The cart reminder nobody has measured

Every guide to abandoned carts hands you the same three intervals. Send the first reminder within one to three hours, a second within a day, a third a few days later if the customer still has not come back. I went looking for the study behind those numbers, because a figure repeated that consistently usually has a field experiment underneath it that everyone is citing at second hand. I did not find one.

What is underneath instead is a chain of citations. Each page quotes the page before it, and the furthest back the thread goes is an aggregate published by an email service provider about its own customers, with no stated methodology, no described sample and no control group to measure the result against. You cannot set a send time on a number like that, and you certainly should not hand it to a clothing brand as though it were research. The useful material sits somewhere none of those pages point at: in two Saudi instruments, one governing what you show before the customer pays, the other governing what you send after they leave.

What a reminder actually does

A cart reminder, when it works, works as a bookmark rather than as persuasion. The shopper who put a hoodie in the cart and then lost signal, or got called away, did not reject the piece. They lost the path back to it. The message restores that path and nothing more: the piece they chose in the size and colour they chose, a link that opens the cart as they left it, and one line saying what is left to finish.

That explains why a reminder can work. It says nothing about when to send one. The interval between the abandonment and the message is a purely empirical question whose answer moves with price, with category, and with how people buy that kind of garment, and it resolves only through a test run on your own list. The hard part of that test is not the timing but the control group. Some shoppers come back on their own, with no message at all, and unless you hold a slice back and send them nothing, your dashboard will credit the reminder with revenue that was arriving anyway.

The price the law puts before checkout

Here the guesswork stops and a written text begins. Article 7 of the E-Commerce Law requires the store to give the consumer, before the contract is concluded, a statement of the contract terms: the steps to conclude it, the provider's details, the essential characteristics of the product, the total price inclusive of all fees, taxes and delivery-related amounts, and the payment, delivery and performance arrangements. A shipping charge that shows up for the first time at the final step is not only a friction problem.

Which makes the shipping line something to settle before the product page is designed rather than after. What does delivery cost, when does it arrive, and is the number the same in Riyadh and in Abha. A piece printed on demand leaves a production line and then enters a shipping network, and those are two durations rather than one, while the buyer reads only the total. How to reduce shipping costs deals with the cost itself. The cart cares about something narrower: that the figure appears in time, and appears whole.

Consent comes before the reminder

The message you send after an abandonment is promotional material, and it falls under the Personal Data Protection Law, issued by Royal Decree M/19 and in force today. Article 25 bars a controller from using a data subject's personal contact means, postal and email addresses included, to send promotional material without that person's consent, and requires a clear mechanism for stopping the sending. The penalty is not nominal: Article 36 sets a fine reaching SAR 5,000,000, doubled for a repeat breach.

The same logic reaches the pixel you install for retargeting. Article 5 of the Law bars processing personal data without the subject's consent except in the cases the Law provides, and the Implementing Regulation sets what that consent has to be: freely given, not obtained by misleading methods, with clear and specific purposes. Because Meta, Google, TikTok and Snap all process what reaches them outside the Kingdom, the rules on transferring personal data outside the Kingdom apply on top of the consent question.

Saudi Arabia has no cookie-specific law. Neither the Law nor its Implementing Regulation contains the words cookie, tracking, pixel or profiling. What does exist is SDAIA's privacy policy guideline, which names cookies as an indirect collection method to disclose, so legislative silence is not a licence to stay silent on your own page.

A retargeting window also closes, however wide you open it. A Meta website custom audience has an advertiser-set retention measured in days, up to a maximum of 365 days, after which a visitor drops out unless they return to the site and match again. An audience built on incomplete consent stays a live exposure for that whole period, not only on the day the code went in.

When checkout stops at a plan line

Before blaming the design of your checkout, look at what your plan permits. Salla Basic is free and does not include electronic payment; that starts on Salla Plus at SAR 990 a year. Zid's free Starter plan is limited to bank transfer and cash on delivery, with mada, Visa, Mastercard and Apple Pay arriving one tier up at SAR 99 a month or SAR 990 a year.

Shopify has no free tier, and Shopify Payments does not cover Saudi Arabia, so a Saudi merchant runs a third-party gateway and pays Shopify's third-party transaction fee on top: 2% on Basic, falling to 0.2% on Plus. Those pages were read in September 2026 and plan pricing moves, so the number that matters is the one published on the day you read it.

Cash on delivery is the question that returns in every Saudi conversation about abandoned carts, and the measured answer does not exist. No authority publishes the share of orders paid on delivery, and every figure in circulation traces back to marketing content rather than to a primary source. What is published is the direction of travel: SAMA puts electronic payments at 70% of retail payments in 2023, 79% in 2024 and 85% in 2025, and reports e-commerce spending on mada cards at SAR 69.3 billion in the first quarter of 2025 against SAR 98.4 billion in the first quarter of 2026. A trend does not tell you what your own customer will do. It does tell you that the list of payment methods in your store is a decision taken on a pricing page long before its effect shows up at checkout.

What the product page has to settle

A good deal of what gets counted as cart abandonment begins one step earlier. A clothing brand sells something that has to be worn, and the buyer wants the size, the fabric weight, and a sense of how the piece sits on a body like theirs. If the page does not answer, the decision gets postponed and the cart becomes a parking space until someone can ask. A piece printed on demand needs that clarity more than a stock item does, because the buyer cannot handle it before it is made, and cannot return it as casually as something pulled off a shelf.

The return policy gets read in the moment of hesitation, not after the purchase. When the Ministry of Commerce published the ten criteria it evaluates online stores against, announcing first-wave results across 100 stores in September 2025, the exchange and refund policy, the consumer data protection policy, site speed and reachable customer service were all among the things being measured. Write your return policy in two lines a buyer can follow, and put it where the hesitation happens instead of burying the link in a footer.

Where Tamm fits

Tamm is a factory and a warehouse in Saudi Arabia, and the print-on-demand service provider behind your store. You keep selling on the platform you already use; when an order arrives we print the piece on our own line and ship it to your customer under your brand. It is not a store platform and it does not compete with Salla, Zid or Shopify. Start with the catalog.

The only number you own

What stayed with me, after all of it, is that the most repeated figure in this field is the one number nobody has measured on your store. The first hour, the second message, the third day: all of it is inheritance, passed from one writer to the next, and your store is the only laboratory where the claim can actually be tested. Hold back a slice that receives nothing, change one variable at a time, and read the difference between the two groups rather than the total that came back. Then ask the harder question. Is the cart stalling because the message was late, or because the page never told the buyer exactly what they would pay and exactly when the piece would arrive?