The Week After It Goes Live

Why launch is not the end

Everybody treats going live as completion. The invoice is settled, the project closes, attention moves elsewhere.

But a site has only ever been used by the people who built it, on their machines, in the way they expected. The week after launch is the first time it meets people who were not involved.

That week produces a specific category of problem: things that work in testing and fail in use.

Budgeting a few hours for it is more valuable than the equivalent hours of polish beforehand, and almost nobody does.

The first hour

1. Load the site from your phone, on mobile data, not office wifi.
2. Submit the contact form as a stranger would and confirm it arrives.
3. Reply to that submission, to check the reply-to address works.
4. Tap the phone number and confirm it dials the right one.
5. Check the address and hours against reality.
6. Click every item in the menu.

The second and third find more real problems than everything else combined. A form that submits successfully and delivers nowhere is the most common launch failure, and it is silent by definition.

Do it from an address unconnected to the business. Internal mail frequently behaves differently from external.

The first day

Things that need doing while somebody is still paying attention.

Confirm the site is not blocked from search engines. Sites are commonly hidden during development and the block is occasionally left in place, which is a problem that takes months to notice and minutes to fix.

Submit the site to Search Console so you can see what Google makes of it. Check the analytics is recording, and that it is recording visits from other people rather than only from you.

And check the old addresses if this replaced a previous site. Every page that used to exist should land somewhere sensible rather than an error.

The first week

Now real people are using it, and their behaviour tells you things testing could not.

Watch which pages they actually visit. It is rarely the order you designed for, and the page you thought was central is frequently not the one they land on.

Watch where they leave. A page everybody exits from is either finished or broken, and it is worth knowing which.

Count the enquiries and, more importantly, read them. Enquiries asking a question the site answers mean the answer is not findable. That is a navigation problem wearing a content costume.

A worked example

A new site for a domestic cleaning company, launched on a Monday.

Everything tested correctly before launch. On Thursday the owner mentioned she had not had a single enquiry all week, which was unusual.

The form was working. Submissions were being delivered. They were being delivered to the address of the person who built the site, because that was what had been used for testing and nobody changed it afterwards.

Four enquiries had arrived and been sitting in somebody else's inbox for three days.

The fix took a minute. The lesson was that the test which mattered was not whether the form submitted, but whether the submission reached the business, and only somebody at the business could check that.

What to expect to find

So none of it feels like a disaster.

A typo somebody spots immediately. A phone number with a digit wrong. A page missing from the menu. A photograph that looks wrong on a phone. An old page still linked from somewhere. A service you forgot to include.

All of that is normal. A launch with no follow-up items usually means nobody has looked properly.

What is not normal, and worth escalating: the form not arriving, the site not appearing in search at all after a fortnight, or pages returning errors. Those three are defects rather than snags, and they are the supplier's to fix rather than yours to live with.

Who should be looking

Worth stating because it decides whether any of this happens.

The person who built the site cannot do most of these checks. They cannot tell whether the phone number is right, whether the hours match reality, whether an enquiry reached the business, or whether the service list is complete.

Only somebody who works there can. Which means the week after launch needs an hour from you, not from them.

The arrangement that works is a short list, agreed before launch, of what each side checks and by when. Otherwise both parties assume the other is watching and nobody is.

Telling people it exists

The step that gets forgotten in the relief of finishing.

Update the address on your listing, your social profiles, your email signature, your invoices and your vehicles. Tell existing customers, who are the people most likely to look and the easiest to reach.

If you have a list, one short message. If you do not, mention it to the next twenty people you speak to, and put the address on the next batch of invoices you send out.

A new site with no traffic is frequently not a search problem. It is that nobody has been told.

The month after

One more look, once there is enough use to be informative.

By then you know which pages matter, which questions still arrive by phone, and whether the enquiries are the kind you want. That is the moment to make the second round of changes, with evidence rather than assumption.

It is also the moment to write down what you learned, because in a year you will be considering changes and will have forgotten all of it.

Keep it to half a page: what people actually asked, which pages did nothing, and what you would add if there were budget. That note is worth more than any report, because it is written from a month of watching rather than from a dashboard.

Then leave it alone for a while. Sites benefit from a settled period, and changing things weekly on the strength of a fortnight's data produces churn rather than improvement.

What the site was supposed to achieve, and whether it is achieving it, comes back to the reasoning in deciding whether to build a site at all.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.