I have traffic but no sales: a 12-point diagnostic

Twelve checks to find out why your store gets traffic but no sales: measurement, final pricing, checkout, payments, and accessibility. With specific thresholds.

It is almost never the design. It is the measurement, the final price, or the checkout. Go through these twelve checks in order: first, verify that your analytics are tracking correctly, then measure the funnel friction, and only at the end touch the aesthetics. Each check has a concrete threshold and can be answered with a yes or no in an afternoon.

We receive the same message several times a month. Decent traffic, flat sales. And almost always, the request that accompanies it is the wrong one: redesigning the store.

We start by measuring, not redesigning. A redesign without a diagnosis just moves the symptom elsewhere.

This text is a self-administered diagnostic. Twelve checks, four blocks, in the order they must be performed. Do not skip them: check 9 means nothing if check 1 is broken.

Before you start: which number are you looking at

The conversion rate of a store is orders divided by sessions. That aggregate number is useless for diagnosing anything because it mixes brand traffic with cold traffic, mobile with desktop, and after-sales service visitors with actual shoppers.

What is useful is the segmented number. Non-brand organic traffic conversion, on mobile, on the product page. That one can be compared and that one can be fixed.

We publish our own records instead of citing international averages that do not distinguish by country or channel. And a warning about the number that is verified and that everyone quotes incorrectly.

That 70% is not your problem: it is the industry floor. Baymard also attributes 42% of abandonments to people who were just looking. Chase the remaining 58%, not the headline.

1. Does your analytics count what you think it counts?

Since the consent banner became mandatory, a portion of the traffic does not enter the analytics. If your store uses standard tagging without a properly configured consent mode, the sessions you see and the sales you see do not come from the same universe.

Check: compare the orders from your store dashboard with those in your analytics over the last 90 days. If the deviation exceeds 5%, your conversion rate is fiction and the next nine checks have no basis.

Fix the measurement before the store. A miscounted number is not optimized: it is corrected.

2. Are you looking at the aggregate or by channel and device?

Separate at least four segments: brand organic, generic organic, paid, and direct. Cross each one by mobile and desktop.

The pattern we see most often: desktop within range and mobile tanking. That is not a traffic problem; it is a checkout problem on small screens.

3. Is your checkout instrumented step-by-step?

You need the number of sessions that reach each step: product page, cart, data, shipping, payment, confirmation. Without that, "I am not selling" is a complaint, not a diagnostic.

With that, "I lose 61% between data entry and shipping" is a task with a deadline.

• Orders in analytics and store dashboard differ by less than 5%
• You have conversion separated by channel and device, not just the aggregate
• Each checkout step emits its own event and you can see the drop-off between steps
• You know what percentage of sessions reject cookies and take it into account when reading the data

4. Do your visits come to buy or to get informed?

Some stores rank well for informational queries and count them as commercial traffic. An article on how to clean a suede jacket brings visitors who were never going to buy today.

Check: take the ten queries that bring you the most sessions and classify them into buy now, compare, or learn. If more than half are for learning, you do not have a conversion problem: you have a mix problem. The fix is content and internal linking, not buttons.

5. Is the product available and findable?

Count how many clicks separate the home page from the best-selling product. If it is more than three, the navigation grid is working against you.

And check out-of-stock items. A product page with no stock that offers no alternative or replenishment notification is a wasted visit.

6. Is the price the user sees the final price?

This is the check with the highest return of the twelve, and it is also the only one that is a legal obligation.

Article 60 of Royal Legislative Decree 1/2007 requires informing the consumer before they are bound of the full final price, including taxes, breaking down increases, discounts, and additional expenses. In other words: the shipping cost cannot appear for the first time in the final step.

And the 21% general VAT (Law 37/1992, art. 90) must be included in the price shown if you sell to a final consumer, not announced separately.

The industry standard: Clean price on the product page, shipping revealed in step 4 of the checkout. The user recalculates, feels cheated, and leaves. The store registers an abandonment and blames the design.

Our criteria: Price including VAT on the product page, shipping cost and free shipping threshold visible on the same screen, shipping calculator before the cart. The drop-off happens earlier, costs less, and is honest.

7. Can one buy without creating an account?

Baymard places abandonments due to the obligation to register at 18% and those caused by a checkout that is too long or complicated at 17%.

Binary check: open your store in a private window and try to buy without registering. If you can't, you already have your first task for the quarter.

8. How many fields does your checkout have?

Count them. Literally, with your finger on the screen.

Each field is a decision you are asking someone to make who had already decided to buy. The second surname, the company name, the "how did you find us?" and the landline phone are fields a form inherits and no one reviews.

We publish our own count here because external averages mix markets with very different postal addresses.

— = data being measured; we will publish it once we have our own sample.

9. How long does your product page take on mobile?

The thresholds are not a matter of opinion. Google defines them and measures them at the 75th percentile of loads, separating mobile and desktop: LCP equal to or less than 2.5 seconds, INP equal to or less than 200 milliseconds, CLS equal to or less than 0.1.

Measure with field data, not with a lab test from your laptop on fiber. And pay special attention to the CLS on the product page: a cookie banner that pushes the buy button half a second after the finger was already moving toward it produces erroneous clicks that no analytics will explain to you.

