Do I need a website before running Google Ads?

You need one page that can convert and tracking that works, not a finished site. Paid traffic should land on a page built for the specific search rather than on a homepage, so the page you need for advertising is one you would have had to build anyway.

When should I wait before advertising?

When the offer itself is still changing, when the site is too slow on a phone to keep the visitors you paid for, or when nothing can be measured. The first is the most important: advertising an unsettled offer buys data about something that will not exist next month.

Can I send ads to a social media page instead?

You can, but you give up control of the layout, the message and the measurement. For anything you expect to produce enquiries, a page you own will nearly always outperform it.

How fast does my site need to be?

Fast enough on a phone over a normal mobile connection that a visitor does not leave before seeing anything. Test on an actual device rather than on office broadband, because that is the condition most of your paid traffic arrives in.

Should I rebuild the whole site first?

Usually not. A full rebuild takes months during which you acquire no customers and learn nothing about what message works. One landing page corrected by real traffic will tell you what the eventual site should say.

Should you run ads before the website is ready.

16 October 20267 min readBeyond Vision Digital

An arrow arriving at a page outline that is only half drawn, the missing half dashed in orange.

The short answer

You do not need a finished website, but you do need one page that can convert and tracking that can tell you whether it did. A single properly built landing page beats a half finished site, because paid traffic should never arrive on a homepage anyway. Wait only when the offer itself is undecided, because advertising an offer you are still changing buys data about something that will not exist next month. If the offer is settled, one page and working measurement are enough to start.

The real question is whether the offer is settled

A website can be improved while campaigns run. An offer cannot, or rather it can, but every change resets what the data means.

So the thing to check before spending is whether you know who this is for, what it costs and why somebody would choose it over what they use now. If those are still moving, wait. If they are settled, the state of the website is a smaller obstacle than it looks.

One page is genuinely enough

Paid traffic should land on a page built for the specific search, not on a homepage that has to serve everybody. That is true for mature sites too, which means the page you need for advertising is a page you would have had to build regardless.

So the honest sequence is not website then ads. It is one landing page, then ads, then the rest of the site once you know what the market responds to. The alternative is spending three months building pages around guesses that advertising would have corrected in two weeks.

What has to exist before the first click

Short list. Everything else can wait.

A page that matches what the advertisement promises, with one action on it.
Conversion tracking that fires on a real enquiry and has been tested by submitting the form yourself.
A way to reach you that works on a phone, because most of the traffic will be on one.
Somebody who answers enquiries quickly, since a lead nobody calls for two days is a lead you paid for and threw away.

The cases where waiting really is right

If the site takes many seconds to load on a phone over a normal connection, fix that first. Speed is not a refinement here. It decides whether the click you paid for ever became a visit at all.

Wait as well if there is nothing on the page that would make somebody choose you, and nobody has decided what that should be. Advertising does not create a reason to buy. It finds people who already have one, and sends them to a page that has to confirm it.

And wait if nothing can be measured. Spending without tracking produces a confident opinion after a month and no evidence, which is the most expensive outcome available.

Do not let the site become the reason nothing ever starts

The opposite failure is just as common and more expensive. A full rebuild takes months, the launch slips, and a year passes with no customers acquired and no idea which message works.

A landing page built in a week and corrected by real traffic will teach you more about what to put on the eventual website than any amount of internal discussion. That is also the cheapest way to find out whether the problem was ever the ads or the page.

Is your offer settled, or is your website unfinished

Those are different problems and only one of them is a reason to wait. Write down who this is for, what it costs and why somebody would choose it. If you can, you are ready to advertise to one page.

If you would like us to look at what you have and say whether it is enough to start on, send us the page and the offer through the enquiry form. Sometimes the answer is that a week of work on the page will save you a quarter of wasted spend, and we would rather say that before you spend it.

Questions we get asked.

Do I need a website before running Google Ads?: You need one page that can convert and tracking that works, not a finished site. Paid traffic should land on a page built for the specific search rather than on a homepage, so the page you need for advertising is one you would have had to build anyway.
When should I wait before advertising?: When the offer itself is still changing, when the site is too slow on a phone to keep the visitors you paid for, or when nothing can be measured. The first is the most important: advertising an unsettled offer buys data about something that will not exist next month.
Can I send ads to a social media page instead?: You can, but you give up control of the layout, the message and the measurement. For anything you expect to produce enquiries, a page you own will nearly always outperform it.
How fast does my site need to be?: Fast enough on a phone over a normal mobile connection that a visitor does not leave before seeing anything. Test on an actual device rather than on office broadband, because that is the condition most of your paid traffic arrives in.
Should I rebuild the whole site first?: Usually not. A full rebuild takes months during which you acquire no customers and learn nothing about what message works. One landing page corrected by real traffic will tell you what the eventual site should say.