How to Drive Traffic to Your Website, and Know Which Part Worked

There is no shortage of advice on getting traffic. There is a serious shortage of advice on working out which of it worked, which is why so many sites do five things at once and then argue about which one mattered.

This is both halves: what tends to work, and what you need in place to know.

Write the pages people are actually looking for

The single most common problem is not quality, it is that a site has nothing on it matching what anyone types into a search box. A homepage, a pricing page and an About page cover perhaps three queries between them.

Look for the questions your customers ask you before they buy, then write the page that answers each one. If people repeatedly ask whether you work with something they already use, that is a page. If they ask how you compare with an alternative, that is a page too, and it will attract people much closer to a decision than anything at the top of the funnel.

How you will know it worked: organic search terms, with impressions and click-through rate. Connect Google Search Console and you get the actual queries, which is the only reliable route to them since Google stopped putting search terms in the referrer in 2012.

Go where the conversation already is

Answering a question properly in a forum, a subreddit or someone's comment section brings better visitors than most link building, because you arrive with context. It does not scale, which is exactly why it still works.

How you will know it worked: the referrers panel. Unfamiliar domains sending real traffic are worth following up, usually by writing more where that came from.

Send email, and tag the links

If you have any kind of list, it is almost certainly your most reliable source per hour spent. It is also the one people most often fail to measure, because an untagged link from an email client shows up as direct traffic and disappears into the pile.

How you will know it worked: UTM parameters. Tag the campaign, and the sources panel splits it out by source, medium and campaign instead of lumping it in with everyone who typed your domain.

Fix the pages that already get traffic

The cheapest traffic win available to most sites is not new traffic. It is the page that already receives visitors and sends none of them anywhere useful.

Find your top entry pages. For each one, ask what the obvious next step is and whether the page offers it. An old post that ranks well and links to nothing is a leak, and closing it takes ten minutes.

How you will know it worked: entry pages and exit pages, plus scroll depth to see whether people reach the point where you ask them to do something.

Buy traffic only once you can measure it

Paid works, provided you can tell the difference between a click and a customer. If you cannot, you are funding an experiment with no readout.

Before spending, have at least one real goal set up, and export it to your ad platform so bidding optimises against something that matters rather than landing page views.

The measurement setup, minimally

Five things, and none of them take long:

1. Search Console connected, so you see real queries rather than guesses
2. UTM tags on every campaign link you control
3. One goal, matching the action you would be pleased to see
4. Scroll depth, if you publish anything longer than a screen
5. Session replay, so you can see why the traffic you worked for did nothing

That fifth one is the one people skip, and it is the one that most often explains the others. Traffic arriving and leaving is a puzzle in a dashboard. Watching four of those sessions usually turns it into a fifteen-minute fix.

What not to do

Do not run five tactics in the same fortnight. You will not be able to attribute anything and you will end up believing whichever one you liked best.

Do not chase volume for its own sake. Ten thousand visitors who bounce are worth less than two hundred who read the page, and they cost more to serve.

Do not judge anything in under two weeks. Traffic is noisy, weekdays differ from weekends, and the temptation to declare victory on day three has ruined more marketing decisions than bad tactics ever have.

See what Galytics measures, or read how we compare with Google Analytics if that is what you are running today.