Three Jobs Every Website Needs. Dallas Finishes One.

For most of one year my own contact form was dead, and I had no idea.

I built the thing myself and I tested it the day it went up. Then I never tested it again, and somewhere in the middle of that year it quietly stopped emailing anyone at all.

I found out when a man in Dallas mentioned in passing that he’d written to me twice in the spring and decided I wasn’t interested in his job.

So I’m not asking you the next three questions from any great height. I’m asking because I’ve failed all three of them myself, in my own business, with my own hands on the keyboard.

Ready?

What does Google print underneath your name in the search results? Could a stranger contact you without leaving the first screen? And what did your website do last Tuesday?

I ran all three of those across Dallas, on every small business homepage in my sample that opened for me.

The city finishes the first job. The other two mostly go undone, and the cost of that never turns up on a report anywhere.

1) What Google prints under your name

78.2% of Dallas sites write a meta description, against 77.9% nationally. Put the 24 cities in order on that check and Dallas lands twelfth.

A client of mine in Oak Cliff never wrote hers, so Google went and picked a line off her page for itself. For two years, everybody who searched her business name got “This website uses cookies to improve your experience” where her offer should have been.

Google’s documentation explains why that can happen. It says snippets are built mostly from the page itself, and that your description gets used when that version looks more accurate.

So you aren’t issuing an instruction. You’re making an offer Google is free to turn down, which changes how the thing ought to be written.

Most of the ones I read are a summary of the page, and a summary is worth nothing to you. Nobody in the history of the world has hired a plumber because his description gave a fair overview of his homepage.

I write mine like the first line of an advertisement instead. What you fix, where you do it, and one reason to tap you rather than the two companies sitting above you.

That gray line isn’t the only line worth a few minutes. Dallas does better on the headline across the top of the page than on anything else I count, and that one is written the same way.

Go and look at yours right now, on the results page, the way a customer would meet it. Would that sentence make a stranger pick you over the other results on that screen?

2) The form nobody built

28.4% of Dallas homepages carry a contact form, where nationally the figure is 28%.

Level with the country, and still miserable.

A contact form is a few boxes a stranger can type into and send you without picking up a telephone.

The man I built for in Garland had a phone number in the footer and nothing else on the page. He was certain that was plenty, on the grounds that everybody who wants him calls him.

Everybody who wants him and is willing to phone a stranger calls him. That is a much smaller group of people than he thinks it is.

There’s the woman on her lunch break who can’t have that conversation at her desk. The man comparing three companies at eleven at night. Everyone who quietly hates the telephone.

Each of them arrives ready to hire you, finds nothing on the screen to talk into, and taps back to the next result. You never hear about any of them, and most of the reasons people abandon a page are that ordinary.

How many of those people walked past your homepage last month, and what would you have paid to get any one of them back?

A phone number in the footer doesn’t count, by the way. That’s a phone number in the footer.

Real estate is the trade that settles this argument. Agents carry a form on nearly half their homepages, and they still tell me nobody has ever once filled one in.

The version I build is three fields on the first screen with no scrolling: a name, a phone number or an email, and one line about what they need from you.

Every extra field you add costs you people who were halfway to becoming customers. I’ve never once needed a stranger’s company size before I’d agree to talk to them.

3) What your site did last Tuesday

58.0% of Dallas sites run analytics, and nationally that’s 58%.

One client of mine wanted his services page deleted. He’d written it himself years ago, he’d hated it ever since, and he was sure nobody ever read it.

Nothing on his website had ever counted a single visitor, so every page on it was a matter of opinion, including that one.

So we switched analytics on and waited. His services page turned out to be the second busiest thing he owned, and the last page most people opened before they rang him.

He was one afternoon away from deleting his best salesman on a hunch.

Which of your pages do people give up on? What did that Facebook post actually send you, if anything?

Setting analytics up takes an afternoon and reading it takes ten minutes a month, which is cheaper than one wrong guess about where your advertising money goes this quarter.

It won’t tell you why anything happened, mind you. It tells you the page people quit on, and then you have to open that page yourself and look at it honestly, which is the harder half of the job.

Count the gap in people, not percentages

Two of these three jobs go undone across this city, and the bill never arrives as a number. It’s the shape of the whole place, honestly, because Dallas finishes fifth of 24 cities without winning a single check.

It arrives as a woman who read a cookie notice under your name and chose someone else. As a man who wanted to write to you at eleven at night and couldn’t. As a page you were an afternoon away from deleting.

Not one of them ever tells you they were there, which is why all three jobs are so easy to leave for another month.

I’d rather you didn’t take my word for any of this, and the explorer on that report will hand you Dallas with every rate and its own sample size beside it if you’d like to check me. What it can’t tell you, and what my own checks can’t either, is whether the form it counted on your homepage still reaches a living person.

Mine counted as a pass for most of a year while it was quietly deleting people. When did you last send a test message through your own form and wait around to see whether it arrived?

All three of these jobs sit on the checklist I run before any site goes live, which should tell you how basic they are. None of it is strategy. It’s closing the till at night.

Do them before Friday

You can knock all three out in one sitting, and I’d take them in this order:

1. Search your business name, read the description underneath it, and if it turns out to be a summary of your homepage, rewrite it as an offer.

2. Look at the top of your own homepage and ask what a customer who will never phone you is supposed to do with it, and if the answer is nothing, put three fields there today.

3. Install analytics if it isn’t there, send yourself a test message through your own form, and then leave the pair of them alone for a month.

That’s your afternoon gone. None of it is design work, none of it needs a developer standing over you, and all of it outlasts whatever you were planning to spend on advertising this month.

If you’d rather the whole thing was simply handled for you, that happens to be the work I do around Dallas.

I won’t promise you the phone rings twice as often. What I’ll promise is smaller than that, and I’ve watched it happen enough times to say it out loud.

A month from now you’ll know which page loses people. The next stranger who won’t call you will have somewhere to type. And nobody will ever read a cookie notice under your name again.