Your customer clicked. What are they still worried about?

You are paying for clicks, people are reaching the product page, and the orders are not following.

So you start wondering whether the website needs a redesign, whether the ads need a better hook, or whether another discount would get people moving. Those are expensive answers to a question that has not been asked properly yet.

Before you spend money on any of them, walk through the purchase yourself. Open the live ad on your phone, follow the link, pick a product, and go all the way through checkout. Then compare the orders sitting in your store against what your ad reporting says you got, because you do not want to spend a month fixing a sales problem that turns out to be a tracking problem.

What follows is how I rank the usual options when an ecommerce campaign gets clicks but struggles to turn them into purchases. These are priorities to investigate rather than guaranteed results. S tier deserves real attention and real budget. F tier is almost always the wrong first move.

Every fix on this list is ranked by the same test, and it is worth stating plainly before the list starts, because it is the part that carries over to whatever situation you are in.

We ask two questions about any proposed fix.

The first is whether the people you are trying to convert will even see the change, on the path they take. A click on a product ad lands on a product page. A change made somewhere else on the site is invisible to that buyer no matter how good it is.

The second is whether the change removes a reason not to buy, or only adds another reason to want. The click already proved that the buyer wants the thing. What stopped them was something they did not know, could not find, or did not trust.

A fix that answers yes to both questions belongs in S or A tier. A fix that fails the first question belongs in D or F, however good the work itself may be. That is the whole ladder, and every rung below is that test applied once more.

I wouldn't automatically blame the hook when people are already clicking. The hook did its job. It failed the second question before we even ask the first one, because a new opening line adds another reason to want rather than removing a reason not to buy.

There is one exception, and it is the only reason this sits at F rather than off the list. An ad can attract people who liked the video, wanted the giveaway, or misread the price, and getting attention is not proof that you are reaching buyers. So look at what the ad promises and at who that promise is likely to attract. Is the product clear? Are you holding back a price that would change someone's decision? Does the creative suggest the product does something it does not?

If the wrong people are clicking, the ad moves straight to the top of your list. If the right people are arriving and getting stuck on the product page, another opening line is a guess, and you should find out what they are missing before you write it.

A home page redesign is hard to justify as a first fix when your campaign sends people somewhere else. It fails the first question outright. If you are advertising a particular jacket, the work belongs on the jacket's product page. If you are advertising a collection, it belongs on the collection page.

None of that means a home page is never a sensible destination, and visitors do use it to work out who you are. But before you approve a redesign, ask whether the people you are trying to convert will ever see it. If the answer is mostly no, there is a better place to start.

The discount wheel sits near the bottom of the list, and the reason is the second question rather than the first. A popup certainly lands where the buyer is. What it adds is another reason to want, in the form of a code, at the exact moment the buyer is still trying to understand the product.

It will collect email addresses, and usually more of them than a plain signup form does. That gives you a number that goes up, which is not the same as a business that gets better. The test I care about is whether those subscribers become profitable customers. Compare the wheel against a plain signup offer, follow both groups to their first order, and account for the discounts you gave away to get them.

Then look at the experience on a phone. Does the popup cover the product before anyone has had a chance to understand it, is it easy to close with a thumb, and is it competing with a second signup form or a sitewide promotion? There is nothing wrong with testing an incentive, but I would not install one and call the conversion problem solved.

A discount deserves a clear job. Clearing old stock is a job. Giving someone a reason to try a starter order is a job. Being nervous because sales are slow is not a strategy, and a discount offered in that spirit trains people to wait and eats at your margin at the same time.

Take a simple example, and treat it only as an example. You sell a product for $100 and the product itself costs you $60, which leaves $40 before advertising, fulfillment, payment fees and everything else. Offer 20 percent off and the selling price falls to $80, which leaves $20 before those same costs. The customer saves a fifth. You give up half of the gross profit you were working with. We wrote the full version of that arithmetic up in our profit math article, and it is worth reading before you sign off on a promotion.

There is a longer cost as well, and it does not appear in this month's report. A study published in Marketing Science that tracked advertising and promotion over many years reported that increases in price promotion were unprofitable over the long run. People learn your discount schedule and they wait for it.

The promotion can still be the right call. It has to earn its place, and you have to measure what you keep rather than how many orders arrived. Before you lower the price, check whether price is the problem at all. A cheaper jacket still does not answer a question about whether it fits.

