The Basics

0:00Welcome to Basic Estimates.

0:02I'm Bob, your hostess with the mostest.

0:04We're going to talk about a number of things

0:06related to project estimates in no particular order.

0:10And I'm going to do this with a range of theory

0:13to a range of practical use.

0:16No particular order.

0:18Here we go.

0:18So what are we talking about when we talk estimates?

0:21We're talking cost and time.

0:24But here's the burn.

0:26Time is money.

0:27So I always like to talk about schedule estimates first,

0:30or the estimates to do the work.

0:33Activities.

0:34But there's another relationship here

0:36you need to be aware of.

0:37Do not forget it.

0:38These are your triple constraints.

0:41You have scope, schedule, and budget.

0:44Why does it matter?

0:46Because if you change one of these,

0:48you almost always affect one or two of the others.

0:53And let's face it.

0:54Let's be real here.

0:55If you change one of them,

0:56there's a 99.9% chance you're going to affect the others.

1:00So if you change something, just be aware,

1:03you should be looking in the other directions.

1:05But those aren't the only constraints.

1:08Those are the classic triple constraints,

1:11or as I like to call them, the triple complaints.

1:14There are other factors involved that impact your estimates.

1:18They are risk.

1:20If you're doing something well known,

1:22the risk level is likely low.

1:25But if it's a brand new task, brand new team,

1:28brand new location, something complex, risk goes up.

1:33And if risk goes up, it's going to take more time.

1:36If it takes more time, it's going to cost more dollars

1:40in cents or whatever currency you're using.

1:43What about quality?

1:44Higher quality, more time, more money to achieve it.

1:48If you go from one sigma to three sigma to six sigma,

1:53six sigma means you have less than three and a half incidents,

1:56I think, or three and a half failures

1:58out of one million possible failures.

2:01And the only quality standard higher than six sigma

2:05is 100%.

2:07What do you think it would cost to achieve 100% perfection

2:11in anything?

2:12And the answer is, it depends.

2:15I don't know.

2:16So when you start talking about estimates,

2:18it's not just cost, it's not just time.

2:21There are other things involved.

2:23Let's clear the board and let's start to dig into this.

2:26Let's talk about time.

2:28Since time is money, I want to start here.

2:31I've estimated all kinds of projects

2:34largely in the telecommunications and IT space.

2:38And when I say IT space, I'm talking about IT stuff

2:43inside of a telecom switch.

2:45Servers, data migration, new equipment turn on, retirement,

2:51all kinds of stuff.

2:52I've seen the good, the bad, and the ugly, mostly good.

2:55But if I'm talking time estimates,

2:58there are a number of questions I like answered right out

3:01of the gate.

3:02For starters, who's doing it?

3:04Why does it matter?

3:05Well, if it's someone who's very experienced,

3:08they might do it faster than the baseline.

3:11And let's say the baseline is eight hours,

3:13because that's the normal amount of time

3:15you have in any particular shift.

3:17If you're working 12s, that number changes, obviously.

3:21But don't think for a second that you're

3:23getting eight hours of actual work out

3:26of any particular person.

3:27What about breaks?

3:28What about lunch?

3:29What about people who smoke?

3:31What about water cooler talk?

3:33What about doing other administrative stuff?

3:36Do you really think you're going to get eight hours

3:39of productive work out of every worker?

3:41Because unless you're using some form of automation,

3:44that answer is a big fat no.

3:46It's not going to happen.

3:48But let's just focus on experience level first.

3:51More experienced people get more work done.

3:54It's a fact.

3:55Less experienced gets less work done.

3:58And then you have mixed experience teams.

4:02So who's doing it really matters?

4:04Then you have dependencies.

4:07That is terrible writing.

4:09But you know what I mean.

4:10Dependencies.

4:11What are they waiting for?

4:13Are they waiting for someone else to finish their work

4:15so they can begin?

4:17Are there multiple people doing this work?

4:20Because you've probably heard that expression,

4:22too many cooks spoil the broth.

4:24Meaning you can't just throw a bunch of people at something

4:27and get work done faster.

4:29Because then you can have confusion

4:31and you introduce other risk factors.

4:34Other issues.

4:35Is there new equipment involved?

4:37New equipment means there's a learning curve.

4:39I can use a backhoe.

4:40I can use heavy equipment.

4:42It really only takes me about 15 minutes to five days

4:45to figure out how to use it.

4:48How much time do I have to learn how to use it

4:50before I actually have to implement its use?

4:53And it really depends.

4:54What about quality?

4:56And I already talked about that.

4:57I can get a ton of work done if I'm not worried about quality.

5:01But if there's the quality standard,

5:03I normally have to take more time.

5:05What about risk?

5:07Risk is a big deal.

5:10If I'm around energized equipment,

5:12and most people in the IT world are,

5:15unless they're writing code, there's an issue there.

5:18What about lockout, tag out procedures?

5:21Am I doing the work?

5:22Is a vendor doing the work?

5:24Is it in the maintenance window?

5:26Is it during the day in a live, hot environment?

5:30Because here's the real issue.

5:31You can work around energized equipment

5:33in the maintenance window

5:35and not worry about bringing down the entire network

5:38or taking customer information or applications offline.

5:41But if you're doing it in the middle of the day,

5:44during the heaviest data usage time,

5:46that's a whole 'nother animal.

5:48You understand my point.

5:49I'm comparing apples to apples

5:51and I'm throwing in some oranges here,

5:53but there are a lot of issues related to time.

