Is it the ads or the landing page.
Four numbers tell you whether the problem is the traffic or the page. If click through rate is healthy and conversion rate is poor, the ad is promising something the page does not deliver, and the page is the problem. If click through rate is poor and the people who do arrive convert well, the targeting or the ad is wrong and the page is fine. If bounce rate is high and time on page is very short, the page loaded slowly or looked nothing like the ad. Check all four against the same period before changing either.
Everybody blames the half they do not own
Ask the ads agency and the traffic is fine, the site is the issue. Ask whoever built the site and the site is fine, the traffic is unqualified. Both are answering honestly from inside their own square, and neither is looking at the join.
The join is where almost all of the money goes. It is also the cheapest thing to diagnose, because the four numbers you need are already in your accounts.
The four numbers, and what each pair means
Click through rate tells you whether the ad matched the search. Conversion rate tells you whether the page matched the ad. Bounce rate and time on page together tell you whether the page even had a chance.
Read them in pairs rather than one at a time:
- Click through healthy, conversion poor. The ad is writing a cheque the page does not cash. This is the most common case and the most fixable.
- Click through poor, conversion healthy. Your page is fine. You are showing the wrong ad, or showing it to the wrong search.
- Both poor. Usually the offer rather than either asset. Neither more traffic nor a nicer page fixes an offer nobody wants.
- Both healthy but few leads. Look at the form and the tracking before anything else, because the drop off is after the point both of these measure.
Short visits are a page problem wearing a traffic problem's clothes
If most visits end in a few seconds, the visitor did not read your page and reject it. They never saw it. That is either speed, or a page that looked nothing like the ad they clicked.
Message match is the part people underestimate. Somebody clicks an ad about same day delivery and lands on a homepage that leads with a company history. They have not decided against you. They have decided they are in the wrong place.
The test that settles it in a week
Take your best performing ad group, the one with a click through rate you are happy with, and send it to a page built for exactly that search. Not the homepage. Not a category. A page that says the thing the ad said, and asks for one action.
If conversion rate moves, the page was the problem and you now know what to do with the rest. If it does not move, you have eliminated the page and the problem is upstream in the targeting or the offer. Either outcome is worth the week.
Why this is worth settling before spending more
Every pound added to a campaign pointed at a page that does not convert buys more of the same result. The account looks busier. The reporting looks better. The bank account does not change.
So the question to answer before the next budget increase is not whether to spend more. It is which half of this is losing the money, and whether anybody has actually checked.
If you have the four numbers and they do not agree with each other, or if you have been told for six months that the traffic is fine and the site is fine while the leads stay flat, that is the conversation worth having with somebody who owns both halves. Send us the account and the page and we will tell you which one is costing you.
Questions we get asked.
How do I know if my landing page or my ads are the problem?
- Compare click through rate against conversion rate for the same period. Healthy click through with poor conversion means the ad promised something the page does not deliver. Poor click through with healthy conversion means the page is fine and the targeting is wrong. Both poor usually means the offer rather than either asset.
Should I send paid traffic to my homepage?
- Rarely. A homepage is built for somebody who arrived knowing your name. Somebody clicking a specific search wants the specific thing they searched for, and every extra decision between them and it costs conversions.
What conversion rate should I expect?
- It depends entirely on price, considered purchase length and how warm the traffic is, which is why benchmark numbers are close to useless. The number worth comparing against is your own last quarter, and whether a change you made moved it.
My bounce rate is high. Is that always bad?
- Not always, but a high bounce rate with very short visits almost always is. It means the page did not load in time or did not look like what was clicked. A high bounce rate with a long visit can simply mean the page answered the question.
How long should I test a new landing page?
- Long enough to tell a real difference from a quiet week. That usually means a few hundred visitors per version at least. Reading a test after two days is how teams end up shipping the version that happened to have a good Tuesday.