This is a real improvement and not a cosmetic one, and it passes both questions. A slow product page is in the path, and waiting is a genuine reason not to buy.

In a published A/B test, the Japanese retailer Rakuten 24 put a faster, more stable version of one landing page against the original, with no other change to how the page looked or worked. The faster version produced a 33.13% higher conversion rate and 53.37% more revenue per visitor. Vodafone ran the same kind of test on one of its own landing pages and reported an 8% increase in total sales. Those were real experiments rather than observations, which is why I am willing to say the speed caused the difference. They are not a promise that your store will see the same numbers.

I would start with the obvious problems. Oversized images, scripts nobody remembers adding, and features that make the page slow or awkward to use on a phone. There is a large difference between making a slow store usable and rebuilding an entire site to chase a better test score. If shoppers can barely load the page or tap the buy button, speed moves to the top of the list. Otherwise, weigh more work against the other problems you have found. B tier does not mean unimportant. It means fix the real problem, then decide whether more is worth it.

Put the ad and its landing page side by side and ask whether somebody would immediately recognize the product and the offer they clicked for.

Picture an ad for a lightweight jacket that packs into its own pocket. The product page opens with a large lifestyle photograph and a headline about everyday adventure, and the packable feature is halfway down. I would move that feature up, show the jacket packed, and say how big the bundle is, so the customer can pick up where the ad left off. The same applies to price, bundles, delivery offers and availability. Google recommends making an advertised offer easy to find and easy to buy once somebody arrives.

You do not need to repeat every sentence from the ad. You do need to follow through on the reason somebody clicked. For a small mismatch this is a useful improvement. For an ad that promises something the page does not deliver, it is urgent, and it fixes more than everything above it on this list.

I would rather show a shopper relevant customer feedback beside the product choices than expect them to find a separate testimonials page. A trust signal where the decision is happening is worth more than the same signal one click away.

Northwestern University's Spiegel Research Center looked at how shown reviews relate to sales. It reported that the purchase likelihood for a product with five reviews was 270% greater than for a product with none. That is a study of what shoppers did rather than a controlled test, so read it as a strong link and not a guarantee. It also says nothing about where reviews belong on the page, which is why placement is a test rather than a rule.

My starting point is the real rating and the real review count near the product title, with an obvious route to the full reviews. Then I test whether a relevant review sitting beside the purchase controls helps as well. Relevant is the word doing the work there. For a jacket, look for feedback about fit, warmth and how it holds up in the rain. For a bag, look for what fits inside and how it feels after a full day. Use genuine customer feedback and keep the whole collection reachable. The goal is to help somebody evaluate the product, not to surround the button with praise.

How much is shipping? When will it arrive? What happens if it does not fit? Those are reasonable things to want to know before spending money, and every one of them is a reason not to buy that you can remove in an afternoon.

Baymard Institute asks shoppers why they abandoned a checkout. In their survey, 40% said the extra costs, meaning shipping, tax and fees, were too high, and 13% said the returns policy was not satisfactory. Those are reasons people give rather than causes anyone measured, and they both point at the same place. The cost and the conditions turned up too late to be part of the decision.

I'd put a short, readable block near the price and the product options. For a clothing store it might read like this.

Your wording has to match your real delivery estimates and your real policy, the details need to be one tap away, and the sizing guidance needs to be useful rather than decorative. Do not squeeze it onto one thin line to keep it above the fold either. On a phone, a few readable lines beat a strip of text nobody can comfortably read. Customers do not read carefully, so make the important answers easy to find while somebody is deciding.

Shop with confidence tells me nothing. A clear explanation of what happens if the size is wrong tells me a great deal, and it removes the reason not to buy instead of decorating it.

A meta-analysis in the Journal of Retailing gathered the research on return policies and how strict they are. It reported that easing a policy was linked to a bigger rise in purchases than in returns. That is worth knowing before you assume a generous return window will be abused.

Start from the concern you are trying to address. Is the worry about fit, about damage in transit, about whether the product works with something the customer already owns, or about whether it will do what you say? Then decide what you can offer. It might be a first size exchange with the shipping covered, or a clear replacement process for a damaged delivery. State the conditions plainly, including any cost to the customer, because a guarantee with an unpleasant surprise attached is worse than no guarantee at all.

This is where I would spend the most thinking time, and it is the only item on the list that can change what the business sells rather than how it is described. The usual advice stops at "listen to your customers", which is not something anybody can act on. So here is how we run it, and it takes about a week.