5:56And there's one more big factor.

5:59Tolerance.

5:59Your stakeholders are gonna have a very different risk tolerance

6:04than you might, or maybe yours is higher.

6:07Different risk tolerances change the amount of time

6:10and the environment in which you're trying

6:12to get this work done.

6:13All of these factors are in play

6:16as you begin to estimate the amount of time

6:19it takes to get work done.

6:21So I promise you, if you feel like this is easy,

6:25if you feel like you're off and running

6:26and it's no problem,

6:28you're probably walking yourself into a trap.

6:30So don't do it.

6:31So now that we've talked about time a little bit,

6:34let's clear the board.

6:35Let's wander over to the cost issues.

6:39Let's talk about cost for just a moment.

6:41Right out of the gate, what's the rate?

6:43Different people doing different work

6:45have different cost rates.

6:48I know what my hourly rate is.

6:50A vendor's rate might be different.

6:52Someone who's brand new might have a different rate.

6:54So you need to figure that out.

6:56And then which schedule are you using?

6:59Is it normal eight to five?

7:01Is it over time?

7:02Is it the weekend, which could be double time

7:05or holiday?

7:06I once spent eight hours in a sell site on Christmas day

7:09and I got paid double time and a half.

7:11Don't think my boss was happy about that pay,

7:14but he was happy that I was able to fix the issue

7:16and get the network back up to speed.

7:18What about resource availability?

7:20Their availability can mean a different cost.

7:23It also impacts the schedule.

7:25What about material?

7:26Material costs matter as much as the resource

7:29actually doing the work.

7:30What about change requests?

7:32If you have a contract, there may be a penalty cost

7:36to actually doing this work.

7:37It just depends and I've seen that.

7:39What about inflation?

7:40That's a factor.

7:42And once again, risk plays a big role here as well.

7:45Yeah, that's gonna take me more time

7:47in the schedule to get some work done.

7:49But here I have contingency and then I have management.

7:54These automatically mean more cost.

7:57Think of contingency risk, additional money

8:00in this particular case,

8:02as additional funding for known or identified risk is,

8:06as I like to call them.

8:07But then you have management reserve,

8:10additional funding for unknown risk.

8:13These all factor into your cost estimates.

8:17There is a very clear, dynamic,

8:20indisputable relationship between time and cost.

8:24And now that I've brushed over, excuse me,

8:27just a couple of the questions that if I were an estimator

8:30that I would begin to ask,

8:32I wanna start to dig into which estimates we could be using

8:35based on where we are in the project.

0:00But throughout my project management career, plenty of people have come to me

0:04and said,

0:04"Look, we know we have a lot of estimating tools.

0:08Which one should we use?

0:09What's best?"

0:10And the answer is, I don't know.

0:12It depends on each project, each deliverable, each team's experience level,

0:16each team's

0:16tolerance level.

0:18There are a ton of variables here.

0:19But what is known is you have a lot to choose from.

0:23So let's take a look at it.

0:24The first one, and one of the best ones, is analogous estimating.

0:29Big assessments are based on something you've previously done, something

0:33similar.

0:34For example, and we're going to go over these again a little bit later on with

0:37some examples.

0:39But in this example, let's say you had a previous server migration project and

0:44it took you two

0:45months and it cost $100,000 just as arbitrary numbers.

0:48Well, if the scope will work as similar on this one, then you can probably

0:52start estimating

0:53right there.

0:54And it's just a start though.

0:56It's not the end all, but it's a good start.

0:58Another good one is parametric estimates.

1:01Parametric estimates are usually x times y equals something.

1:05For example, if I know I can build 10 feet of fence an hour and I have 200 feet

1:10of fence

1:11that I can safely build, I simply divide one by the other.

1:16200 divided by 10, approximately 20 hours.

1:19And then it's a question of how much time do I have based on availability?

1:24X times y equals something.

1:26This one's one of my favorites and it's really good for things like equipment

1:31rentals, times

1:32days and weeks and months, salaries, stuff like that.

1:36If I know everybody's salary and I know how much time you're working with me on

1:40a week,

1:41it's pretty easy for me to calculate what your cost is going to be.

1:44Well, then you have bottom up estimates.

1:46Bottom up estimates are slow and detailed.

1:49Now riddle me this.

1:50The final guess is fast, then this must be slow, which makes you wonder, where

1:57are we

1:57doing these?

1:58Well, probably in planning.

2:00And we're going to talk some more about this.

2:03Early planning, also planning.

2:06There is no one size fits all and there is no one in done.

2:10You can use them over and over and over as you elaborate on the details.

2:17Another fantastic one is three point.

2:20Also known as PERT, program evaluation interview technique.

2:24We're going to have some great examples on this one.

2:27This is a good one to use, but this one will get you into trouble.

2:31Another one that will really get you into trouble and one of my favorites is

2:35expert

2:36judgment.

2:37An expert said so.

2:39Well, who said he was an expert and why is he saying so?

2:42We're going to talk some more about that.

2:44But since we're talking about experts, you can get into trouble and it's called

2:49group

2:49think.

2:50What I'm talking about, if you've ever been in a room, as someone has said, you

2:53know

2:53what?

2:54This would be a good idea.

2:55Let's do that.

2:56And everyone else who wasn't speaking said, yeah, that's a great idea.

3:02Meanwhile they're thinking, I'm just going to go ahead and go with what this

3:04person is

3:05saying because number one, I didn't have an idea.

3:08And number two, he's on the hook for it.

3:11So everyone raise your hand.

