Got a Problem?

The book Thinkertoys by Michael Michalko is one of those business books that is so full of great ideas, you feel like taking action on many of them as soon as you read them. One of them is called “Phoenix.”
Say you have a problem, a really serious problem within the organization. What to do? How to approach it? According to Michalko, The Central Intelligence Agency (CIA) developed a series of questions designed to drill down into problems, in part to make sure that a problem is looked at from many different angles. The list of questions presented here should not be viewed as “the last word,” but rather “yours to improve.”
Getting you mind wrapped around this technique may be more effective if you consider for a moment a problem – a tough one – that you face. Write it down so that it is clear what exactly you are saying the problem is, not something vague. Now, answer these questions about the problem:

The Phoenix Checklist
1. Why is it necessary to solve the problem?
2. What benefits will you receive by solving the problem?
3. What is the unknown?
4. What is it you don’t yet understand?
5. What is the information you have?
6. What isn’t the problem?
7. Is the information sufficient? Or is it insufficient? Or redundant? Or contradictory?
8. Should you draw a diagram of the problem? A figure?
9. Where are the boundaries of the problem?
10. Can you separate the various parts of the problem? Can you write them down? What are the relationships of the parts of the problem? What are the constants of the problem?
11. Have you seen this problem before?
12. Have you seen this problem in a slightly different form? Do you know a related problem?
13. Try to think of a familiar problem having the same or a similar unknown
14. Suppose you find a problem related to yours that has already been solved. Can you use it? Can you use its method?
15. Can you restate your problem? How many different ways can you restate it? More general? More specific? Can the rules be changed?
16. What are the best, worst and most probable cases you can imagine?

Presumably your brain has been exercised by this process, but I think this approach works even better when done by a team. The idea is to harass the problem so much, that it “gives up its secrets” and becomes easier to deal with. In a group setting you could write the answers on big sheets of white paper and attach them to the walls as you work through the questions. This way people looking around the room could continue to keep coming back to what has already been said, letting the information soak in by repetition. Seeing different aspects of the problem side by side may cause a flash of insight or realization.

Now that the problem has been identified, how about a plan to deal with it? Here is another set of questions designed to uncover the solution or solutions:

The Plan
1. Can you solve the whole problem? Part of the problem?
2. What would you like the resolution to be? Can you picture it?
3. How much of the unknown can you determine?
4. Can you derive something useful from the information you have?
5. Have you used all the information?
6. Can you separate the steps in the problem-solving process? Can you determine the correctness of each step?
7. What creative thinking techniques can you use to generate ideas? How many different techniques?
8. Can you see the result? How many different kinds of results can you see?
9. How many different ways have you tried to solve the problem?
10. What have others done?
11. Can you intuit the solution? Can you check the result?
12. What should be done? How should it be done?
13. Where should it be done?
14. When should it be done?
15. Who should do it?
16. What do you need to do at this time?
17. Who will be responsible for what?
18. Can you use this problem to solve some other problem?
19. What is the unique set of qualities that makes this problem what it is and none other?
20. What milestones can best mark your progress?
21. How will you know when you are successful?

As with so many things, the time and effort invested in the process will have a lot to do with the value derived. But it seems like a good methodical approach to solving problems. Asking good questions, and constantly striving for better questions, seems an unbeatable approach to dealing with problems as opposed to living with them.

A Method of Attributing Overhead

In an earlier article I described what a better pricing model might look like and in another defined in very general terms what “overhead” means within a component manufacturing company. Suggesting a way to put the two together will be the purpose of this article.

In my Technical Support days I used to help explain to component manufacturers how the costing part of the design software worked. Setting up the materials with their proper costs was straightforward, and I could get through the labor estimation with its “setup” and “run” factors without much of a problem. With overhead, the program seemed to want a percentage of something – of material or sale price. We’ll go a different direction and describe a method of attributing overhead based on time.

Two Bins

Let’s say we’ve looked at our business and have put our costs into two large bins. Into the ”direct” bin we’ve put the shop labor, everything that goes into the actual manufacturing process. Into the “overhead” bin we’ve put everything else: rent, office salaries, taxes, and a whole lot more. The important thing is that every cost be accounted for in one of the two bins. The overhead is then added up and converted into a total dollar figure “per month.” Let’s say our figure adds up to $52,000 per month.

Total Direct Hours

Next, we look at the production workforce; everyone who helps us produce deliverable product. Let’s say that’s 18 people, including people who saw, catch, move, build, and stack. Now we’ll multiply the number of people by the number of hours we are working. If we are not too busy and not working overtime, that would be 8 hours x 18 people x 22 working days per month = 3168 hours per month. We need to  double check this against the payroll to make sure that that’s about the total number of hours we are actually paying people.

