Author: Matt Tuttle

Since the last posting, we now know that an Order of Magnitude estimate can be derived by using this basic formula:

Estimate = (a + 4m+ b) / 6

a= most optimistic estimate

b= most pessimistic estimate

m= most likely estimate

We also know that in order to generate such an estimate we need some general knowledge on the subject because there is such a heavy emphasis on Value M in the formula. Without any knowledge on the subject at hand, providing a most likely estimate would be like coming up with a Silly Wild-Assed Guess—or a SWAG.  However, even with such general knowledge, our Order of Magnitude estimate will still yield us a 35% accuracy level from the onset.

Not good enough?  Next we introduced the concept of standard deviation, a measure of variability or dispersion of data points in comparison to the mean.

Standard Deviation = (b – a) / 6

We said that by calculating one standard deviation of the Order of Magnitude estimate, we could get the estimate from 35% accuracy to 68.7% accuracy.  Why is that?  Let’s look at a normal bell curve.

The red area represents one standard deviation (as know as one sigma) and how much of a normal curve it covers.  If the exact average of a normal curve is right on the 10-mark, we see that one standard deviation ranges from the 9-mark to the 11-mark.  If this were a piece of pie, you’d cover basically 69% of the pie leaving 31% not covered.  For most project managers, that leaves too much uncertainty and error; they like answers that cover most of the curve.

We see in the green area that if we want to cover 95.4% of a normal curve then we need to provide for two standard deviations, between the 8-mark and 12-mark.  This will only leave 5% uncovered and is a percentage that most business managers are ultimately comfortable in working in.  95% is pretty good bet and provides managers with statistical comfort; getting to 99.7%, or three standard deviations, between the 7-mark and 13-mark, certainly only leaves 0.3% of the curve uncovered, but estimating purposes, such coverage is not needed as 95% gets us where we want to be without much error left to factor in.

In Part III, we derived the following estimate and standard deviation for the length of the Golden Gate Bridge (in kilometers):

Estimate = 2.42 kilometers; standard deviation of 0.583 kilometers.  In short, we could say this:

“I am now 68.7% certain that the Golden Gate Bridge is 2.42 kilometers in length plus or minus 0.583 kilometers—or the range of the distance of the bridge is between 1.837 kilometers and 3.003 kilometers.”  With one standard deviation applied to our estimate, we are basically covering 68.7% of a normal curve—but we are still leaving out 31.3%.  That’s a lot of room for error.

If we want to cover the 95.4% range and reduce our chance of error to only 5%, all we simply need to do is add more one more standard deviation.

One Standard Deviation = (b –a) / 6

Two Standard Deviations = 2 [ (b-a)] /6

Three Standard Deviations = 3 [ (b-a)] /6

So, from our previous example, to provide a 95.4% accurate, two standard deviation estimate, we take our standard deviation and simply multiply by two.  Estimate now equals 2.42 k +/- 1.168k or “I am now 95.4% certain that the Golden Gate Bridge is 2.42 kilometers; that the range of the distance of the bridge is between 1.252 kilometers and 3.588 kilometers.”

Turns out the Golden Gate Bridge is exactly 1.9 kilometers in length and that our one standard deviation estimate actually covered that, but without looking it up or knowing for sure, we could only be 68.7% statistically sure of that fact.  By applying one more standard deviation, by going to two sigma, we moved up to 95.4% sure and that is a much more comfortable area to be in when making important decisions.  Yes, the range of the estimate increased dramatically, but in return, your chance for error reduced dramatically as well.

In Part V of the High-Level Estimating series, we are going take our knowledge of deriving an estimate and applying standard deviations and we are going to calculate a 95.4% accurate estimate for the length of a 4-task project schedule.

Author: Matt Tuttle

We have previously explained how inaccurate and invalid a SWAG estimate can be and we have laid the foundation that a solid Order of Magnitude-type estimate truly does require that the estimators have a general knowledge of the subject. It is this general knowledge that dramatically separates the two types of estimates and allows us to begin with a 35% confidence level.

