How accurately do you track your employees hours? I’ve been reading about what guys like Jerrald Hayes do for years and have always been in agreement. Currently we have 21 categories for production employees and 5 more for the supervisors.
The records are fairly detailed but still leave a lot to be desired.
I’ve been looking at the job clock system http://forums.taunton.com/tp-breaktime/messages?msg=41327.1
with interest. The main problem I see is I would have to reduce my 21 classifications down to about 5 or 6 at best. Ideally, (from the job clocks perspective) you would just record straight time.
So the question is, we spend a lot of time tracking, recording, assimilating, and applying job data. The problem is, as Sonny would proclaim, everything we build is a prototype!
I know that most, if not all, projects can be broken down into definable and predictable categories, but at what cost? I’m not a specialty contractor so the knowledge that crown takes x per cut, x per corner, x per lineal ft., x per etc… is not very useful for me.
Where is the line drawn? At what point do you realize that your going overboard with your data?
This is hard for me to verbalize, as I would love to plot everything on my spreadsheet and tell you to the 100th of an hour what every minute task takes to complete. That is my nature.
But I have another nature that really like profit. If I can boost our profitability (and at the same time make my job more simple) then sign me up!
OK, let me have it!
Jon Blakemore
Replies
Where is the line drawn? At what point do you realize that your going overboard with your data?
When you see eye to eye with Jerrald Hayes!!! ;~`)
Eric
I Love A Hand That Meets My Own,
With A Hold That Causes Some Sensation.
[email protected]
Eric,You've got guts. If anybody can write someone into the ground with HTML, links, white papers, databases, etc. it's Jerrald.If you get hammered don't blame me!
Jon Blakemore
I've actually spent some time with the man, been in his office, had some looooong conversations.
While that was tongue in cheek, it's nothing he would disagree with.
He IS a collecter of minutae!
EricI Love A Hand That Meets My Own,
With A Hold That Causes Some Sensation.
[email protected]
Eric,If you had Jerrald's database could you use it in your remodeling work? I think his company mainly does trim and millwork so he can focus a good bit.A guy like you who does it all, do you think you would ever be efficient with the numbers or would it be more trouble than it's worth?
Jon Blakemore
Jerrald has more in his data base than you know and he also has some product available for resale. Much of what he he has would be applicable to remodeling.
I love data, just not THAT much.
As a one to three man show, I feel I don't have a lot of use for it although I am sure that he would beg to differ and be more than happy to show and explain why.
Collecting the historical data first is the biggest part of getting a data base to function correctly for you. I think much of that data varies too much by region or operational technique to make it common enough for anyone to jump on board and start pushing buttons.
For a company with a sales staff, I would say it's a tool that is VERY useful, and needed.
EricI Love A Hand That Meets My Own,
With A Hold That Causes Some Sensation.
[email protected]
Where is the point of diminishing return?
You can record each task to the nanosecond. Measure materials with a set of micrometers but as you specify and document more you spend more time and effort recording the job and less time and effort actually doing the job.
IMHO nothing takes less than 15 minutes. It only pays to record actual physical progress. Which means the results of a task or job. All jobs include planning, setup of both tools and materials, actual work, replacement of tools and materials, cleanup and evaluation.
I have gotten a lot of flak saying this. Snickering that installing a wirenut takes 15 minutes. That is until they understand the larger issue. I had to find the fault that required the wirenut and that required I gain access to the house, get my tools on and grab some test equipment. I had to extricate the wirenut, assuming I had the nut on me, remake the joint, install the nut and inspect my job before tucking the wires back in and reinstalling the cover plate. Then I had to get the tools, belt and test equipment back on the truck. Document what had been done, write up a bill or otherwise record the job.
I use electrical work as an example as it is what I do. But the principle holds for any job in any trade.
Trimming baseboard on a chop saw is not just the five seconds it takes the saw to cut. Moving and setting up the saw had to be done. you had to size up and plan the baseboard installation. You had to measure. Carry the wood to the saw. Examine the cut. Perhaps make some adjustment. Test fit and nail it in place. At the end of the day you have to reorganize, clean up and document what was done.
The problem tradesmen have is they do so much by rote. Automatically. Without conscious effort and consideration. Which is why apprentices sometimes make good estimators. They see the individual steps because, for them, it is all still new.