3:12Yay, I'm not in trouble.

3:14If you want to eliminate group think, there's another technique called delphi.

3:19And it's anonymous.

3:21Anonymous.

3:23I can get your opinion in an anonymous manner.

3:27For example, if I ask five people to write down their opinion on a piece of

3:31paper about

3:32which technique is best from one estimate to another estimate and they're not

3:36allowed

3:36to say it out loud, I'll get their honest opinion.

3:39I simply poll the results and I can see who said what.

3:42That's a great way to do it.

3:44So let's clear the board and keep going.

3:46Another good one is top down, which I said earlier, but I like to call this one

3:52directed.

3:53Meaning you were given a budget, you were given a time frame and now it's your

3:58job to

3:58fit everything in the middle of it.

4:00That one's a little more difficult.

4:02And now you have a constraint.

4:03And don't forget about the triple constraints.

4:06Scope schedule cost.

4:07As soon as they say can only cost this much, now this is impacted and possibly

4:12how much you

4:13can actually produce is impacted.

4:15Transfer bids is also great in the form of analysis.

4:19Sometimes it may cost you a little bit more to actually use vendors, but it's a

4:24fantastic

4:25risk strategy called transference.

4:28It's pretty easy for me to put some sort of contract language in there that

4:33says if you're

4:34doing an IT upgrade or a cut over and you bring the network down, there's a

4:38financial

4:39penalty.

4:40Or if you don't have it back up in time, there's a financial penalty.

4:44Or if we get sued, we simply transfer that to you.

4:48Great way to do it.

4:49And because of risk, you also have reserve analysis.

4:53You must analyze how much additional money you need based on the risk.

4:58More risk, more money.

5:00Less risk, obviously.

5:02Less money.

5:03These are just a couple.

5:05These are not the end all be all.

5:07And since we have a few of them, let's talk about when we might use them.

5:11When to use, that is the question.

5:14Very early on, when the project is initiating, this is very early.

5:21What do we know?

5:22Very early.

5:23We know very little.

5:25And because we know very little, we need to use tools that take advantage of

5:30that.

5:30Expert judgment.

5:31That's an easy one.

5:33Analogous, something similar.

5:36And if it's very early, it could be top down directed.

5:39One day one, they could say, "This project is due August 1st, and you can't

5:43spend more

5:43than $1 million, just as an arbitrary number."

5:47So initiating is very, very early.

5:50Then we get into planning.

5:51Does that mean we can keep using these tools?

5:53Absolutely.

5:54But we have better tools.

5:57We know more, I hope.

5:59And when we know more, we can move on to three-point estimates.

6:02Or pert.

6:04We can also move on to bottom up.

6:06Think of it this way.

6:07If I know the cost of the wood to build a house, the framing, the drywall, the

6:13electrical

6:13cost, the plumbing, the shingles, all of the components, and I know the labor

6:19cost to do

6:20it.

6:21I can quickly aggregate it, sum it all up to get me to a total cost.

6:28Bottom up is slow, but it is usually very accurate, which is why you're doing

6:32it later

6:33in planning.

6:34Vendor bid analysis can happen in planning as well, normally nearing execution.

6:40And the reason I say that is because most vendor bids are only good for 30 days

6:44.

6:44Otherwise they'll change them.

6:46Look at the cost of material.

6:47Look at inflationary costs and stuff like that.

6:50You also need to worry about reserve analysis.

6:53The majority of risk occurs in execution, which is why we need to know how much

6:58management

6:58and contingency reserve we need.

7:01All of this is done through something called rolling wave planning.

7:06It means we simply revisit and revisit and revisit.

7:11Here's another way to look at it.

7:12I try to plan my projects from four different levels.

7:16I have the strategic level, usually very early.

7:20Then I have the operational level.

7:23What's the impact to the company?

7:25Can we support it because we have normal operations and we have project

7:28activities happening

7:30at the same time?

7:31Then I have activity levels based on people who are doing it, when they are

7:36doing it,

7:36and then I have task level.

7:39Here I'm worried about in one example construction.

7:42Yeah, I know the people that are doing it, but what about the material?

7:45And I'm not just talking about construction equipment.

7:48I'm talking about personal protection equipment.

7:50Gloves, boots, hard hats, vest, hammers, stuff like that.

7:55All of this stuff gets rolled into your estimates.

8:00All of your estimates get rolled into your budget.

8:04Let's take a pause for cause right there and when we come back, we're going to

8:08start digging

8:08into these individual estimates.

0:00Okay, let's dig into some of these estimates and right out of the gate we're

0:05going to talk about

0:05initiating. Initiating is very early in the project. Think of it as the project

0:11has basically just

0:12come to life. It's got a heartbeat. What can we do? Well, the first thing we

0:17can do is look at

0:18analogous estimates. How similar is this project compared to what we've done

0:23before? And I've got

0:24a great example coming up. Well, then we have experts. I have been called an

0:29expert and it scares me

0:31every single time because trust me, when you're an expert, you can't get it

0:35wrong too many times

0:36and they strip that expert title right away. And then you just go back to above

0:40average, which,

0:40you know what, I'm pretty comfortable there. Well, then you have top down

0:44directed. Basically,

0:45someone told you this is it right out of the gate. You have a constraint, but

0:49there are things you

0:50can do about it. And then I have a bonus resource we're going to get to in just

0:55a moment. So let's

0:56clear the board and walk down the road of awesomeness. Let's look at an

1:02analogous estimate. So it's

1:04based on something you've previously done. So let's look at something you've

1:08previously done.