1. Pull the last fifty pre-purchase messages. Support email, live chat, Instagram and Facebook direct messages, comments under the ads, and the reason field on returns. Pre-purchase only. A complaint about a late delivery is a real problem and a different one.
2. Write one line per message, recording the question the person asked rather than your interpretation of it.
3. Tally those lines by theme. Fifty is enough, because the themes start repeating well before you get there.
4. The top theme is your objection. If the top two are within a message or two of each other, treat both as live.
5. Answer it first where the decision is made, as one line on the advertised product page. Do not touch the offer yet.
6. Give it two weeks. Watch the share of product page visits that add to cart, and watch whether that question keeps arriving in support.
7. If the question keeps arriving after the page answers it plainly, the objection is not about your copy. It is about the product, the offer, or the audience. That is the point at which we change what is sold.

That last step is the one that matters, because it separates a writing problem from a business problem, and most stores never separate the two. Say you sell a concentrated cleaning product and shoppers keep asking how much they need. If explaining the dilution on the page stops the question, you had a writing problem. If people keep asking anyway, you have a product problem, and the answer might be a starter kit with a marked bottle, a smaller trial size, or a different audience.

Sometimes the honest answer is that the product is not right for that customer. Better copy should not disguise that, because there is no prize for persuading somebody to buy the wrong thing and return it a week later.

This deserves its own line because it will not fix today's missing first orders. It is here because the second order carries no advertising cost, and that is what makes an expensive first sale survivable.

Once people are buying, the question is whether there is a good reason for them to come back. For a product with a natural repeat cycle, I would look hard at the experience after delivery. Did the customer understand how to use it, did it match what the ad promised, and were problems handled properly? A refill reminder makes sense for a consumable and a relevant accessory makes sense for something durable. A generic promotion three days after delivery makes sense for neither.

Measure repeat purchases over a window that fits the product, compare customers who have had the same amount of time to come back, and account for the cost of the discounts, the messaging and the fulfillment. Do not assume every first-time customer will become a repeat buyer. Give them a good experience and then see what happens.

The ladder tells you what is usually worth doing. Your own analytics tell you which rung to start on, and this is the table we work from.

For scale, Baymard Institute puts the average documented cart abandonment rate at around 70% across fifty studies, and Littledata put the average checkout completion rate for Shopify stores at 45% across 2,800 sites in 2023. Those are the industry backdrop. The numbers below are ours, and they are the first read we use rather than a benchmark anybody should be measured against.

We start treating an add-to-cart rate on paid traffic as a product page problem when it sits below roughly 3 to 5 percent. We read fewer than about four in ten carts reaching checkout as a cost surprise. We read fewer than about half of started checkouts finishing as friction inside checkout. And we read fewer than about one in six customers buying again inside the product's own repeat cycle as a problem with what happens after delivery. They are ranges on purpose, because a store selling mattresses and a store selling coffee should not be held to the same line.

Most stores cannot do all of this at once, so here is the order we use when there is one thing at a time.

In the first week, spend nothing. Walk the purchase on your phone starting from the live ad, confirm that the orders in your store match what the ad platform is reporting, and read the last fifty pre-purchase questions your customers asked.

In the first month, make one paid change, and make it the A tier your own numbers point at. Most of the time that is the shipping, returns and fit block on the products you are advertising, the real rating beside the buy button, and the ad-to-page match on the campaign spending the most money.

In the first quarter, work through the rest in order. The guarantee you can support, then the offer rebuilt around the objection you keep hearing, then the second-purchase path. Speed sits outside that order. Fix it when the page is slow on a phone, start with the images and the scripts, and do not rebuild the store for it.

The things we would not fund first are the home page redesign, the new hook and the bigger discount. Each of them can be the right call later. None of them is the right call while the product page still leaves a question open.

Before your next conversation about refreshing the creative, take the journey your ad asks somebody else to take. Write down everything you had to search for, guess at, or leave the page to understand, then compare that list against what customers have been asking you.

You may find the ad needs work. You may find the product needs work. You may also find something far smaller, like a missing measurement, an unclear policy, or an important feature buried under three paragraphs of brand writing. I'd rather investigate that than offer everyone another 20% off.

If you want another set of eyes on your ecommerce ads and your product pages, talk to us about what to investigate first.