IMHO it seldom pays to estimate in anything less than quarter hour blocks. Materials too need a large measure. Ten foot sections work well for electrical work. When in doubt round up. A three foot run of conduit is often as time consuming as a ten foot run. The fiddly bits eat up time on a scale disproportionate to their size. Registering these as larger sections for time makes up the difference.
I already round to the nearest quarter hour.Still, when you consider that we're not a specialty contractor we do a lot of varied tasks. A lot. I could compile a mountain of data if I really wanted to.I don't know that I could ever get to the point where I could remain on top of my data enough to be able to pinpoint exactly how long the trim for Johnson will take. I'm beginning to think if I can get within half days I would be doing well.
Jon Blakemore
I would think you could graduate the estimates in as course a set of unit as the customer will accommodate. Within reason of course. Knew a guy who once gave a verbal estimate along these lines:'We can wire your house for $8000. Or $10,000 if you want to know how many wirenuts we used.'What he didn't include is the number of wirenuts used would still only be an estimate. Anyone who demands to know how many nails you used, beyond being assured it beats code, probably doesn't trust you. Which means you might not want to work for them. Or at least you should charge a surcharge for aggravation.Quarter hour time blocks seems to work well enough. But half hour blocks might be smoother. As long as everybody is singing from the same hymnal you might get by with a handshake and a simple examination of the finished product and a guarantee that the joints won't open up for ninety days or whatever is customary.Electrical work, where most of it is hidden and performance based, demands a little more detail. But we also have a fairly strict set of standards in the code and inspectors in case we miss something.T&M jobs are the fly in the ointment. Still a matter of trust. Reference to the trade manuals for estimating, to make sure your within a median range seems like a good step. Ironically a master of paperwork can fight either side. He can document that a stalled job is within industry standards. Or he can show a efficiently run job is being sandbagged for profit.On a large job +/- half a day may be as good as it gets.
4,I'm not really interested in justifying our prices to the customer. We have not done cost plus to date and I'm not sure how often that will come up.Primarily I'm concerned about estimating future fixed price projects based historical data. The problem I'm up against is can my historical data ever be accurate enough to really be useful.And if so what is the cost of that accuracy?
Jon Blakemore
My opinion is that you should go for rough data, rather than spending a lot of time trying to collect and catalog the minutae. The minutae is only useful if the jobs are exactly alike and the same guys are doing the work. That's never the case, so it's easier and more accurate to deal in round numbers. For instance, I would not say... OK... it'll take me 6 hours and 43 minutes to do task X.... I'd say 8 hours. Cutting things too close leads to lost dollars, and isn't useful unless you're in a tight bidding situation, which no one wants to be in any way.
David,How do you break down your time records? I'm interested in the job clock system. I could maybe get 5 categories out of it. That would give me:demo/cleaning
framing
interior finish
exterior finish
???Could I get away with just these?
Jon Blakemore
Jon,
I mostly work solo or with one other guy so it's easy for me to keep track in some detail. I mostly use it forensically to see how I do compared to what I estimated, rather than trying to use data from the last job on the next estimate. The main thing I've learned is that I'm accurate on the actual work and often forget to include enough for supervision, ordering & delivery, mobilization & demobilization, etc.
I've supervised a lot of guys at once, and the challenge then is to get them to fill out timecards accurately. It's usually too much to ask.
Would the job block system let you change the categories as you change jobs? Why is it limited to 5? How does the employee tell the system which task they are doing?
So you don't really use historical data for estimating?Do you build the project in your head? Do you use unit costs?The job clock system consists of a reader that scans key fobs that all the employees have. The simplest would be a one green and one red fob, for start and stop of work. I could expand up to about 6 fobs (the stop fob plus a few categories), but any more than that and it would be cumbersome.
Jon Blakemore
Most of my jobs have new and different tasks and/or circumstances. I am quite accurate with my "it'll take about" method, and I estimate in round numbers (usually days) anyway, so it doesn't matter if I can hang a door in 37 minutes or 48 minutes or whatever. Admittedly, on a very large job I would be hard-pressed to be accurate, but on the smaller stuff, estimated in weeks, I come out fine.
As I understand it, Jerrald's work is trim and stairs. With a limited scope it would be easier to start collecting data and using it to estimate.
With the 50+ cost codes on my estimating spreadsheet it would be very unwieldly.