I have also introduced the following formula, which can help provide a fast Order of Magnitude estimate any time one is needed:
Estimate = (a + 4m+ b) / 6
a= most optimistic estimate
b= most pessimistic estimate
m= most likely estimate

Sure, deriving an Order of Magnitude estimate does not require this formula; someone could be so expert and proficient in a specific area that they can come up with Order of Magnitude estimates simply based on past experiences. I’ve seen $100 million IT roadmaps initiated solely off of this concept in which IT leadership laid down all the projects they wanted to possibly consider in an upcoming fiscal year and the initial creation of the first round budget was developed by quickly gauging the size of the individual efforts (like small, medium, large, XL, etc.) and assigning it a rounded dollar value based on that size—a small effort was worth $50k, a medium effort was $250k, etc. This could be done with high confidence because the IT experts in the room had great knowledge on what it would take to complete an effort, assign it a size, and then assign its predetermined dollar value. Very quickly you have a first round budget and now you have a good idea of what next steps are needed, like does the budget need to be reduced, what projects take priority, and determining which projects need additional research to narrow down effort assumptions. If you lack that type of expertise, this formula is a great tool to get you there.

As mentioned earlier, my father loved to teach this formula by asking the person he was “training” to estimate how long is the Golden Gate Bridge? Let’s walk through how he helped derive an estimate for this. For sake of this article, I want the estimate to be made in kilometers, I’ll explain why later.

First, he would start with variable A, the most optimistic estimate. He would say, “Okay, you have a good picture of the bridge in your mind, for Value A, I want you to provide me a distance that you know for sure the bridge is AT LEAST this long. For example, you know for sure the bridge is at least 0.1 kilometers long, right?” They would agree, but the point is, Value A can be a ridiculous answer on the low side, the formula does not care. For today’s article, let’s go with 0.5 kilometers.

Next, he would go to Value B, the most pessimistic estimate. “Okay, using the same line of thinking, for Value B, provide me a distance in which there is no way the bridge can possibly be THAT long. For example, the bridge can’t possibly be 4 kilometers long, can it?” Probably not, so let’s go with it, Value B is 4.0 kilometers.

Now, Value M is critical. Value M, which is the most likely estimate, requires all the estimator’s general knowledge and logic be applied into a very decent response. “For Value M, really think about how long in distance is the Golden Gate Bridge—this is your best, most logical guess based on what you know about the bridge.” It does not take a math major to see that the formula relies heavily on the value for M. Simply put, the numerator of (a + 4M + b) could be written like this: (a + m + m + m+ m + b). The fact that there are four M’s in the top line dictates that this value will factor greatly in the final answer. This M value is what instantly separates this from being a SWAG; with so much emphasis on M, general knowledge or expertise is required to start with that magical 35% accuracy. For today’s exercise, we are going to estimate that the Golden Gate Bridge is 2.5 kilometers long as our best guess.

So, we have A at 0.5k, B at 4.0 k, and M at 2.5k. Let’s plug and play and get our estimate.
Estimate = 0.5k + 4(2.5k) + 4.0k / 6
Estimate = 14.5k / 6
Estimate = 2.42 kilometers

Whether or not we are right, wrong, or fairly close to the actual distance of the bridge, we now have an estimate that actually gets us to being around 35% accurate. 2.42 kilometers is surely a better answer than “I don’t know” yet allows us to build a mental picture in our head about how long is the bridge without truly knowing.

Okay, so a 35% estimate is not good enough for you; you still want a little more certainty. In the next article, we will apply some standard deviation around our estimate using the formula: Standard Deviation = (b – a) / 6. In the meantime, here’s a quick preview so you can see where we are heading:
Standard Deviation = (4.0k – 0.5k) / 6
SD = 3.5k / 6
SD= 0.583k

Using our original estimate of 2.42 kilometers with 35% confidence, we can now say, by using one standard deviation, “I am now 68.7% certain that the Golden Gate Bridge is 2.42 kilometers in length plus or minus 0.583 kilometers—or the range of the distance of the bridge is between 1.837 kilometers and 3.003 kilometers.

What is the actual answer? The Golden Gate Bridge is 1.9 kilometers in length. Our 35% estimate was off by little more than a half a kilometer, but our 68% estimate using one standard deviation puts us right where we want to be.

Why I use Kilometers?
My dad loved using the Golden Gate Bridge as his estimating teaching lesson, but when you begin applying estimates for an object that is only 1.1 miles in actual length, it really limits the “teaching effect” because the estimator is going to hang around the 0.5 m, to 1.0 m, to 2.0 m range and that doesn’t provide great separation in teaching standard deviation a little later. I choose kilometers for this example because it allows us to expand the numbers we will use in the lesson, which then allows for great standard deviations.

