Websites

What a new website should do in its first 90 days

Most small business websites are treated as finished on launch day and never touched again. The three months after launch decide whether it was worth building.

A website is not a brochure. A brochure is finished when it is printed. A website is a salesperson you have just hired, and nobody hires a salesperson, shows them the door on day one and never asks how it is going.

Here is what the first three months should look like.

Day zero: what “launched” should actually mean

Before anyone celebrates, six things should be true. They take an hour between them and they are the ones that get missed:

  • Every form has been submitted by a real person and the email arrived, in the inbox, not in spam. Test it from an address outside your own business.
  • The site is in Google Search Console and Bing Webmaster Tools, and the sitemap is submitted to both.
  • Analytics is installed and recording, and somebody can name what counts as a conversion.
  • The old URLs redirect to the new ones, if this replaced a site. Skipping this throws away every bit of ranking the old site had.
  • It loads properly on a phone on cell data, not on the office wi-fi.
  • There is a backup, and somebody has checked the backup actually restores.

That last one sounds paranoid until the first time a plugin update takes a site down on a Friday afternoon.

Weeks one and two: make sure it can be found

A new site is invisible by default. Google has to discover it, crawl it and decide what it is about, and none of that is instant.

Check in Search Console that your main pages are actually indexed, not just submitted. There is a difference, and “crawled, currently not indexed” is a message worth understanding rather than ignoring. Make sure your business name, address and phone match your Google listing exactly. Get the site linked from your own social profiles and any directory you already appear in, because a site with no links pointing at it is a site Google has little reason to trust.

Weeks three to six: find out where people give up

By now there is enough traffic to see a shape. You are looking for one thing: the point at which people leave.

Three questions answer most of it. Which page do most people land on? Which page do most people leave from? And how many of the people who reach your contact page actually send anything? That last number is the one that matters, and on a lot of sites it is well under one in twenty, which usually means the form is too long or asking for something people are not ready to give.

This is also the point to read the site on a phone, slowly, as a stranger. Most owners have never done that. It is uncomfortable and it is the cheapest research available.

Weeks seven to twelve: fix the two pages that matter

There are almost always exactly two: the page most people arrive on, and the page where they decide. Everything else can wait.

On the arrival page, the first screen has to answer what you do, where, and what to do next. Not a slogan. On the decision page, remove one field from the form, put the phone number where a thumb can reach it, and add the one piece of proof a real customer would ask for.

Change one thing at a time and leave it for two weeks, or you will never know which change did it.

The numbers worth watching, and the ones that are noise

Worth watching: inquiries, calls, where the inquiries came from, which pages are indexed, and what your site looks like on a phone.

Noise: total visits, time on page, bounce rate on a one page visit, and your position for a phrase nobody searches. A number that goes up while inquiries stay flat is a number that is measuring the wrong thing.

If you are already past 90 days and none of that happened

That is the usual case and it is recoverable. Start with the launch checklist above, in order, because it is common to find that the contact form has been silently failing for months. That is not a rare failure. We have seen a business lose nearly a month of inquiries to it without a single error message anywhere.

Then pick one page and fix one thing. Websites are not rescued in a rebuild. They are rescued in a habit.

If keeping that habit is the bit that never happens, it is worth handing off. Our managed growth program covers the watching, the updates, the backups and the monthly report, so the site stays looked after rather than remembered.