How to Tell If a Contractor Can Actually Do the Work - practical ways to evaluate contractor capabilities on industrial projects, from someone who reads these documents for a living.
A few years ago I sat at a table with two proposals for the same job: a drive and motor control upgrade on a production line. One number was roughly half the other. The plant manager slid the cheap one across and asked the question everybody asks.
"Why would anyone pay double for that?"
Fair question. So we spent an afternoon on it. By the time we had written down everything the low bid did not include, commissioning hours, startup support, drive parameter files, as-built drawings, and travel for a crew based four hours away, the two numbers were sitting a great deal closer together than the cover pages suggested.
Nobody had lied. The cheap bidder had simply answered a different question than the expensive one.
That is the thing about evaluating contractors. Most of the time you are not comparing companies at all. You are comparing documents that describe different amounts of work.
So let us talk about how to actually tell the difference. Not the checklist you have already read ten times. The parts that have genuinely saved me money.
Start by evaluating your own scope
Uncomfortable one first. Most contractor disasters I have been called in to clean up were not caused by a bad contractor. They were caused by a scope of work vague enough that a good contractor had to guess, and then got blamed for guessing wrong.
If your request for quote says "upgrade the stacker controls," you will get five prices for five different projects. One bidder assumes new drives. One assumes you are keeping them. One assumes your existing drawings are accurate, which is optimistic in most plants I have walked through.
Before you evaluate anybody, write down five things:
What exists today, accurately. Not what the drawings say. What is actually in the cabinet.
What the machine has to do when you are finished, in numbers. Pieces per minute, tolerance, uptime.
Who owns the tie-in points, mechanical, electrical, and network.
What "done" means, and who signs the paper that says so.
What you are not asking for. This one prevents more arguments than anything else in the document.
A tight scope does something beyond preventing fights. It turns your bids into an actual comparison. And it lowers your price, because every hour a contractor spends guessing at your intentions gets priced as risk, and risk is the most expensive line item in any quote.
Credentials: what actually means something
Certificates are the easiest thing to ask for and the easiest thing to over-read. ISO 9001 means a company has a documented process. Not a good one. Documented.
Here is what is both verifiable and genuinely useful, at least on this side of the continent:
Firm-level licence to practise engineering
In British Columbia a firm needs a Permit to Practice from Engineers and Geoscientists BC. It is mandatory, it applies to the firm and not just to the individual engineer, and it is searchable. Alberta has an equivalent Permit to Practice through APEGA. In the United States it varies state by state, and a stamp from one state is not a stamp in the next. If your project crosses a border, ask early rather than at submission time.
Electrical contractor licence, and the FSR behind it
In BC, anyone hired to perform electrical work needs a contractor licence from Technical Safety BC, and that licence carries a designated Field Safety Representative whose certificate class sets the scope of what the contractor can legally declare. Class A is unrestricted. Class B covers installations under 1000 volts. Class C is limited to under 150 volts to ground, 200 amps, single phase. It is worth confirming the class covers your installation before somebody discovers otherwise at permit time.
Panel shop listing and SCCR
If a contractor is building you control panels, ask whether their shop holds a UL 508A certification, or the Canadian equivalent under CSA C22.2 No. 286, and whether they can apply their own panel shop label. Then ask what short-circuit current rating they intend to mark on it. The National Electrical Code requires an industrial control panel to carry an SCCR, and a panel rated below the available fault current at its installation point is an inspection problem waiting to happen. The cULus mark is accepted in both countries, which saves an argument.
Safety certification and standing
The Certificate of Recognition, COR, is the certification most owners in Canada prequalify against. It is voluntary, audited, and issued through provincial certifying partners. Ask for it, and ask for a current WorkSafeBC or provincial clearance letter alongside it. The clearance letter tells you they are in good standing on their premiums, which quietly tells you something useful about their cash flow.
Insurance, with dates
Not "we are insured." A certificate of insurance, current, with limits appropriate to your project, and naming you as additional insured if your contract calls for it. Read the expiry dates. I have seen a certificate expire in the middle of a shutdown more than once.
None of the above predicts good work. All of it predicts whether the finished work can be legally connected, inspected, and closed out. Those are two different problems, and the second one bites harder because it bites at the end, when your line is down and your budget is gone.
The reference call that actually works
Everybody hands you three references who adore them. Of course they do. So ask a different question, and ask it of both sides.
Tell me about a project that went sideways, and what you did about it.
Ask the contractor. Then ask the reference the same thing. If the two stories match, you are dealing with an honest contractor and an honest reference, which is worth more than a wall of glowing testimonials.
Four more that tend to produce real answers:
Who was actually on site, and are those people still with the company? Firms do not do the work. People do.
How many change orders, and were they genuine scope changes or things that should have been in the base bid?
How long after startup were they still answering the phone?
Would you hire them again for the same job, or only for a different one? It is remarkable how often that produces a yes and a no in the same sentence.
And the quiet one that tells you the most: what did the handover package look like?
The end of the job is the real test
I once walked into a plant to troubleshoot a machine and was handed drawings that described equipment which had not existed there for about a decade. It is my favourite recurring nightmare, because it is completely avoidable and almost never avoided.
A contractor’s real capability shows up in what they hand you on the last day:
As-built drawings, marked up and reissued. Not the bid drawings with a fresh date in the title block.