Author: Matt Tuttle 

As mentioned in my previous article, my father was a nut when it came to asking for estimates. He wanted estimates to be made quickly. He wanted them to be at a high-level and not exact, and he wanted enough logic applied toward them to be around 35% accurate right from the onset—in short, he wanted Order of Magnitude estimates. It didn’t matter who you were or what you did, he set this expectation for everyone, both from the young teenager to the seasoned professional. If you failed to provide an Order of Magnitude-type of response when asked for an estimate, he let you know about it…actually, even worse, he’d teach you how to do it right then and there.

My father took it as a teaching moment; he wanted people around him to be well-versed in how to provide him estimates moving forward whether they reported to him at work, were the neighbor next door, or one of my girlfriends. It was an amazing and fascinating discussion to watch when he was mentoring a new engineer in his facility; it was downright uncomfortable watching him grab a pen and paper and begin providing an estimating dissertation to my new girlfriend who simply came over to meet my parents and enjoy a relaxing dinner.

First thing he would always start out with is the difference between a SWAG and an estimate (keep in mind he never knew of Order of Magnitude, so every time he referenced an “estimate,” an OoM is what he was truly seeking). In this case, SWAG has nothing to do with stealing bounty off a pirate ship or collecting knickknacks and souvenirs at a business convention. SWAG, in the estimating world, is an acronym that stands for a “Silly Wild-Assed Guess”—a form of estimating that is seldom used in the profession of project management.

The main reason why SWAG is not a commonly-used estimating technique is because the person or people providing the SWAG are simply guessing with no logic or expertise behind it—it is a “wild-ass” answer in which the degree of certainty hangs close to zero. How this differs from an Order of Magnitude estimate is that the estimator, in such a case, knows enough about the subject to put some kind of logic behind the answer to get close to at least starting at 35% confidence. The person does not have to be an expert or a professional in a specific field; they basically need a general understanding about a whole range of topics to be able to at least put a little logic into the answer. A SWAG, on the other hand, lacks this completely. So, how did my dad start by proving his point when teaching somebody about estimating? He simply asked them to estimate how long is the Golden Gate Bridge?

If you ask this question to somebody who has never seen the bridge in person, has never seen the bridge in a picture, has no idea what the geography of San Francisco looks like, wasn’t aware that the city was home to such a famous bridge, and doesn’t know much about bridges in general, you will obtain a SWAG for an answer. However, with an Order of Magnitude estimate, you can start off estimating the length of the famous red bridge by merely having some common knowledge about it. You don’t have to have ever driven across it or have seen it person, but chances are, most people you talk to know about the Golden Gate Bridge in some way or some form—and it is this small piece of knowledge that quickly takes an estimate from being a SWAG to an Order of Magnitude. Ask me the circumference of the earth’s moon and I can Order of Magnitude it. I’ve seen it in person, I know the formula for circumference of a circle, and I have a general thought is my mind about the diameter of the moon compared to that of the earth. With some quick logic, I can derive an estimate for the diameter, in miles of the moon and therefore provide a circumference—and ultimately a 35% accurate Order of Magnitude estimate. Ask me to estimate the diameter of the largest moon circling Jupiter and I will have to SWAG it. Why? I just don’t know enough general knowledge about it. I don’t know how many moons there are around Jupiter and I have no size parameters from which ones are the biggest moons compared to the smallest—and I certainly don’t have a perspective on how that relates to our own moon. Whatever answer comes out of my mouth on this estimate will lack any logic or knowledge and therefore would be useless to base any initial decisions on. So, when my father asked for an estimate (an Order of Magnitude estimate, mind you), the first assumption he always made was that the person providing the answer at least had enough common knowledge on the topic to generate such an estimate. Once that was established, he could start asking you to fill in the variables in the easy high-level estimating formula:

Estimate = (a + 4m+ b) / 6
a= most optimistic estimate
b= most pessimistic estimate
m= most likely estimate

This was my father’s “bread and butter” formula for providing an Order of Magnitude estimate! Short enough to quickly memorize and use forever and simple enough algebraically to perform in your head or on a scratch piece of paper. It was the formula that immediately got you away from a SWAG yet rigid enough to yield an answer with at least 35% validity to it. It was his perfect formula. He never told me when he learned the formula, but as long as I can remember, he was synonymous with it; the formula was a part of his DNA. He used it all the time and he wanted those around him to use it.

