Why show a coupon before the home screen?

When I opened Taobao Instant Commerce, I did not get much time to look at the home screen. A coupon pop-up had already taken over the display.

The background was dimmed, leaving a white modal at the center. Red and pink light effects surrounded it, while small sparkles created the feeling of winning a prize. The most prominent elements were not the terms, but the numbers “24” and “13.” The spending thresholds appeared in much smaller type. Two “Use now” buttons led to the corresponding offers, while a larger button at the bottom invited me to “Accept happily.” The close control was still available, but its visual weight was clearly lower.

Before asking what I wanted to order, the product told me something else: ordering now could be cheaper.

After opening Taobao Instant Commerce several more times, however, I noticed that the coupon did not appear on every launch. Some sessions began with the offer; others went directly to the home screen. That makes the product question more interesting. The platform is not only designing a pop-up. It is also deciding when the pop-up should appear, who should see it, and how often it should return.

What looks like a simple promotion is therefore connected to a much longer product chain: how a platform captures the first second of attention in selected sessions, how it chooses its audience and frequency, how it proves that a coupon created incremental orders, who ultimately funds the discount, and which strategic objective the subsidy is meant to serve.

For readers unfamiliar with the Chinese market, Taobao Instant Commerce is Alibaba’s on-demand delivery business. It sits inside the broader Taobao ecosystem and connects consumers with restaurants, retailers, merchants, and delivery capacity. The “red packet” shown in Chinese apps is not necessarily cash. In this context, it is better understood as a promotional coupon with eligibility rules and spending thresholds.

1. Sell the saving before selling the product

The most direct function of this screen is to establish a price expectation before the user sees any merchants or products.

In food delivery and quick commerce, opening the app does not always mean the user has decided what to buy. But it does signal a degree of immediate purchase intent. Showing a coupon at that moment adds a reason to continue precisely when the intention has begun to form.

The visual hierarchy is built around that goal.

Red creates a sense of reward. Saturated motion captures attention. Large numbers create a value anchor. Smaller terms postpone the effort of understanding the rules until after the user has perceived the benefit. The sequence is deliberate: first, “I can save money”; second, “What do I need to do to qualify?”

The copy also shapes the decision. Language associated with “opening” or “winning” a red packet turns a fixed discount into a small reward event. A message inviting the user to return tomorrow gives the next visit an element of anticipation. A label such as “newly received” creates a sense of ownership before the coupon is used, making dismissal feel more like giving up something already acquired.

Both coupons in the observed screen were restricted to food delivery. This indicates that the pop-up was not only stimulating consumption. It was also using a high-traffic entry point to route demand toward a category the platform wanted to grow.

The button hierarchy pushes users forward in two different ways. “Use now” serves people who already have purchase intent. “Accept” lowers the psychological commitment for people who are not yet ready to order: they can secure the benefit first and decide later. The close control remains available, but it does not compete for attention on equal terms.

In the first second, the screen can therefore perform three jobs:

1. Establish the belief that Taobao Instant Commerce offers attractive prices;
2. Add momentum to the current browsing and purchase journey;
3. Create a reason to return through the promise of another reward.

The coupon connects immediate conversion with future retention.

2. A high click-through rate does not mean the strategy works

My first reaction was straightforward: if the pop-up is this visually dominant, does the platform simply want users to enter the promotion flow, even through an accidental tap?

That interpretation is only half right.

Strong visuals, large buttons, and a full-screen interruption will increase clicks, and some of those clicks may be accidental. But accidental interaction is not meaningful growth. If a user enters the offer page, immediately goes back, views no merchants, adds nothing to the cart, and completes no payment, the click improves a surface metric without creating transaction value.

The real question is whether the coupon moves users through the full journey:

Coupon impression
→ Claim or use
→ Enter food delivery
→ Browse merchants
→ Add items to cart
→ Reach checkout
→ Complete payment
→ Purchase again

Click-through rate and claim rate can tell us whether the page attracts attention. They cannot tell us whether the strategy creates value.

Three outcomes matter more:

• Did the coupon cause more users to complete payment?
• Does the value of incremental orders cover the subsidy cost?
• Did the interruption damage experience, trust, or longer-term retention?

Looking at only one of these outcomes can produce the wrong conclusion.

3. How can we prove that the coupon created orders?

The simplest approach is to compare conversion before and after launch. If payment conversion rises from 10% to 12%, it is tempting to attribute the two-point increase to the coupon.

That conclusion is weak. Weekends may naturally produce more delivery orders. Rain may increase demand. Lunch users behave differently from afternoon users. The platform may have launched another campaign at the same time.

A better test begins by defining which users and sessions are eligible for the coupon. Among comparable eligible users, one group can enter the home screen directly while another sees the coupon first. The experiment should cover the same cities, dates, and dayparts, with other product conditions held constant. This separates the effect of showing the coupon from the effect of the platform’s targeting rules.

Quick-commerce experiments have an additional complication: treatment and control users share the same merchants and couriers. If the coupon creates more demand, restaurants may prepare orders more slowly, courier capacity may tighten, and popular products may sell out. The treatment can therefore change the experience of the control group.

