Principles of Estimating

[rockstar.esq]

"So...how do you figure out how much it'll cost?"

There's a lot to it but everything builds on one concept; estimating is about systematically getting closer to the answer.  The most simplistic method is bounding the answer.  By defining the range that contains your answer, you've reduced the problem.  The next step is to determine what you need to know to reduce that range further.  As counter-intuitive as it may seem, asking yourself what you don't need to know can be very helpful.  The idea is to reduce uncertainty by systematically answering questions that divide the range sort of like playing "I spy". For example: a client asks for conceptual pricing on an office remodel.  The number of occupants and what paint colors they'd choose are irrelevant.  The square footage of the space won't change in a simple remodel and the cost of paint isn't typically driven by color choice.

 

Right off the top it's important to understand that it's very hard to remove all uncertainty.  Better design, or past-project similarity can help to reduce the uncertainty but some will always remain.  I like to think that estimating is actually about controlling risk rather than pricing stuff.  There are lots of ways to arrive at a price - heck you might even win a competitive bid by throwing darts at numbers.  But here's the thing that makes the estimating mindset different from an Entrepreneur.  It's never the job that you lose that puts you out of business, it's the job you win.  Look at it this way, the total bid amount is the company's minimum risk for not completing the job.  That risk goes down as the project reaches milestones, and only goes away entirely after the warranty period.  All the projects a company has underway have risk which added together amounts to running risk.  More than one company has had to drop everything to jump on a project that was going badly.  That can make every job suffer which is why it's important for someone to be thinking about this at the bid stage.  Every time I think about the risk to reward ratio in the construction market, my respect for the entrepreneurial spirit grows.

 

So how do you reduce risk?  As a bidder there are several approaches.  The most common is to define what is included and what is excluded from your bid.  Contract parlance refers to these as inclusions, and exclusions which appear on the bid proposal.  These can range from standards like "daytime working hours", to more project specific details like "excluding carpet on floor two". Remember they're called "General Contractors" instead of "Builders" for a reason!

Another way to reduce uncertainty is to put part of the work out to bid. Things go out to bid for several reasons.  The first and most obvious is to use market competition to keep the price down.  A  less obvious reason is to reduce risk.  Let's say that three subcontractors bid on a project.  The two low bidders are 3% apart.  If you win the bid with the low bidder amount and later learn they're missing something huge, or they back out, you can hire the 2nd low for 3% more which makes your minimum risk 3% for that trade.

 

A good principle of business is to have a policy of "the record is always on".  Anything you put in writing, you should expect to be saved and used later.   The subcontractor bids will have inclusions and exclusions on them.  Comparing them against each other is very illuminating.  It won't take long to see that exclusions are the embodiment of the expression "The devils in the details"!  I'll get more into reading bids which is called "Bid scoping" in a later post.

 

For now, it's important to see that risk is contained by knowing the spread (difference between bids), and  knowing the differences in the exclusions.  Sometimes the high bidder picked up on something significant that the competition didn't which spells disaster if you'd hired the low bidder.  Remember to call your clients attention to anything you've included that was tricky to see, or understand.  For example: the plans may show something is existing that you find missing during your job walk.

 

It's a terrific illusion that the construction documents will provide enough information to know every quantity, every time.  In the commercial construction world, the owners and architects expect the estimator to "make reasonable assumptions" often based on "standard means and methods" when a design fails to cover something.  The consequence of these expectations is the practice of stating assumptions via inclusions and exclusions on the bid form or proposal letter. Control risk by clearly defining what you are and are not including in your scope of work.

 

This brings us to the Estimators Paradox which is:

When you win you worry about what you overlooked, but when you lose, you worry it's because of something you shouldn't have included.

Next I'd like to cover a few principles of effective estimating.  It's hard work to count and measure everything on a project.  Conceptual or budgetary efforts for a client or an Architect are "free" services that consume valuable resources.  Many times historical data, allowances and minimal research will provide adequate accuracy for the purpose.  Having Subcontractors price conceptual work should be studiously avoided whenever possible.  Every bid should be retained for use as historical pricing.  I'll get more into how to track your files to make this easier in a later post.

 

Historical pricing is only as useful as your records, and your efforts to improve on what you've learned.  At the General Contracting level there is a tremendous range of acceptable detail  for estimating measurements called Quantity Take offs.  (QTO).  In my experience, a more detailed QTO is a more useful QTO provided the detail exists on the plans.  For example, If the plans resolution is to the nearest foot, there is no advantage to QTO's carried to the nearest inch.  While on the topic of inches, it bears mentioning to those inclined to the metric system that a decimal foot is a far more useful system for QTO's i.e.twelve feet, six inches would be notated as 12.5'.  No accuracy is lost and the spreadsheets are immensely simplified.

 

Measurements alone are not useful without showing how they relate to cost.  I'll provide some simple template ideas in a future post.  Speaking of units, the unit of measurement for materials aren't always obvious.  For example carpet is measured by the square yard whereas floor tile is measured by the square foot.  The "RS MEANS" series of books will provide valuable insight into both the units of measure, and what ballpark price to use.  There are other similar resources, but I'm most familiar with RS Means.  Beware of trusting one source implicitly.  There are many factors that must be adjusted to reflect the exact situation you're facing.  Anything that strives to be all things to all people fails on both fronts.  Get used to the idea that you'll have to use multiple references to check accuracy.