1:09Let's say the HR department came to you and said, Hey, we want an internal

1:13knowledge database for

1:14the HR team using confluence. And you said, Okay, we can do that. So you set it

1:20up, you created 25

1:22pages, there was customized layouts, you trained 10 people. And here's the cool

1:27part. It only took

1:28two weeks with one team member working full time. So not too overwhelming. So

1:34that was the baseline.

1:35That's the past project. Well, someone else came to you and said, You know what

1:40, you did such a

1:40bang up job. We'd like you to do it for finance. And you said, Okay, same thing

1:46, similar setup, 25

1:48pages, similar layout, only this time there's eight users. You can expect two

1:53to two and a half

1:54weeks with the same team member. So the first one took two, we've got the same

1:59team member. So

2:00that gave us the benchmark. Why are we doing that? We're reusing the page

2:05templates in the branding,

2:07if we can, because we certainly don't want to reinvent the wheel. And finance

2:12may need a little

2:13more customization. So we're adding a small buffer. And that's why you're

2:17ending up with slightly

2:19larger schedule. That's how an alic access to making works. And you're probably

2:24thinking, Well,

2:25how similar is it? And the answer is, I don't know. But there is a way out of

2:30it. So let's take a

2:31look at it. Let's add a twist. And here's the twist. They're all remote. What

2:37do we do about that? Well,

2:39it depends. How big of a deal is it to have remote users? And what's the impact

2:43of the schedule? And

2:45if there's an impact to the schedule, what's the impact to the budget? Luckily,

2:48we have the same team

2:49member again. That may not happen very often. And based on the success we've

2:54had before, we're only

2:56going to add an additional week. And that's because remote coordination, it

3:00really comes down to

3:02similarity. And you can't put a square peg in around hole, folks. Don't say,

3:07Well, this IT project

3:09is very similar to this construction project, because we're going to have the

3:13same person running

3:14it. That is not similar enough. And if you're looking for a cool tool to help

3:18you stay out of

3:18trouble, I have one for you. So let's take a look at it. As soon as I figure

3:22out how to do it,

3:23ah, here it is. What you need is a similarity scoring model. And this becomes

3:30very important

3:31because a whole bunch of people need to agree on this thing and sign off on it.

3:36Here's a great

3:37example. Now, these are random weights. I came up with them. I came up with

3:42these components. And

3:44I just came up with a random example. So let's take a look at them. Is it the

3:47same lead? That's

3:49very important, particularly if it's someone experienced. Is it the same tech

3:54stack? Are you

3:55reusing templates or reinventing the wheel? Never reinvent the wheel. The

4:01problem is, people take

4:03that wheel and they put it someplace on their hard drive and they forget they

4:07have it and they

4:07never share it with anybody. Don't do that. Share the wheel. Are the

4:11requirements similar? Easy peasy,

4:14but in this case we're going to say no. Is it the same number of users? This is

4:18really important.

4:19It's rarely the same. Is it the same team doing the work? Also very important.

4:25Anytime you have a

4:26new team member, particularly with Scrum, there's usually a loss of

4:30productivity while that team

4:32member integrates into that team. Not always, but quite often. Are there the

4:36same security needs?

4:38I don't know, but could be. And is it the same integration? In this example, we

4:44had two nodes,

4:45so we basically took the weights, we moved them over here and we added them up.

4:50And then we had a

4:51score that everyone agreed to prior to our using this. In this case, it looks

4:57like we can use it

4:58with some adjustments. Again, you need to do this without incorporating and

5:04injecting bias into

5:05your scoring mechanism. So if you create this on your own, make sure you float

5:11this up to leadership

5:12and get buy in across the board. Do not keep it in house into your own

5:17department and then

5:18surprise everybody else later on, because I promise you, they're just going to

5:22push back.

5:23They're not going to do it. It's easier for you to create this and ask for

5:27their buy in. Excuse me,

5:31then it is to ask everyone to create something together that will take far too

5:37long. Do not do

5:38that. Create it, float it, make some adjustments, defend yourself, justify, get

5:45everyone to sign

5:47off on it, and then use something like this on a small project first. Do not go

5:53crazy and try to

5:54implement it across the board because you will lose your mind and people will

5:58push back to the

5:59status quo, which is really easy to maintain. I love this tool. Come up with

6:04one. And there's one

6:05more point here, folks. 80% of what you do is going to be very similar. So

6:10there's no reason for

6:11you not to have one of these. It's that last 20% that's going to require some

6:15movement to the left

6:17into the right to get buy in. Those one offs are no big deal. Aim for the 80%

6:22and let the 20%

6:24work itself out. This will save you a ton of time. So let's look at one more.

6:29As I figure out where

6:30my cursor went, let's look at a few more tools here. Well, expert judgment.

6:35There's a reason we

6:36have experts. They've been there and done that. They have experiences outside

6:41of your company.

6:42And here's the cool part. They are not concerned with the politics and the

6:48internal environmental

6:49factors. That's why you need to bring someone from the outside in for really

6:54big changes. If it's

6:55an enterprise level change, you should have an expert. But when you get them,

7:00ask them what other

7:01projects have you worked on that are similar to this and what was the cost?

7:05Make them justify.