A calibration warning, because the industry exaggerates here. Going from 4 to 2 seconds in a store with a poorly filtered catalog does not fix the catalog. Performance amplifies what is already working; it does not replace what is missing.

10. Are the payment methods the ones your buyer uses?

In Spain, this is not solved by copying what a US store does. Here, card via Redsys with the bank, Bizum, and international gateways coexist.

Stripe's public commissions for Spain, consulted today, serve as a comparable reference:

Source: Stripe Spain pricing page, consulted 08-27-2026. Currency conversion adds an additional 2% when applicable.

Against that, a virtual POS with your bank through Redsys is negotiated on a case-by-case basis and does not publish a comparable rate. The two sources are not homogeneous and anyone who presents them as an equivalent table is oversimplifying.

Gateway commissions and platform plans verified 08-27-2026. These are list prices that change without prior notice and are negotiated by volume: check them before deciding, do not cite this article as the current price in six months.

11. Are shipping, delivery times, and returns where people look?

Article 71 of Royal Legislative Decree 1/2007 sets the withdrawal period in distance contracts at 14 calendar days. That right exists whether you publish it or not. The difference is that publishing it sells, and hiding it costs.

Check: on the product page, without scrolling on mobile, are the delivery time, shipping cost, and return policy visible? If you have to look for it in the footer, it's not there.

12. Can it be purchased with a keyboard and screen reader?

Law 11/2023, of May 8, which transposes Directive (EU) 2019/882, set June 28, 2025, as the effective application date for e-commerce accessibility requirements. This is not a style recommendation: it is an obligation with a sanctioning regime.

Quick check, five minutes: put the mouse away and complete a purchase using only the tab key. If you get lost in the size selector, the province dropdown, or the cookie banner itself, there are real buyers getting lost there every day.

A purchase that cannot be completed with the tab key is not an inconvenient purchase: it is a sale that does not exist.

In what order to fix it

The expensive mistake is not picking the wrong check: it is getting them all right and attacking them all at once.

And a disclaimer, because the industry promises the opposite. We cannot guarantee a percentage improvement: anyone who guarantees it without having seen your funnel is selling you a number, not a diagnosis. What we do agree on in writing is the fixed price per phase before starting, with no surprise hours.

A store that receives visits and does not sell is not broken on the inside: it is asking for too much at the wrong time. And these twelve checks are not a list of improvements: they are an order. Check 4 is not touched as long as 1 still does not match.

Do the first three today and keep two things: the 90-day order reconciliation between analytics and dashboard, and the field count of your checkout. Send us those two numbers and we will tell you, in the first 30-minute cost-free call, which of the four blocks is yours.

Frequently asked questions

What is a normal conversion rate for a Spanish store?

It depends on the sector, the channel, and the device, and any single figure you are given mixes all three. We publish ours segmented with visible n and period: still being measured.

I have a lot of traffic and zero sales. Is it the design?

Almost never. Start with check 1: if your orders in analytics and in the store dashboard do not match, the problem is that you are not measuring. If they do match and the drop-off is between the cart and payment, it is the checkout, not the aesthetics.

Is it worth it for me to change platforms to sell more?

Changing platforms solves cost and maintenance problems, not conversion. If your checkout asks for fourteen fields, it will ask for them just the same on the new platform.

How long does it take for a checkout fix to be noticeable?

The effect on the funnel is seen in days; reliable reading needs enough volume to not confuse noise with improvement. If your store does not reach a hundred orders a month, we will tell you: with that volume, you will not be able to distinguish a real improvement from a good week.

What if the problem turns out to be the product or the price?

We will tell you and we will not take the project. Redesigning the shop window for an offer that the market does not want is spending your budget on having the problem better presented.

Sources

• https://web.dev/articles/vitals — "good" thresholds for Core Web Vitals: LCP ≤ 2.5s, INP ≤ 200ms, CLS ≤ 0.1, evaluated at the 75th percentile and separately on mobile and desktop. Verified 08-27-2026.
• https://baymard.com/lists/cart-abandonment-rate — average cart abandonment rate 70.22% aggregated from 50 studies (2006-2025, range 55%–84.27%); reasons for abandonment: extra costs 40%, slow delivery 20%, security 19%, mandatory registration 18%, complex checkout 17%, technical errors 17%; 42% were just looking. Verified 08-27-2026.
• https://stripe.com/es/pricing — published commissions for Spain: standard EEA card 1.5% + €0.25, premium EEA 2.8% + €0.25, outside EEA 3.15% + €0.25, SEPA debit €0.35, +2% for currency conversion. Verified 08-27-2026.
• https://www.boe.es/buscar/act.php?id=BOE-A-2007-20555 — Royal Legislative Decree 1/2007: art. 60 (prior information of the full final price with taxes and additional expenses) and art. 71 (14-day withdrawal period in distance contracts). Verified 08-27-2026.
• https://www.boe.es/buscar/doc.php?id=BOE-A-2023-11022 — Law 11/2023, of May 8, transposition of Directive (EU) 2019/882: effective application of accessibility requirements since June 28, 2025. Verified 08-27-2026.
• VAT Law 37/1992, art. 90 — general rate of 21% applicable to sales to final consumers. Regulatory reference cited in the text.