When enquiries drop or advertising stops paying off, the owner's first thought is usually: the site is outdated, we need a new one. Sometimes that is true. But more often, five checks that take about an hour will show what actually needs fixing: the site, the traffic, or what happens after the order. We will walk through them using our own numbers, and the first one is not in our favour: on our store IWANT, which already runs on a new engine, advertising brought 1,831 clicks and zero payments in two weeks. A new site would not have helped there, because the store already had one.
1. Where the traffic comes from and how much of it takes action
Open the traffic sources report in your analytics tool (Google Analytics or similar) and look not at the number of visits but at how many visitors from each source take the target action: add a product to the cart, fill in a form, call. Paid and organic traffic on the very same site can differ several times over.
At IWANT, from 6 to 20 September, 11 of 1,835 paid visits reached the cart (0.60%), against 34 of 837 organic visits (4.06%). It is the same site, and only the source differs. That is a 6.8-fold gap. The conclusion: the problem was the traffic we were sending, not the site. We switched the advertising off on 20 September.
Limits. This is one store, with only 11 and 34 carts, and zero payments from paid traffic and one from organic. It is enough to show the order of magnitude, but not enough for a universal rule.
2. Is there an action on the page
Land on the page that your advertising or search sends people to, as a visitor would, and answer one question: what can I do here in a single step? Call, leave a request, add to cart. If the answer is "go to another page and look for contacts there," that costs you visitors.
3. How the site looks and works on a phone
Open the site on your phone over mobile data, not on the office Wi-Fi. Look at the first screen: is it clear what you sell and how to get in touch, without scrolling? Tap the request button: does the form open and send? We have no numbers of our own for this point, it is a manual check.
4. Does the promise match what the person sees
Type into a search engine the query you want to be found for, or click your own ad. Does the first thing you see on the page answer that query? If the ad promises a specific offer and the visitor lands on a generic home page, some visitors may leave immediately. We have no numbers here either.
5. What happens after the order
This is the most expensive finding. Look at how many orders over the last three months were actually completed: paid, received, not returned.
At IWANT, of 19 orders over 90 days, 9 were completed. Five were cancelled, and all five were placed with payment on delivery; two were returned. Counted from visits, there are about 1,200 visits for every completed order: nine orders from 10,770 visits, or 0.084%. The figure is small, and nine orders is a thin base, but the order of magnitude is clear. Buying more traffic or building a new site is unlikely to change this if orders fall away after checkout.
When the site really is the problem
A new site is needed if these signs remain after the five checks:
- the person reaches the right page, sees it and does nothing, while the traffic source is a good one;
- on a phone it is impossible to place an order or leave a request without effort;
- the platform does not allow a function you need at all, rather than making it "expensive" or "inconvenient";
- texts, prices and sections can only be changed through a developer, and every change has to wait.
If none of these applies, first fix what you found in points 1, 4 and 5. That is usually cheaper than a redesign.
When this check is not needed
If the site is long outdated, does not open properly on a phone, and the engine is no longer updated, there is nothing to calculate: it will have to be rebuilt either way. The check is for those whose site works but whose results do not.