7:06If you're not doing that, they could make anything up. And there are plenty of

7:09people out there

7:10calling themselves experts. I'm one of them, but I'm not going to show up

7:14without past performance

7:15approved you. I know what I'm talking about. Other things, is it feasible? I've

7:20had people ask me

7:21about projects before and it's completely unrealistic. And you have to be able

7:26to tell them that and

7:27you have to be able to hear that. What are the typical risks involved? Remember

7:32, when you're inside

7:33the business, you're not seeing what's happening outside. And so you can be

7:38looking at the world

7:39through what they call rose colored glasses. It looks fantastic until you take

7:44the glasses off.

7:45And then it looks like chaos. Everything is burnt to the grounds. Experts have

7:50a different

7:50point of view. If you don't have a solution, ask them what they recommend.

7:55There's a reason you

7:56are paying them. Get their opinion. And there it is again, I forgot. What is

8:00the solution you

8:01recommend? If it's completely bonkers and it's outside your budget, it's just a

8:06recommendation,

8:07folks. I've seen companies bring in two or three different experts and get

8:12their opinions.

8:13It's worth the money spent upfront to prevent the money you're going to pay

8:18later on reacting

8:20to all the problems that came with it. So think about it. Other issues you

8:24might have,

8:25top down directed constraints. Is it the budget or is it cost? It's almost

8:31always one of those.

8:32And as soon as you have a constraint, one of them that normally means you have

8:37flexibility.

8:38Think about your triple constraints. Scope, schedule, cost. If you tell me,

8:44cost is the primary

8:45constraint, that means I have flexibility in the other two. What I don't want

8:49to hear is, well,

8:50all three are as important as one another, because that's just not true. I've

8:54never seen it in 30

8:55plus years of project management. There is always flexibility. What I prefer to

9:00hear is this one.

9:01The scope is absolute. It must be delivered completely, no matter what, which

9:07means right out of the

9:08gate, you're going to give me more money to do it. You may not believe it right

9:12now, but you will

9:13later on if the scope is most important. That's a fact. So if it is cost, that

9:19means scope is

9:20immediately impacted, which means you need to ask, what are the absolute must

9:24have deliverables

9:25and what are the nice to have? You know what the difference is between wants

9:29and needs, cost.

9:31No matter what, why is it the budget or why is it the schedule or the scope?

9:36You need to find out

9:38why they need to give you a concrete reason. And usually when you figure that

9:42out, you can begin

9:43to understand where the flexibility is. If they don't know, that means it's all

9:48up in the air and

9:49they just made it up. How did you get to these metrics? How did you get to this

9:53date? These are

9:54very important. Project managers, as it relates to estimating, needs to be able

10:00to ask hard questions.

10:02I feel like I misworded that, but do not be afraid to ask hard questions. The

10:08best project

10:09managers I know are impersonal. They are simply gathering information so they

10:14can draw conclusions.

10:16And if there's a fixed budget, are there pre-approved vendors and licenses we

10:20need to be aware of?

10:21Because if there aren't, this is going to be a problem. And if it's a

10:25predetermined budget,

10:27did someone already build in risk reserve? Risk reserve for contingency, for

10:32known risk,

10:34and management reserve for unknown risk is incredibly important. As soon as you

10:39do not have

10:40risk reserve built in, you are going to have a risk. And a risk becomes an

10:45issue. An issue is a

10:47problem. And a problem is something that must be solved. These are hard

10:50questions and I recommend

10:52you have a list of them when you interview your key stakeholders about any of

10:56these estimate

10:58issues. So one more. Whoops, I did the wrong thing there. This is one of my

11:03favorite estimating

11:04tools. Historical information is the best source of estimation that you can

11:10possibly have. What

11:11happened before? What was good? What was bad? Between the original and the

11:16final, you are always

11:18looking for the variance. What was it? Why did it happen? If you can figure

11:24that out, you can figure

11:25out how to prevent it or do it again. And the only kind of variance you want to

11:29do again is a

11:30positive variance. And that doesn't happen very often. But when it does, it is

11:35spectacular. So ask

11:37these questions. And here's the final one. If we had a negative variance and

11:41there were lessons

11:42learned, did we implement any of those changes? Because if we didn't, whatever

11:48happened before,

11:49it is absolutely going to happen again. So initiating is your opportunity to

11:56ask the hard

11:57questions. And if you're not doing that, these problems are going to follow you

12:02deep into planning

12:03and all the way into execution. And if you get into execution and you have

12:09questions unanswered,

12:10if you have magic floating around your head, if you're hearing, we'll just

12:15figure it out and

12:16cross that bridge down the road that bridge is going to burn beneath your feet.

0:00Now that we've taken a look at some initiating estimates when we know very

0:03little,

0:04let's take a look at some planning estimates when we know more.

0:06Let's start with parametric.

0:08Usually means x times y equals something.

0:12Bottom up, which is slow, but usually very accurate.

0:16And this typically happens deep in planning.

0:18Three point, also known as pert.

0:21There's the formula.

0:22P plus 4r plus o divided by 6,

0:25pessimistic, optimistic, realistic,

0:29sometimes called most likely.

0:30Then we're going to look at standard deviation of our three point estimate,

0:35as well as reserve analysis related to risk.

0:38So hold on, here we go.

0:40Let's start with parametric.

0:42And because my writing is sometimes questionable,

0:45I figured I would just go ahead and map this out for you so you can follow

0:48along.

0:48But don't think it's going to be easy.

0:50This is going to get complicated pretty quick.

0:52Let's take a look at it.

0:53Your company needs a new support portal.

0:56Okay, everyone needs support.

0:58There's a ticketing system of frequently asked questions section,

1:02some live chat, log in, log out stuff, and a dashboard.

