the gap from Here to There. If you want to be—let's say—the number two supplier of refrangulated widgets, what does that tell you about your warehousing and logistics systems? What level of performance do they have to reach, in order to support the overall corporate goal? But also, what is their performance today, and how far does it have to improve? Can you spell out achievable interim milestones towards which your logistics and warehousing personnel can aspire, that will get them where you need them at the right time?

And of course reaching this goal isn't even primarily about warehousing and logistics, though doubtless those play an important role. Every single department in the company should contribute to these goals somehow. So the CEO has to delegate to the respective department heads the task of working out maps for each of their functions which will support the common strategy.

As an aside, you should check that each department map is consistent with all of the others. If Engineering plans to develop the Next Generation Widget in Year Two using a special technical tool that IT doesn't plan to install until Year Four, somebody has to change his map!

Cascade downwards 

It doesn't stop there, but the next steps are pretty straightforward. You as a department head (or functional VP, or whatever your title is) now have a strategic map for what your department has to achieve in five years, and also in this year. Take it to your section managers or group leads, and go through the exact same exercise. Ask each of them to spell out how their group will contribute to meeting your goals. Notice that they don't have to reach all the way back to the company's goals, because your goals have already been aligned with the higher level. So as long as they support achieving your goals, they are also supporting the company as a whole.

Cascade this exercise down through the company organization, all the way to the shop floor. (Yes, this is exactly the same procedure I recommended three years ago for business continuity planning.)

The result is that I, as an employee, have personal goals to achieve that are directly related to my job. But if I achieve them, that supports my supervisor in achieving his goals, which in turn supports the department manager in achieving her goals, which ultimately rolls all the way back up to supporting the company in achieving its strategic targets for the year.

End of the year

Then at the end of the year, you evaluate how you did. This means everyone, at all levels. But the point isn't just to assign a grade, like in so many performance review systems. The idea is rather to carry out a root cause analysis on each missed target, to learn why it was missed, and then to update your plans with this new information. This way you—individually and as an organization—keep in touch with reality, learn lessons from experience, and adapt your strategy pragmatically.

I will admit that it is hard to remember to do this last step. Even when I worked at a company that did all the rest of it, that step was sometimes missed. But of course it is important.

           

Make your matrix certificate work

Last week I introduced the concept of a matrix certificate, and I thought it could be useful if I explained the approach in a little more detail. Mostly the idea is self-explanatory, but it is worthwhile to keep one or two points in mind so they don't trip you up.

When ISO first introduced its management system standards, each certificate was tied to a specific location where specific work was done. So the certificate was printed with an address and a scope statement, as in the example to the left.* 

But many organizations have more than one location, and often the processes are identical from one site to the next. There is not a lot of variety, for example, between this McDonald's and that one. So organizations began asking their Certifying Bodies** if there were any way to group multiple locations under a single certificate. Since the CB's, for their part, had to manage a lot of redundant information for all of these sites, they were keen to agree. And so CB's began to introduce the "matrix certificate."

A matrix certificate works just like a standalone certificate, except that the address and scope are replaced by a matrix: Site 1 does this kind of work, Site 2 does that kind of work, and so on. There is no requirement for a literal matrix on the certificate itself, so long as the information is there. A matrix is a convenient format when there is a large number of sites to consider. When there are only two or three, the CB might simply use paragraphs, as in the example to the right immediately below.***  

The other place you see a genuine matrix is in the internal paperwork that supports the certification: at the CB, and at the organization itself. These internal matrices look something like the one that I show at the bottom of this post, and they correlate the infor­mation needed to manage the audit program. This starts by listing the sites and their scopes or func­tions, just like on the certif­icate. But then the internal matrix also tracks how many people work at each site. This is because there is a formula used by all the CB's to determine how many audit-days**** they have to schedule at each location. The formula uses both number of employees and type of work as inputs to the calculation: a big plant needs more audit-hours than a small plant, and manufacturing requires more attention than HR or sales. Finally, the matrix should track how long the audit cycle is for each site. Normally with a multi-site certificate, the company headquarters is audited every year; also, manu­fac­turing is likely to be audited every year. The example I give below has one site for each of those functions. But then it also shows three sites that mostly do engineering or product design, that follow exactly the same procedures, and that are therefore more or less interchangeable. In order to reduce the overall audit costs (and to simplify scheduling), the CB has put each of these on a three-year cycle, so that they rotate. In any given year the CB plans audits in three locations; but over the three years before recertification, they visit all five sites.

On the whole, using a matrix certificate is a good way to control your certification expenses if you are in a situation where it works for you. But of course there are risks. The main risk is that one of the critical entries in your matrix will change: for example, you might have heavy layoffs at one site, or (conversely) hire a whole new department. Either way, you have to make sure to report the change to your CB in a timely way, so that they can recalculate your audit days. One way or another, this might change your plans for the year.

Or you might move functions from one site to another. Since each function makes its own contribution to the calculation of audit days, this move might make a difference to the result even if the number of employees involved is not high. A specialized technical function, for example, might employ only a few people; but it might require a lot of Quality equipment to make sure it operates correctly, and in some cases all that infrastructure might have to be audited.