Part III of High Level Estimating posts soon.  We’ll put the formula to use in order to derive an estimate on the Golden Gate Bridge—and then we’ll apply a simple trick of using standard deviations to put even more certainty around our final Order of Magnitude answer.

Author:  Kristen Backstrom

I recently heard that iPads are replacing textbooks in schools.  Can you tell me a little more about this? Do you think this will be good for students?

                                          – Gina, Denver, CO 

Gina,

From Kristen:  Here’s a video link from apple.com detailing how the iBooks Textbooks for iPad will work within the educational system. http://www.apple.com/education/#video-textbooks

As an educator, I think Apple, McGraw Hill and Pearson are right on for having iBooks Textbooks available on the iPad!  These companies have found an innovative way to enhance the product lifecycle of the iPad technology, and it’s going to make some significant changes to our culture too. Changes won’t only be to the textbooks currently being used, I’m sure we’ll also see a decrease in the need for school supplies like paper, pens, binders, and post-it notes. Imagine the cost savings!  Education has reached an era where its meant to be a self expression, and it’s great that these companies created a form of learning entertainment.

From Matt:  As a parent, I want learning to be as interactive, fun, and easy as possible.  So, if you ask me if I’m in favor of going from printed textbooks to iBooks Textbooks, I would say yes, and here’s why:

• My children’s generation is geared towards computer-based learning; a transition to reading and learning from a textbook on the iPad should be seamless.
• Instead of regular graphs or pictures in a static book, iBooks Textbooks probably have graphs that can change or video to support a concept–making the learning experience even better.
• The cost of iBooks Textbooks will be cheaper for the consumer.  Yes, content development will still be the main cost driver, but materials, printing, and shipping costs should be reduced dramatically.
• iBooks Textbooks will also be more environmentally conscious than a printed book.
• Carrying an iPad around is certainly easier than lugging a backpack full of heavy textbooks.

_________________________________________________________

We’re interested in hearing your questions relating to technology trends, or workplace behavior issues.  Please send your question via email to kristen.backstrom@pnta.net and reference Readers’ Question as the subject.

Author: Matt Tuttle

Growing up, one thing my father continuously reinforced to my sister and I is that when someone asks you for an estimate, give them exactly what they ask for—an estimate. While most kids were taught to look both ways before crossing the street and while others were learning to measure twice and cut once, my sister and I had the fortune of learning to provide estimates to my father who expected them to have some logic behind them. He expected the answers to be in the ballpark, and he expected them quickly; we basically had to follow the same requirements as the engineers who worked for him at a local high-tech company. We may have been the only kids on the planet allowed to provide approximations to most of the questions he asked.

Nothing would drive my father more nuts than to ask for an estimate and then watch someone crunch through a set of numbers, take several minutes to derive an answer, and then provide a number with a bunch of crooked figures and decimal points associated with it.  He inherently felt that such an exercise defeated the purpose of asking for an “estimate.” If he had asked for an exact answer, he would expect someone to take their time and provide numbers that had a high degree of accuracy. Estimating held a definition that was synonymous with speed and providing answers “somewhere in the ballpark.” He also wanted rounded numbers. When asked what time I would be home from a party, answering 11:20 P.M. did not suffice. Dad wanted 11:00 P.M. or 11:30 P.M. to come out quickly. He wanted a good idea of the parameters to his query; he did not care about specifics.  I later learned I could answer 2:00 A.M. and get away with it as long as I never answered 1:53 A.M.  In short, any answer with a modicum of detail was frowned upon. Certainly this caused me some frustration in elementary school—as a teenager I used it as a means to get his goat. I’ll never forget the day he asked me the value of pi, and I provided 10 digits after the decimal point. He literally paid me $20 to get out of the house because he was expecting an answer of 3 or 3.1, and he could not work with me any longer because he was so frustrated with the answer. It was a planned response on my part; I ran out of Mountain Dew and knew my answer would get me out of doing my homework for a while.

My father was always strategic in nature both professionally and in his personal life, so the estimates he sought never needed to be exact. He never realized that there are three classical types of estimates ranging from pie-in-the-sky answers to those with more detail and certainty. He did not know that when it came to estimates, there were actually several options: Order of Magnitude, Approximate Estimate, and a Definitive Estimate—each one providing an estimate, yet one more exact than the other.  In the world of project management, each type of estimate has a purpose and a timeline to when they should be applied.  Sure, grabbing a number out of the air based off an educated guess has its purpose in life and on a project, but so does providing an estimate that holds a 95% confidence interval—as well as estimates somewhere in between.  True, my father may have always wanted Order of Magnitude estimates in his line of work and in his life, but eventually, somewhere down the road of a project, human and capital resources will have to be assigned, budgets will have to be allocated, and completion dates set.  By utilizing a proper project management methodology, the three various forms of estimating allows a manager to work through the logic at key intervals to eventually provide the details to answer those crucial questions.