1:06Approximately 12 functional modules,

1:09and it takes about 12 hours per module based on what you've been able to do so

1:14far.

1:14So far, this is pretty easy.

1:17So let's do the math.

1:18If we do the math, we come up with about 144 total hours.

1:23That is fantastic.

1:25But what about risk?

1:26What might happen?

1:27Maybe we should just go ahead and add 10% for any complexity problems

1:31or third party integrations, or anything else that may come to mind

1:35based on what you've previously done.

1:37Let's just go ahead and add it.

1:38Let's do the math.

1:40And when we do the math, we come up with about 158.4 hours.

1:45So let's just go ahead and round that up because no one works 0.4 hours.

1:48That's just ridiculous.

1:50Now here's where it gets interesting.

1:51What about the skill level of the person doing it?

1:54Are all of your employees equally as skilled as one another?

1:58Because I'm betting the answer is probably not all the way to a big fat nope.

2:04So let's take a look at what to do about it.

2:06We have a skill multiplier.

2:08Let's say the baseline is eight hours, just as a base.

2:12And it would be fantastic if any employee could be productive for a full eight

2:16hours a day.

2:17But I'm going to leave that for another time.

2:19So based on what we know about the productivity of our employees,

2:23our senior levels has a multiplier of 0.8.

2:26Well, what does it mean?

2:27They do work faster than the baseline.

2:30And they should.

2:30They're very experienced.

2:32So let's do the math.

2:34160 hours down here times 0.8.

2:38Let's see, 0.8.

2:401.28.

2:41Put the decimal.

2:42It takes approximately 128 hours for our senior most person to do this work.

2:48Why?

2:49Because they've proven that they get it done faster than the baseline.

2:53If we go to the mid-level person, it's 1x.

2:56It's going to take them approximately 160 hours.

2:59That makes sense.

3:00But what happens if we assign a junior person to it?

3:03It's going to take one and a half times.

3:05Let's take that 160 hours, split it for 80.

3:10And if my math is right, I think we're up to 240 hours now to do this work.

3:15And what about the rookies?

3:17People who are shadowing.

3:19People who are at the very beginning of their career.

3:212x?

3:22This is real stuff, folks.

3:24And if you're not paying attention to skill multipliers,

3:27you're going to have over-estimated or under-estimated estimates all the time.

3:35But what you're really doing is setting a false expectation on when this work

3:41is going to get done.

3:42And it's going to have a cascade effect all the way through your project.

3:45But this is not the only issue.

3:47Let's drive forward.

3:48What about availability?

3:50And you're probably thinking, "Well, wait a minute.

3:52They work 40 hours a week on just this?"

3:56Let's say you have a shared resource.

3:58That senior person with a productivity multiplier of .8,

4:02they're going to be in demand.

4:04And we know they can get 160 hours of regular work done

4:07in about 128 if memory serves.

4:10So let's just go back to 128 hours.

4:14Well, what if this person is a shared resource and is only available 4 hours a

4:18day?

4:19Let's take a look at it.

4:20128 divided by 4, there's 3, 12, 2.

4:26If they're only available 4 hours a day,

4:29the activity duration for them to do this is 32 days.

4:34It's freaking me out too.

4:36And I've seen this a hundred times.

4:38But it doesn't just stop there.

4:40Let's drive on and throw some more problems at this.

4:44Can your resource sustain that pace?

4:48Most of them can.

4:49Some of them can't.

4:50What if they get derailed with other higher priority stuff?

4:54You've just added time to that activity duration.

4:56What about upcoming vacation or business travel?

4:59Happens all the time.

5:00What if they get sick?

5:02Okay, well, let's look at the other end of it.

5:04What if we just add more resources?

5:06That's fantastic.

5:07You probably should.

5:09You should probably just start with more resources.

5:11Or ramp up along the way.

5:14There are a ton of reasons why level of effort and activity duration

5:20and availability and skill productivity are in constant conflict.

5:26And if you're not paying attention to these,

5:28you're going to have overly optimistic estimates

5:31that simply aren't going to meet the mark.

0:00It's time to talk bottom up estimating.

0:02Bottom up estimating is estimating from the bottom up.

0:06It's slow, but it's very detailed.

0:09And you can only do this when you know a lot of information

0:12or you can figure it out.

0:13These are the result of brainstorming sessions

0:17and walkthroughs.

0:18So let's take a look at an example.

0:20In our example, we're going to set up a new IT help desk space on site.

0:25Easy enough.

0:26Let's go to a workspace and figure out what we're going to do.

0:30So what are we doing?

0:31Well, it depends.

0:32Let's start with workstation setup.

0:35What might we do to set up the workstations?

0:38The places where the IT peoples are going to sit.

0:41Well, we need to order some PCs.

0:44And when they get here, let's put them together.

0:46Assemble.

0:47Let's call that three hours.

0:49And you may look at this and go, no way, dude.

0:51I can totally order PCs and set them up in 15 minutes.

0:55Well, if you can prove that, that would be awesome.

0:57But for now, let's just stick with three hours.

1:00And let's configure those accounts.

1:02Let's call that six hours.

1:04And again, these are arbitrary numbers.

1:07Hopefully, we have some supporting information to back this up.

1:11Well, let's connect them.

1:12They're not going to be much good if they're not connected to all of your

1:15systems.

1:15Let's call that three hours.

1:17And then we have to install the software.

1:19Actually, it's ticketing software.

1:21I was trying to be very specific.