 

.

[Ratel10mm]

Excellent post, thank you. :)

[rockstar.esq]

Thank you Ratel10mm .  

[HIGHSIDER]

Top class advice, many thanks for sharing & the effort made here.

[engrick]

I tend to agree with everything you said as I estimate projects on the design side.  It is much harder to estimate when there is no design (as I do the designing).  Having a quantifiable design make estimating much easier as opposed to a client that "thinks they know" what they want. If you do not have something on paper, put it on paper and make sure it agreed upon.  

 

From there I tend to take two approaches - piece by piece and as a whole, along with some 'gut feel'.  Example would be  - a box of cookies should cost $1 for a box of ten(the whole/gut feel), but then I would figure a single cookie takes 1/2 oz of this, tablespoon of that, and 1 minute of mixing, 2 minutes of baking, electricity.....times 10 (piece by piece).  If the two come out to about a dollar I run with it.  If not, it is back to "what have I missed?" or "what have I over assumed?" Sometimes if they do not agree I make sure I document exactly what I am giving and not.

 

Documenting what you are going to give and not give is how some contractors give "less of a product" such as "I will provide 5 chocolate chips per cookie" where you will provide 10/cookie.

 

My nickles worth.

 

PS do not forget your overhead - this computer I am using was not free!

[rockstar.esq]

engrick,

 

You're absolutely right, it's important to use multiple methods to cross-check your estimates.  Square foot pricing can be skewed pretty severely by seemingly mild changes.  An office remodel for example could range significantly depending on whether they're using cubicle systems or framed office walls. Restrooms involve nearly every trade a project may require - adding or subtracting them makes a profound impact across the board.

 

Using large assembly/ unit costs like $X amount per office, $X amount per restroom allows a quick tally to approximate what's driving the project cost. Wherever you're using historical or conceptual data, the overhead and profit should be included in every cost component.  Stuff get's added and subtracted left and right, you can't afford to end up with a budget that doesn't pay.  I generally skew higher on conceptual pricing because most design teams view budgetary exercises as an insight into how much more money they can spend.  It's incredibly rare for them to consider showing restraint early on to pay for stuff they inevitably fail to mention.

 

Design-build bids are very difficult since bid clarifications ,exclusions, and inclusions define what you've designed.  Cheapskate clients love to take a well developed professional proposal to "Dog and a Truck" contractors looking for a discount.  It's a balancing act to protect your interests, encourage apples to apples comparison's and yet prevent your work from helping the competition.  Although, as I write this it occurs to me that the best protection from bid shopping is to decline invitations from shady clients.

 

It may sound crazy but when I bid Design-build, I don't price it conceptually.  I literally develop the design, solicit bids from engineers, and price the whole shebang by the deadline.  This stuff is way to squirrely to expect a job I won to go well if I don't have everything ready for prime time when I hand it over to Project Management. I have way too many bids going to accommodate constant interruptions from managers asking what I figured for every little thing.  That makes design-build bids more expensive, which in turn reduces the number of them I'm willing to consider pursuing at any given time.

 

Broke clients love design-build because they see it as a way to push change orders back on the bidders.  "You should have thought of that" is their battle cry. They never take responsibility for misleading narratives, omitted information, short deadlines, and small budgets.  About the only good thing I can say for design-build is that build teams hire the designers which gives them contractual leverage to get tight-budget, and short-duration jobs done without all the adversarial nonsense.

 

Of course, everyone being more professional is a better answer but perhaps it's best not to speak of things that may never be!

[Samuel51533]

Thank you for this post, I have had problems with estimating what my work is worth since I got started, thank you this well help 

[BeaverNZ]

This is really good stuff Thankyou

I have always kept all my job sheets for the last 16 odd years of working for myself and find them handy at times, The part of the job I always struggle with is the amount of labour to include in a job. I am fortunate to have established clients who trust me to provide the best outcome for a job they ask me to do alot of the work I do are one offs and the job often evolves as I think of a better way of doing the job as It goes. I am not all that interested in dealing with Joe public wanting stuff done right now for no money Cheers Beaver

[Brian Hibbert]

At my work, we use an estimating formula that I believe can be used for any manufacturing process.

Each job is broken down into steps (different options for the end product have more steps).

You might break down a project into cutting, forging, heat treating and final assembly.  If there are multiple separate component types to be made for a project they are each a separate step. 

Each step has costs associated with it. 

Setup fixed labor costs+ setup fixed material costs + setup labor per unit * number of units + Setup material per unit * number of units + run labor per unit * number of units + material per unit * number of units

Then add each step together for your total costs and multiply by your markup for the final price. If you want different markups for different parts of the job, do the markup calculation on each step before adding them together.  

Setup costs include the time and material needed to adjust machinery to get things right before the production run starts and any other costs that are per job rather than variable per unit.  In the case of blacksmithing, it might be the time and material needed to make a jig or other special tool for the job.  Run costs are pretty straight forward. They're the actual cost of the materials in each unit (including wastage) and the time it takes to do the step.   Note that some tasks can be pure fixed setup costs (0 for everything else) or pure labor costs have no setup costs, etc.  Just put zeros in where the costs aren't needed. 

This is a pretty flexible way to break down the cost of a project and come up with a close estimation of a price.