There is a meeting you are never invited to, and it decides most of your year.
Somebody at a prime contractor, a rural electric cooperative or a state broadband office has a list of subcontractors to put in front of a committee. They have a deadline, an open spreadsheet, and eleven names to get down to four. They are not calling eleven companies. They are opening eleven websites, one after another, and spending about ninety seconds on each.
That is the whole competition. Not the bid, not the interview, not the relationship you have with someone who used to work there. The first cut is made by a person clicking through your site at speed, looking for a small number of specific facts, and the companies that survive it are the ones whose facts were easy to find.
What they are actually checking
This is not a mystery and it is not subjective. Anyone who has filled in a prequalification packet knows the list, because the packet asks for the same things every time. The person building the shortlist is doing an informal version of that packet from your website, and they are looking for four things.
Whether you can be bonded, and for how much. A single bonding capacity figure and the name of your surety answers a question that otherwise takes an email and three days. If the job is a two million dollar build and your site says nothing about bonding, the safest assumption is that you cannot carry it, and safe assumptions are what shortlists are made of.
Whether you are going to hurt somebody. EMR, and a recordable incident rate if you have one worth publishing. An EMR under 1.0 is a number you should be putting in front of people rather than waiting to be asked for. Primes have insurance requirements they cannot waive and a subcontractor who looks like an underwriting problem gets removed early and quietly.
Whether your crews are certified for the actual work. Not “experienced team.” The certifications: OSHA 10 and 30, the manufacturer programs you hold, BICSI credentials if your people carry them, CDL classes, confined space, flagger, whatever the work requires. A list of certifications is boring to write and it is the single most scannable proof on a contractor website.
Whether you actually work where the job is. A county list, written as text. Not a phone number and an invitation to ask. If a coordinator in Lincoln cannot tell in four seconds whether you cover the three counties in their build, you are not a maybe, you are a skip.
A buyer who is not ready to inquire is not a lost buyer. They are a buyer who has not been given a reason yet.
Why all of it ends up behind the contact form
Almost every contractor site I have looked at hides at least three of those four, and it is rarely an oversight. It is usually one of three beliefs, and all three are wrong in the same way.
The first is that specifics are boring. Someone at some point decided the homepage should say “quality craftsmanship since 1987” instead of “EMR 0.78, bonded to $5 million, 140 crew, 22 counties.” The first sentence is on ten thousand contractor sites and it persuades nobody. The second sentence is on almost none of them and it does the entire job.
The second is that giving the numbers away removes the reason to call. This gets the sequence exactly backwards. Nobody calls a contractor to find out whether they are eligible. They call once they have decided you are, and everything you withhold before that point is a reason to stop rather than a reason to reach out.
The third is the one that actually costs the most: a belief that the website is for people who already know you. That is true of your existing relationships and it is precisely why a website matters, because those are the jobs you were getting anyway. The website’s only job is the buyer who does not know you and never will unless the first ninety seconds go well.
What to put above the form instead
The fix is not a redesign. In most cases it is one screen of content moved from a PDF or a prequalification packet onto a page, in text, above anything that asks for an email address.
- A capability block on the page they land on. Bonding capacity, EMR, crew count, years operating, and the certifications, as readable facts rather than a downloadable capability statement. The PDF can stay; it should not be the only copy.
- A service territory answered in one screen. A map is fine, but the county list has to be text beside it, because a map image is invisible to a search engine and useless to someone on one bar of signal.
- Projects described by a person who was there. Scope, footage, conditions, what went wrong. Twenty minutes with a foreman produces better project copy than any marketing brief, and it reads as true because it is.
- References you can name. One named prime or cooperative who will confirm the work is worth more than six anonymous testimonials, and the permission is easy to get at project close while everyone is happy.
If you do not have the numbers
Some of this will not apply to you yet. A three-year-old company does not have a decade of safety data, and a small operation may not carry a bond large enough to want to publish the figure.
The answer there is not to go quiet, because quiet reads the same as bad. The answer is to publish what you do have and be specific about the rest. “Bonded through [surety], capacity discussed per project” is a real answer. So is “OSHA 30 across all foremen, no recordable incidents since we opened in 2023.” A small number stated plainly beats a large one implied, and a buyer who is fine with a smaller subcontractor still needs to be able to tell that is what you are.
What does not work is the middle ground almost everyone picks, which is a page of adjectives that could describe any company of any size in any state. That page does not fail because it says the wrong thing. It fails because it says nothing a shortlist can use.
The other half of the problem
There is a second reader arriving at the same website, and in this sector they may matter more than the first one. A splicer or a lineman with ten years of experience is deciding whether your company looks like a place that treats crews properly.
In most of this industry the constraint is not winning work, it is staffing it. A contractor who cannot crew a job turns it down, and a careers page that is one line of text and a generic inbox is the reason good applicants go elsewhere. That page deserves what a product page gets: what the job actually is, where it is, what the equipment is like, who you would report to, per diem, rotation, and an application that can be finished on a phone at the end of a shift rather than a PDF that has to be printed.
Both readers, incidentally, are often on a phone in a truck. A site that takes six seconds to load is a site that was never seen, and that is almost always four specific decisions rather than bad luck.
A ten-minute test
Open your own site on your phone, on cellular rather than the office wifi, and give yourself ninety seconds.
- Can you find your bonding capacity without using the menu?
- Can you find your EMR at all?
- Can you name your service territory from the page in front of you, in counties?
- Could an experienced splicer apply, on that phone, in under five minutes?
Four noes is normal. It is close to universal in this sector, which is the good news, because the bar you are actually clearing is set by the other ten websites on that coordinator’s list. Fixing two of the four puts you in front of almost all of them.
This is the argument I make at more length, with what it looks like built, on the web design for infrastructure contractors page. And if you want the version with a real company attached to it, the Central Plains Communications case study is the same set of decisions made on a site I own, which means I had to live with every one of them.