1:23Let's call that just two hours.

1:25And then we have QA, Quality Assurance and Testing.

1:28Let's call that three hours.

1:30So we've walked through all of this just to order the workstations and set them

1:34up.

1:35We come up with 17 hours.

1:37That's step number one.

1:38Everyone write that down or I'll have a summary forward shortly.

1:42So let's clear the board and keep going.

1:44Once we get them, what about the physical space setup?

1:47What does that entail?

1:49Well, where are they going to sit?

1:50We need to order some desks, some chairs, some monitors.

1:54Well, there's a cost associated with that.

1:57Let's call it three grand.

1:58That seems pretty reasonable.

1:59What about cabling and network drops?

2:02Let's just call it cabling.

2:03We need cables.

2:04We need some network drops.

2:06Someone's paying for that.

2:07Let's call that 1,200.

2:09Whoops, not 1,000.

2:11What about setting up the Wi-Fi?

2:13Let's just call that four hours.

2:15Do we have signage?

2:16I have no idea.

2:17But let's call it two hours.

2:19And the last one, and this one might surprise you, is PPE.

2:24Personal protection equipment to actually do this,

2:26like gloves, hard hats, boots, maybe some knee pads.

2:29But what about mats?

2:31Chair mats.

2:32Let's call it another 150 bucks.

2:34So where are we?

2:36We have six hours.

2:37And let's do some math here.

2:39$4,200.

2:41All right, now we're cooking with Greece.

2:42Let's clear the board and drive on.

2:44The next thing we have to do is training.

2:46We have to create guides and SOPs.

2:50Standard operating procedures, four hours.

2:53We also have onboarding.

2:55Yes, I apologize if my writing looks very fast.

2:58It is very fast.

2:59But I'm trying to slow it down.

3:01I've had a lot of coffee, but I'm always doing something very fast.

3:04What about the ticketing walkthrough?

3:06And if you've ever worked with a ticketing system,

3:08it normally requires some kind of walkthrough.

3:10How to create, how to escalate, all of that stuff.

3:13Let's call that another two hours.

3:15Well, then once we have the IT people sitting there,

3:19we have shadowing.

3:21We want to make sure we have someone on site

3:24to support them while they go through their learning curve.

3:27So let's call it a shadowing period.

3:29Yeah, you know what I mean.

3:30Shadowing period.

3:31Eight hours.

3:33That's one day for three days.

3:35Maybe one day a week for the next three weeks.

3:38Or Monday, Wednesday, Friday.

3:39I don't know.

3:40Whatever makes the most sense.

3:42And right now, that looks like 24 hours total.

3:45We're sitting at 34 hours.

3:47Again, these are arbitrary numbers just for this example.

3:51If it were me, I likely got a couple of people together.

3:54We went to the empty space and we walked through this brainstorming,

4:00putting up little post-it notes along the way.

4:02And then we got together and aggregated all of this

4:05down to something that made sense.

4:07So let's keep going.

4:08What about security?

4:09You ever seen an IT space where anyone could just walk up on it?

4:12Yeah, I mean either.

4:13Those folks are usually secluded.

4:15Just like engineers.

4:16We put them someplace where they don't have to worry about anyone else

4:19because they're doing sensitive stuff.

4:21So we have badging.

4:22That's an hour.

4:23Then we might have policy and firewall updates.

4:26That looks terrible.

4:27I'm just going to rewrite that.

4:29I'm sorry about that.

4:30The other part of my brain took over.

4:32Firewall.

4:32That's what I was trying to say.

4:34Let's call that two hours.

4:35And then we have asset tagging.

4:37Can't have someone just walking off with our stuff, can we?

4:40And the answer is no.

4:42So we have another two hours for a total of five hours.

4:46There's a lot involved.

4:47And we're almost done.

4:48But what about all the project management that goes into it?

4:52And there is some.

4:52Well, what kind of project management?

4:54Well, how about meetings?

4:56Coordination.

4:57That's at least four hours.

4:59How many meetings do we have on the project?

5:01All of them.

5:02How many should we have?

5:03None of them.

5:04But I digress.

5:05What about status reporting?

5:07Let's just call that two hours.

5:08And that's being generous.

5:10And then let's add a buffer just because 6.2 hours.

5:15Where did we get that?

5:16Some arbitrary known risk events.

5:19And we just added 5%.

5:21Just as an example.

5:22So now we're at 12.2 hours.

5:25Well, what do we do with it?

5:26Well, we add it up.

5:27So let's do that.

5:28I've taken the liberty of adding it all up over here.

5:32And here's where we end up.

5:33You got the workstations.

5:35That's the wrong pen.

5:36And how did that happen?

5:37Got the workstation set up.

5:39Physical space.

5:40Training.

5:41Security.

5:42Overhead.

5:43And we had our cost of equipment and PPE.

5:45Personal protection equipment.

5:47We end up at 74.2 hours.

5:50Let's just round it up to 75.

5:52And $4,350.

5:55This is our bottom up estimate.

5:58Now here's the beauty of this.

6:00We can take these activities in these costs.

6:03And push them up a level into a work package.

6:07Which is part of scope.

6:10And we can go to the boss and say,

6:12"Here is our deliverable."

6:14This could be part of a larger company stand up.

6:17You might have a finance section,

6:19an IT section, an engineering section,

6:22a visitor section.

6:24There's no end to what you can do here.

6:26And we can keep adding these work packages

6:29and add them all up to get an overall project amount

6:33and an overall project schedule.