Order of Magnitude
This is where my father thrived; these are the answers he expected any time he requested an estimate. While he never officially defined them to me as Order of Magnitude, this is exactly what he was seeking. An Order of Magnitude estimate can be derived very quickly, it does not require expert knowledge to formulate an answer, and without much detail can still put you in a range of being around 35% correct. To him, speed was the key, so in his line of work, having a 35% correct answer always sufficed. After studying estimating, I could see why that was the case. Order of Magnitude should provide you a very quick answer; calculations should be made in your head. If you’re doing that, chances are you’re using round numbers already, so the answer itself will be rounded as well. Perfect! And even with at 35% confidence, it at least provides some very early parameters; for my father, that meant he could make some initial decisions and move forward. Speed! Speed! What I truly learned, however, is that in his line of work, he made high-level decisions based on Order of Magnitude answers that…wait for it…provided a starting point to drill down further and derive more detailed estimates for the next stage of the effort.  I don’t know exactly what my father did day-in and day-out in his profession, but Order of Magnitude estimates were the optimal way for him to operate and make decisions.  It was the gateway for determining whether to stop or move forward and for other teams to vet out more details later.

Stay tuned for Part II of High-Level Estimating.  I’ll look into how to provide a high-level estimate quickly, yet establish levels of confidence around it. Sure, such an exercise would drive my father a little nuts, but with 30 more seconds to think, you can provide an answer that is both quick and fairly accurate—and has a little more teeth to it than a SWAG (that’s an acronym that has nothing to do with booty on a pirate ship).

Matt Tuttle has joined PNTA.net as a content writer for our website.  Matt is a program management professional with over 20 years of experience, and an award-winning Toastmaster.  He has led many major continuous improvement and software deployment projects/programs with companies like Dell, Inc. and NCS Pearson in such areas as IT, Procurement, Finance, and Operations.  Most recently, Matt was the program manager overseeing the installation, deployment, and data migration of the Agile Product Quality Management module into Dell’s global manufacturing operations—a primary aspect of Dell’s entire Product Lifecycle Management portfolio.

Matt enjoys writing about business and everyday life, reading fiction, and following baseball; he is also the President of the Board of Directors for a Texas-based Select 13U baseball team.

Matt has a B.S. in Business Administration from UNLV, an MBA from the University of Phoenix, and a Master’s of Science in Project Management from the University of Wisconsin-Platteville. He resides in central Texas with his wife and two children.

Matt can be contacted at the following email address:  mj_tuttle@hotmail.com

We are looking forward to his contribution!

Author: Kristen Backstrom

Be part of the conversation! PNTA, Incorporated’s language education services, LipLogic , released English to Mandarin Words and Phrases.  This is a native app that’s fun and easy to use.  The app is available in iTunes, and is compatible with iPhone 3G and higher, iPad, and iTouch.  All you need is IOS 5.0 on the device to run it.

LipLogic.com is logic for your lips!

Author:  Kristen Backstrom

Here’s another app from Personal Networking and Technologies Associates.  Our language education services, LipLogic , released English to German Words and Phrases.  This is a native app that’s fun and easy to use.  The app is available in iTunes, and is compatible with iPhone 3G and higher, iPad, and iTouch.  All you need is IOS 5.0 on the device to run it.

LipLogic.com is logic for your lips!

Author:  Kristen Backstrom

This just in!  PNTA, Incorporated’s language education services, LipLogic , released English to Lakota Words and Phrases.  This is a native app that’s fun and easy to use.  The app is available in iTunes, and is compatible with iPhone 3G and higher, iPad, and iTouch.  All you need is IOS 5.0 on the device to run it.

LipLogic.com is logic for your lips!

Author:  Kristen Backstrom

PNTA, Incorporated’s language education services is growing!  Today,  LipLogic released English to French Words and Phrases.  This is a native app that’s fun and easy to use.  The app is available in iTunes, and is compatible with iPhone 3G and higher, iPad, and iTouch.  All you need is IOS 5.0 on the device to run it.

LipLogic.com is logic for your lips!