Finally, there is a risk that two different sites are supposed to be doing the same work according to the same procedures, but one site decides to implement local improvements without telling the other. I have seen this happen, where an organization had two Distribution Centers. One was more or less a glorified shipping dock at the far end of a building that did a lot of other work; the other was a pure Distribution Center with no other functions. Back when the organization first codified their written procedures, someone wrote a single set of logistics documents and told both organizations to use them. But in the intervening years, the Site Manager of the standalone Distribution Center brought in a lot of equipment to modernize his operations. It was exciting to visit, because every time I arrived they had improved something new. But somehow the word never got back to the first operation. I had workers from that one tell me with a straight face, "We have exactly the same procedures they have there," when there was no longer anything in common. And I had to explain, "No you don't. And at the very least your documentation has to be updated to reflect that reality."

Note that when changes like this take place, it can undermine the logic behind a matrix certification. The justification for auditing Site 3 in place of Site 4 and Site 5 is that they follow the same procedures in all three sites so you'll see the same evidence anyway. Once that is no longer true, it is harder to keep down the number of audits.  

It is also true that breakthrough improvements on such a scale are mostly the exception rather than the rule. And if there is a chance for your organization to make breakthrough improvements, having to replan your audit program is a small price to pay!

__________

* I found this illustration online by searching for "sample iso 9001 certificate." I have no professional, personal, or financial connection to the organization in question. 

** A Certifying Body (CB) is the registrar that send out your external auditor, manages the audit paperwork, and issues your certificate. There are lots of CB's in the world. If you look at the certificates that I posted above as examples, the first is from TÜV SÜD, and the second is from DEKRA. These are both well-known CB's. Just by coincidence, I have never worked with either one.  

*** Again, I found this illustration by searching online and have no connection with the company. 

**** One audit-day means one full day of auditing by one auditor. So four audit-days might mean one auditor for four days, or two auditors for two days, or one auditor for three days plus another who joins him for only one of those days, or any other combination that adds up to four.     

           

How do you certify a big corporation?

Last summer, in the middle of a podcast about something else, Kyle Chambers raised a question: how is it possible to certify a really big company to a standard like ISO 9001? His point was that really big companies have so many parts that they can't all play together. (I mention the podcast in this post here.) Kyle might have meant the question rhetorically, but of course he is right. If you have scores of locations spread across multiple continents—engaged in dozens of lines of work—how can you possibly coordinate them all into a single management system? How can you possibly certify anything so complex?

You can't. So you break it in pieces.

More exactly, you divide your really big company into sub-units of a useful size, and then make each sub-unit responsible for its own certification. Then in your sales and marketing literature you speak of the company as a unified whole, announce that "Con­glom­erated Enterprises has been proudly certified to ISO 9001 since ...," and follow with the date of the first certificate that came through.

How big is "a useful size"? It really depends. I've seen it done several ways.

In general, there are two competing pressures at play in determining the right size for an entity to serve as the scope of certification.

• On the one hand, you don't want the scope too narrow, or you'll have to pay for too many audits. In other words, if three offices that are all in the same city all do exactly the same kind of work, you can save some money by letting them share a certificate. You pay for less overhead at the Certifying Body, and you pay for fewer audits (because all three offices are doing the exact same thing). This all looks good on a balance sheet.
• On the other hand, you don't want the scope too broad, because there's always the risk that someone goes crazy one day. If you have a hundred locations all sharing the same certificate, and someone in a tiny office way out in Far Foodle starts violating an important policy, the next auditor might rate it as a Major Nonconformity. Then that Major might put the certificate at risk for all hundred locations!

Where you draw the line while balancing these two imperatives depends a lot on your organizational culture. I've worked for one multi-site operation whose corporate ethos was entrepreneurial, and who had created a number of sites by acquisition. So there was a distinct chance that two sites might not be on the same page. For that company, each site was strictly responsible for its own certifications. The company had a blanket contract with a CB, because they got a volume discount. But inside that blanket contract, we were on our own. I made the arrangements for my location, but not for the others; I worked with the auditor and tracked the findings for my location, but not for the others. We never had a crisis where one location risked losing certification, but the company's management took no chances.

But then later I worked for another company that bundled eight sites across the United States all into one certificate. We supported the same product line, and we were all in the same geographic region. So the company decided that was enough commonality that it made sense for us to share a certificate, and we could sink or swim together. I made the arrangements for eight locations, tracked findings for eight locations, and flew around the country a fair bit to support audits when they happened. 

This meant I carried out eight internal audits a year, but not nearly so many external audits. We had a "matrix certificate," which meant that the CB did a sampling every year: they always audited headquarters, and they always audited our one factory, but then they would pick one or two other sites randomly and leave it at that. Over the years, of course, they ended up visiting every site; but every spring I had a long discussion with their scheduler to agree where they should go. 

"You visited A last year, and you visited B the year before, but you've never audited C. How about seeing them this year?"

"Wait, we've never been to C? What do you even do at that site?"

"It's basically the same work as B, so I don't expect any significant findings. But you haven't been there yet, so it might be good to visit."

"Sure, I guess. How many people work there ...?"       

As long as the processes and systems really are common across locations, this can be a useful way to proceed. Next week I'll say a little more about how matrix certificates work.