6:35That is the beauty of bottom up estimating.

6:38But again, you need a ton of information to make this happen.

6:41If you don't have it,

6:43you simply use progressive elaboration

6:46and rolling wave planning.

6:48Rolling wave planning is where you plan for the things

6:51that are closest to you.

6:53If you have to begin work,

6:54and most of the time we need to begin work much sooner than we should.

6:58So with that,

6:59let's take a pause for cause.

0:00It's time to test your knowledge, knowledge, knowledge.

0:04Okay, that's really me making the echo, but here we go.

0:07You have two resources, a senior IT engineer and a junior.

0:11There's a skill multiplier here and a cost for each one.

0:14Their availability is listed here, two hours for the senior, five hours a day

0:18for the junior.

0:19They have 100 hours of work to complete and each work day is eight hours.

0:25So here's what you need to calculate.

0:27Option one is how long and how much for the senior engineer to do it.

0:32Option two is for the junior engineer and option three, the toughest one of all

0:38,

0:38maybe, is how long and what cost for them to do it together.

0:42Good luck and may the force be with you.

0:00So how did we do? Loved it, no worries, or that was awful, and we just aren't

0:05friends anymore.

0:06Well, let's take a look at it. Another quick review. We've got 100 hours of

0:10work,

0:11two resources, a senior IT engineer with a different skill multiplier and rate

0:16compared

0:16to the junior and his rate, or her rate. We also have their availability over

0:22here, which I hope

0:23you picked up on. If you didn't, uh-oh, let's walk through it together. Here we

0:28go. Let's take a

0:29look at option one, the senior. That is not, let's get rid of that, the right

0:34one. Let's take a

0:35look at the senior. Look at the skill multiplier, not multiplier, that's a

0:39totally different issue.

0:41There's 100 hours of work with the skill multiplier. 100 times 0.09, it's going

0:47to give you 90 hours.

0:49Why? Because they're awesome. There's also a total cost times 55. That's going

0:55to give you a price of

0:57$4,950 because they're $55 an hour, there's 90 hours of work, easy peasy, and

1:05based on availability,

1:07it's going to take you 45 days to get this done. So we end up with 45 days, $4,

1:13950 for option one.

1:16Take a look at that for just a moment. If you didn't figure it out, hit pause

1:21for cause,

1:22take a look at it. If my math is wrong, well, I apologize, it shouldn't be, but

1:27you never know.

1:27So let's clear the screen and move on. Let's take a look at option two, and I

1:32've captured option one

1:33right here. Option two is the junior. Different skill multiplier, 1.2, not

1:39quite as proficient.

1:41If you take 100 multiply by 1.2, you get 120 hours. They have a different

1:48hourly rate. That's

1:50going to give us $3,960. Three, nine, 60. And how long is it going to take

1:57based on availability of

1:59five hours a day? You divide 120 by five, you get 24. So now we're actually

2:05getting it done

2:06faster, and we're getting it done cheaper. So not too tough. Options one and

2:13option two

2:14were pretty straightforward as long as you caught their availability, and as

2:18long as you caught

2:19their skill multiplier. But let's take a look at option three, because this one

2:24might cook your

2:24egg a little bit. Option three is we're using them both together, and this is

2:29where things can get

2:30really weird. This is where productivity really comes into play. If you look at

2:36the skill multiplier,

2:37for the senior, it's 0.9. And if you divide their availability by 0.9, you end

2:44up with 2.2 hours

2:47of productivity based on availability. This is their output. Okay, I got that.

2:55So then we do the same

2:56thing for the junior. He's got five hours a day divided by productivity or

3:02skill multiplier, if you

3:04will. That gives you 4.17. Because the junior is less efficient, the junior's

3:11output is obviously

3:12less. And if you add them together, you get 6.39 units a day or hours a day of

3:21workout put. Well,

3:23we know they're available seven hours a day. And if you divide 100 by 14 point

3:29or 100 by 7,

3:30excuse me, you end up with 14.29 days to get it done. So let's just put that

3:36over here and clear

3:38some of the rest of this. Because figuring out how long it was going to take

3:42was the easy part.

3:44Figuring out what it's going to cost, not quite as simple. So I'm using my

3:48magic eraser here,

3:49and we're going to figure this out. We know that the senior share of the work

3:54is 2.22. We also know

3:57that the total output for both of them was 6.39 hours of productivity a day. If

4:07we move

4:08multiply that, let me just move this back over here. I apologize. 2.22, that's

4:14the senior's productivity,

4:15his output, divided by 6.39, that's total output times 100 total hours, gives

4:24us 34.78 hours that

4:28the senior is charging for, times 55. Are you staying with me? If you do that,

4:33you're going to end up with

4:35$1,913 and looks like 13 cents for the senior. If you take a look at what's

4:44left, 100 minus 34.78,

4:48you get what's left for the senior. Correction, the junior, which is 65.22. And

4:54if you multiply that,

4:56times the junior's rate, you end up with $2,152. And 9 cents. If you add them

5:03up together, you have

5:05$4,065.22 for option 3. And I hope you were able to stick with that. If not,

5:14hit pause, back up,

5:16take a look at it. But here's the real question you have now. Which choice is

5:20best? And the answer is,

5:22it depends. Is speed and getting it done most important? Or is cost? Only you

5:30can answer that

5:30with your stakeholders. I hope you enjoyed this. And when we get into the next

5:35skill,

5:35we will continue to raise the bar and do some more complicated estimates. Get

5:40ready for it,

5:41it's on the way.