A Labor Model that Estimates Time

Here is the hardest part. We want to create a labor estimating routine that accounts for every part of the direct labor that is being used, and we want it to come very close to predicting what labor we end up actually using. At the end of the month (week, day,) we want the total hours that we thought we would use up (using our labor model) to add up to the hours that we will actually be paying people for. This will be challenging at first if we have been using a very simple method to estimate labor costs, but it is far from impossible. We begin by looking at each activity that people are involved with and provide our initial “best guess” as to the time that the activity will take “per” something – per piece or per truss or per plate. We then apply it to all the jobs we did in a given day or week and see how it compares to the “real” time. We adjust, refine, and through the process learn more about our how our people are actually working. We won’t actually use this labor model until we are confident that can reasonably accurately predict our real production on every job, and very accurately predict our real production over many jobs. Nothing good comes easy and this is where the real work – and reward – of this process will come from.

The Endgame and the Big Payoff

Back to overhead. You’ve calculated that your overhead is currently $52,000 per month and your direct hours add up to 3168. That works out to an overhead of $16.41 per direct labor hour. Now, when calculating the cost of a job you will tally the materials, direct labor, and, using the formula we’ve just created, add an overhead cost equal to $16.41 x the number of direct labor dollars that job is expected to consume. If you estimate the job will consume 100 hours of that precious 3168 hours you have to build things with every month, the job needs to “pay back” $1641.00 to pay for its share of the overhead. You’ve identified your break-even point (congratulations!) and with that comes a lot of power to intelligently decide what jobs to take and which to let go.

What is Your Conversion Rate?

In his book “Why we Buy,” Paco Underhill tells the story of senior executive of a multibillion dollar chain who was asked “How many of the people who walk into your stores buy something?” This executive knows a lot about his business. He knows the total sales, the average sale amount, the sales in any given store vs. last year, profitability by item, and many other things. His answer to the question was, “Damn near all of them.” Since their stores were “destination” stores he reasoned, people didn’t go there unless they had some very specific purchase in mind. The reality was 48%, which, as it turns out, is a very good rate for stores of its kind. Underhill’s book is filled with stories about how many things are “hidden in plain sight,” and how obvious some things are once we can “see” them.

The concept of the conversion rate, implying as it does that shoppers need to be “transformed” into buyers, was alien to his organization. That changed when the company realized that by changing some things about the store, they could improve their conversion rate.

Our Batting Average

Most component manufacturers I know can quote their sales figures for any given month. Sales totals and gross profit figures, viewed together, seem to be the most basic indicators of the overall health of the business. Fewer are as conscious of their conversion rate, or “Of the opportunities we had, how many bought something?” It’s the “batting average” for the business.  If we know we sold “100” this month, was it 100 out of 300, or 100 out of 1000? The conversion rate is a third very fundamental way the health of the business can be measured, and also carries with it important intelligence about the market we are operating in.

Keeping Score

If we are tracking quotes and have some kind of a way of calculating what percentage become orders, we have the basic information needed to calculate our conversion rate. We also need to know, “Among the quotes that have not turned in to orders, which are still ‘in play,’ and which went to competitors?” Even better information can be gathered if we also record why they did not become orders. Although many will go in the books as lost to “lower price,” the ones that don’t will be instructive. We might see “Couldn’t meet delivery date,” “Loyalty to other supplier,” “Project was cancelled,” “Other supplier was able to include wall panels” - and others. Gathering this information will, over time, uncover some new insights into the market. Imagine, for example, the market intelligence we’d have in observing the percentage of quotes where the “Project was cancelled” increase over time from an average of 5% a month to 15% month.

Is there any reason not to record and review “Why we didn’t get the order” or “Why we got the order” (!) on every quote? Among those lost to the competition, why shouldn’t we find out and record who the competition was that beat us out and “keep score?”

What we Might Learn

Although looking at the numbers alone would tell us some things, even more revealing might be seeing changes in the numbers over time. For example, if we see that we were getting a conversion rate of about 40% for the 3rd quarter, but it went down to 30% in the 4th, it tells us that something is changing. What is it? Has a new competitor come into the market? Is there less work out there and competitors are lowering their prices (more than we are?) If we track the fate of every quote, the reason we got or lost the job, and who we lost out to, we’d have great insight into the current market.

In a general way, we use conversion rate information to look at:

1) the extremes – the biggest and the smallest conversion rates in any sample and ask, “Why?”

2) the biggest changes over time and ask, “What’s going on and what (if anything) should we do in response?”