PLC and HMI program backups, with commented code and a version number you can still identify six years from now.
Drive and instrument parameter files. The day a VFD fails at 4 a.m., that file is the difference between an hour of downtime and a shift of it.
I/O lists and network addressing that match the installation rather than the plan.
Operations and maintenance documentation, and a spare parts list with part numbers you can actually buy today.
Training for the people who have to live with the machine after everyone else drives home.
Evaluate this before you sign, not after. Ask to see a real closeout package from a previous job, redacted if they need to. Then make documentation a payment milestone. Holding a meaningful piece of the contract until the closeout package arrives is the most effective project management tool I know of, and it costs nothing.
Go and look
You will learn more in twenty minutes standing in a panel shop than in twenty pages of proposal. Wire labelling. Ferrules. Whether terminations look like somebody cared. Whether the shop floor is organized or archaeological. Whether the person walking you around can answer a technical question without leaving the room to find someone.
Same principle applies to the money. Write a Factory Acceptance Test into the contract before anything ships, and then show up for it. I do this personally, not because I distrust the builder, but because finding a wiring error in a shop, with the panel on a bench and a full toolbox within reach, costs a fraction of finding it during startup with the line down and an audience. Every FAT I have attended has found something. Every single one.
How to read a low bid
A low bid is not automatically a bad bid. Sometimes it is a shop that is efficient, hungry, or already knows your equipment inside out. But there are only so many ways to be cheaper, and it is worth knowing which one you are looking at:
They are genuinely more efficient. Excellent. Ask them how, and see whether the answer is specific.
They excluded work. Check the exclusions list, and if there is not one, that is your answer. A good quote tells you plainly what it does not cover.
They assumed something optimistic, usually about your existing equipment, your documentation, or your electrical capacity.
They are buying the job and intend to earn it back through change orders.
They have mispriced it and do not know yet. This is the worst outcome for you, because a contractor losing money on your project starts making decisions you will not enjoy.
Normalize before you compare. Take every bid and add the missing scope back in: commissioning, travel and accommodation, documentation, spares, training, freight, and any duties or exchange on cross-border equipment. Then compare the loaded numbers. The ranking changes more often than you would think, and occasionally it inverts entirely.
A scorecard you are welcome to steal
Gut feel plus a number beats either one on its own. Here is the weighting I work from. Adjust the weights to suit your project, but keep the discipline of scoring every bidder before you look at the prices.
Criterion | Weight | What you are actually scoring |
Relevant experience | 20% | Have they done this on a line like yours, at your scale, in your process? Not "an industrial project." |
Named people | 15% | The specific individuals who will be on your site, their availability, and whether they have done this exact thing before. |
Scope completeness | 15% | Does the quote prove they read yours? Is there a clear exclusions list? |
Documentation and handover | 10% | Quality of a sample closeout package, and willingness to tie it to a payment milestone. |
Safety record and program | 10% | COR or equivalent, current clearance letter, and how they talk about safety when nobody is grading them. |
Licensing and compliance | 10% | Firm permit to practice, contractor licence and FSR class, panel shop listing, current insurance certificate. |
Commissioning and support | 10% | Commissioning hours included, response commitment after startup, and who actually answers the phone at 2 a.m. |
Financial stability | 5% | Clearance letter, credit or supplier references, payment history with their own vendors. |
Price (loaded) | 5% | The normalized number, after you have added every bidder’s missing scope back in. |
Price at five percent bothers people, and it should. It is deliberately provocative. Set it to something you can defend to your finance director. But most owners I have worked with score price at forty percent and then spend the difference twice over on rework, so the honest answer sits somewhere between those two numbers.
Red flags, the engineering-specific kind
They will not name the individual who will be on your site.
There is no exclusions list anywhere in the proposal.
They are cheerful about ambiguity. "We will sort out the details during commissioning" means "we will invoice the details during commissioning."
Resistance to a witnessed factory acceptance test.
Documentation appears as a priced option rather than as part of the work.
They cannot produce a current insurance certificate or clearance letter within a day.
All of their questions are commercial and none are technical. Anyone serious about your project asks awkward questions about your existing equipment. Silence during the site walk is not confidence.
You are being sold a platform when you asked for a solution.
The honest bit
Most owners are not bad at evaluating contractors. They are simply doing it three times a decade, against people who bid work every week. That is not a knowledge gap, it is a repetition gap, and it is why a forty thousand dollar mistake gets made on a four hundred thousand dollar project by a clever person having a busy month.
Closing that gap is most of what I do. Somebody has a project, a stack of quotes that do not line up, and a nagging feeling that the numbers are not comparable. We go through them together, in plain language, and I tell them what they are actually looking at. Then we reach into the HIVE, our network of vetted engineers, contractors, and specialists, and match the project to people who can genuinely do it, against a scope tight enough that they can hit the ground running. That saves them time, which saves you money.
You should not have to be an engineer to hire one.
If there is a stack of bids on your desk right now and something about them is not sitting right, that is exactly the conversation to have. The first consultation is always free. No commitment, no sales pitch, just a straight answer from someone who has been on both sides of the table:
The best projects are not built out of steel and machinery alone. They are built out of decisions made with the right information sitting in front of you.
Comments