My Bismarck Audit, in the Order I Actually Run It

Last week I sat down with a coffee and ran my standard audit across a long list of Bismarck businesses.

It’s the same run I’d do for a client who’s paying me, in the same order I always do it, because I’ve done it enough times now that changing the order is how I start missing things.

Reviews first. Then whether the site loads, and whether it loads safely.

Then the phone. Then the code. Then the parts an actual human being reads.

It ate most of a Tuesday.

First stop: the listing, before I open the site at all

I never start on the website. I start on the Google listing, because that’s where your customer starts, and because a website nobody reaches is a philosophical question rather than a business problem.

Stars, hours, photos, address, phone. If a business has no reviews on it whatsoever, the listing shows up as a plain gray line beside four competitors with gold stars under their names, and the comparison is finished before anybody has clicked anything.

Which is where I stopped for the first time.

77.7% of the businesses I sampled in Bismarck hold at least one Google review, where 88% do across everything I’ve counted.

No city I audited sits lower than that.

It surprised me more than anything else in the whole run, because asking for a review costs nothing at all. It’s one sentence at the end of a job, said out loud, by a person who has just finished doing good work in someone’s house.

Would your own listing get through that first stop with stars on it? Mine wouldn’t have, for about the first three years I was in business.

Second: does it load, and is the door locked

Next I type the bare web address with no https on the front of it, then watch where the browser takes me.

Almost every site in town now has a certificate, so the Not Secure warning I used to hit on these runs almost never turns up any more. That half is fine and I’ve got nothing interesting to say about it.

Locking the door is also the easy half. Financial firms run the tightest websites I audit and the emptiest Google listings of any trade I count, which is a strange pair of habits to hold at once.

The redirect is the sloppy half. 87.2% of the sites here push plain traffic across to the secure copy, against 88% nationally, and the rest will cheerfully hand you the unlocked version if you type the address without the s.

Nothing dramatic breaks when that’s missing. A customer lands on the old copy, gets a warning from her browser, and quietly decides you’re either broken or sketchy, and she will never tell you which one she picked.

Have you ever typed your own address without the s, just to see where a browser drops you?

Third: the phone

Two separate checks go here, and owners mix them up almost every time.

The first asks whether the page fits a phone at all, and Bismarck sits right on the national rate for it, which makes it the least interesting result of the lot.

The second asks how long the page takes to arrive, on a simulated phone over a deliberately slow connection, which is a great deal closer to your customer standing in a parking lot than your office wifi will ever be.

Bismarck averages 56 out of 100 on that one, against 58 across the whole sample. Google’s own scale doesn’t call a score good until it reaches 90, so that isn’t a pass with room to improve.

Fitting a screen and showing up quickly are two completely different jobs, and only one of them is something your customer physically sits through. Which of the two do you think your last designer was hired to solve?

What breaks when a page is slow never shows up in any report I could write for you. Someone taps, waits, gets bored, hits back, and taps the next result down instead, and you will never meet that person or hear a word about it.

Fourth: the code

This part takes the longest and produces the least argument, because nobody in any office has ever had a strong opinion about a tag.

The meta description is the sentence a searcher reads underneath your blue link, in the half second before she decides whether you’re worth a tap. Leave it empty and Google writes one for you by grabbing whatever text it finds first, which on a lot of homepages turns out to be a cookie notice or the words Home Page.

74.3% of the sites here have one written, against 77.9% across my sample. The rest are handing their only line of first impression to a robot with no stake at all in how it turns out.

When did you last read the sentence Google is showing underneath your own name?

Another line down there tells Google what you actually sell, and Bismarck finishes last of all 24 cities on that one.

I owe you one caveat here, and it sits underneath everything else. Bismarck arrives with 2,562 businesses sampled across 37 industries, and no other city I collected covers fewer trades than that. A narrower spread means a couple of unusual industries can shove a city average around more than they could in a bigger town.

Fifth: the parts a person reads

Last of all I go through the page the way a customer with a problem would. What does it ask me to do, and how hard is it to do that thing?

26.6% of Bismarck homepages carry a contact form, against 28% across the sample.

What breaks without one is a person at ten at night who has decided to hire you and now has to go hunting for an email address. Some of them will hunt. Most of them go back to the map and tap whoever is listed underneath you.

Every site I build for a business around Bismarck launches with a form on the homepage, and it’s still the change clients thank me for first, ahead of anything that took me ten times as long.

What the ten checks add up to

Run all of them, weight them the way I always weight them, and Bismarck lands on 70 out of 100.

The score is a weighted average of ten checks, so it isn’t a grade for any single one of them, and a town can lose a couple of points on something its customers will never notice.

Toronto is the line directly above Bismarck in the city rankings. Montreal and Windsor are the only two sitting below it.

Up at the other end, Denver came second on the same ten checks and I ran that city three times before I trusted it.

That’s a lot closer together than the ordering makes it sound, which is my honest read of the whole exercise. Nothing here is broken. A handful of habits are just missing, and every one of them is free.

Habits, not rebuilds. Everything a site asks for after launch is about that size.

Your own site sits somewhere inside that spread, and a position on a table matters a good deal less than the habits sitting underneath it.

Would your own site survive that run in that order? Most wouldn’t, and mine didn’t the first time I turned the thing on myself.

If it were my business, I’d spend Saturday morning asking every customer from this year for a review, and Saturday afternoon putting a form on the homepage. The rest can wait for a week when someone is paying me to care about it.