Looking for suggestions on best way to learn to prepare electrical estimates - outside of going to work for a contractor which would be best way. I'm an electrical engineer for a consulting company and have access to RS Means. I'm looking to better understand cost of my designs for personal knowledge and to accurately provide estimates for owner and to properly evaluate contractor change order requests. Any suggestions, helpful hints, recommended estimating handbooks are appreciated. Thanks.
how to learn estimating
- Thread starter DW98
- Start date
- Status
- Location
- Tennessee NEC:2017
- Occupation
- Semi-Retired Electrician
I'm not familiar with Mike Holt's estimating publications, but I've heard good things about it.
RS Means will only get you close on the very high side (if that make sense). Too many variables, like region, labor force available, volatile commodities, etc, that are all moving targets so a once-a-year RSM book can't keep up.
Any accurate estimating starts with knowing REAL costs. Real cost can only be determined by old school nut & bolt takeoff. Every material item with associated labor units (published), and expenses for a particular job must be taken off and priced.
This is your starting point. From there, you will develop known average costs for repetitive assemblies that will speed things up. Then you can count and enter most items. There is no guessing when it comes to feeders and lighting/controls.
I would suggest getting a book like MH's and learning those basic old school takeoff methods, then purchase an inexpensive estimating program to customize those assemblies and allow you to stay current on pricing.
I come from old school Kuntz method, so I estimate using the cost codes to this day. It's also how we manage our jobs.
Basically, you may not want to work for a contractor to learn, but you need to estimate like you do........if you want real accuracy. But then, there are some old tricks you will never learn unless you do work for a contractor. I can't tell you about those...haha.
Or...do like the architects/engineers do around here, release 20%, 30%, 50% drawings and let the contractors budget for free.
Good luck
RS Means will only get you close on the very high side (if that make sense). Too many variables, like region, labor force available, volatile commodities, etc, that are all moving targets so a once-a-year RSM book can't keep up.
Any accurate estimating starts with knowing REAL costs. Real cost can only be determined by old school nut & bolt takeoff. Every material item with associated labor units (published), and expenses for a particular job must be taken off and priced.
This is your starting point. From there, you will develop known average costs for repetitive assemblies that will speed things up. Then you can count and enter most items. There is no guessing when it comes to feeders and lighting/controls.
I would suggest getting a book like MH's and learning those basic old school takeoff methods, then purchase an inexpensive estimating program to customize those assemblies and allow you to stay current on pricing.
I come from old school Kuntz method, so I estimate using the cost codes to this day. It's also how we manage our jobs.
Basically, you may not want to work for a contractor to learn, but you need to estimate like you do........if you want real accuracy. But then, there are some old tricks you will never learn unless you do work for a contractor. I can't tell you about those...haha.
Or...do like the architects/engineers do around here, release 20%, 30%, 50% drawings and let the contractors budget for free.
Good luck
Gantt charts and PERT charts may help.
Also, there is a formula that pops up a lot, uses a weighted average and is sort of statistically based.
If S = the shortest time a task has been accomplished, A = average and L = the longest time, the likely time it will take in the future is
{S + 4xA + L}/6
so for S = 1, A = 5 and L = 15 your predicted time is 6 hours.
Also, there is a formula that pops up a lot, uses a weighted average and is sort of statistically based.
If S = the shortest time a task has been accomplished, A = average and L = the longest time, the likely time it will take in the future is
{S + 4xA + L}/6
so for S = 1, A = 5 and L = 15 your predicted time is 6 hours.
- Status