In addition to user-level randomization, the team may need experiments separated by commercial zone, geographic area, or time window. Preparation time, delivery time, cancellation rate, and capacity utilization should be monitored alongside conversion.

The primary conversion metric should use all eligible experiment entrants as the denominator:

Payment conversion = users who completed payment ÷ users who entered the experiment

Analyzing only people who clicked or claimed the coupon would select users with stronger purchase intent and exaggerate the effect.

The full funnel still matters: entry into food delivery, merchant visits, add-to-cart rate, checkout arrival, and time from app open to payment. These metrics reveal where the coupon changes behavior.

Because the pop-up does not appear on every launch, the exposure policy also deserves its own metrics: the share of sessions eligible for display, actual reach, impressions per user, conversion after first versus repeated exposure, and whether people who repeatedly close the pop-up continue to receive it.

Frequency can be tested directly. A first impression may benefit from novelty, while the second and third impressions may produce rapidly diminishing returns. Total clicks can continue rising even after incremental payment per impression has fallen below the cost of interruption.

Experience guardrails are equally important: immediate exits after the pop-up, repeated dismissals, quick returns after “Use now,” cancellations, refunds, next-day retention, and repeat purchase.

Four false positives deserve particular attention.

First, the coupon may subsidize people who would have ordered anyway. Orders do not increase; the platform simply earns less.

Second, it may pull tomorrow’s order into today. Daily conversion rises, but weekly demand does not.

Third, the campaign may increase claims without increasing redemption or payment. Users are attracted by the large headline value, then discover that the threshold makes the offer unusable.

Fourth, the coupon may move an order from another entry point or channel without creating new demand for the business as a whole.

The most valuable outcome is not total orders, but incremental orders: transactions that would not have happened without the intervention.

4. Who pays for the coupon?

Once we focus on incremental orders, a practical question follows: who funds the money the user saves?

A coupon is not value created from nothing. It may be funded by the platform’s growth budget, merchant discounts, a brand’s marketing budget, or a combination of all three.

A useful principle is that the party receiving incremental value should fund a corresponding share of the cost.

If the platform is acquiring users, reactivating dormant customers, or establishing a quick-commerce habit, it is the primary beneficiary and should carry more of the subsidy. If the goal is to bring a merchant new customers, raise basket size, or introduce a product, the merchant may participate through store coupons, threshold discounts, or promotional bundles. When both platform and merchant gain, co-funding is more sustainable than shifting the entire cost to one side.

The coupon means something different to each participant.

For users, it reduces the price of a purchase or a first trial and helps convert hesitation into action. For merchants, it is a way to buy traffic, new customers, volume, or utilization of idle capacity. For the platform, the return includes not only the order, but also activity, transaction density, merchant participation, and habit formation.

Those benefits do not always appear together. A user may save money while a merchant absorbs both the discount and the operational pressure of a demand spike. The platform may gain order volume without earning enough long-term value to cover the subsidy. Funding decisions should therefore follow where incremental value actually lands, not simply who participated in the campaign.

Merchants also adapt. To recover discount costs, they may raise platform prices, increase minimum order values, emphasize high-margin bundles, or change portion sizes and product mix. The displayed discount can become larger without meaningfully lowering the user’s final cost. A serious analysis should compare base prices, platform and in-store prices, product composition, and merchant contribution margin before and during the campaign.

We also need to distinguish displayed value from actual cost. A large headline amount may combine multiple coupons or represent the maximum available benefit. Real cost appears only when an eligible user redeems an offer. The analysis should move from advertised face value to claimed value, usable value, actual discount, and finally the share borne by platform and merchant.

One simple decision question is: how much did the platform spend for each truly incremental order?

Suppose 10,000 users generate 1,000 orders without the coupon and 1,200 with it. The incremental volume is 200 orders. If the platform spends an additional RMB 6,000, each incremental order costs RMB 30.

If that order and the customer’s future purchases create only RMB 8 of value, the strategy is not sustainable even though order volume increased. If the acquired customer develops a stable repeat-purchase habit, a higher first-order investment may be rational.

Coupon size should therefore be constrained by incremental transaction value and customer lifetime value—not by how large a competitor’s headline number looks.

5. The coupon is ultimately a platform strategy

At this point, the coupon is no longer merely a UI component. It is the product expression of a business strategy.

If Taobao Instant Commerce prioritizes market expansion, it may trade margin for users, orders, and supply density. If it wants to build an on-demand shopping habit, the coupon may also be evaluated through visit frequency and cross-category purchase. If the business enters an efficiency phase, broad subsidies should give way to offers targeted at users whose behavior can actually be changed.

Different strategic goals create different product rules:

• Market expansion emphasizes new users and order growth;
• Habit formation emphasizes visit frequency and repeat purchase;
• Supply development directs traffic to selected merchants, cities, or periods;
• Efficiency management emphasizes contribution margin and incremental order cost;
• Competitive defense may use intense but limited offers.

For a multi-sided marketplace, the coupon can also balance supply and demand.

When merchants and couriers in a zone have spare capacity, a targeted discount can pull demand into that period. Merchants receive more orders, staff and equipment are better utilized, couriers may receive a steadier flow of work, and greater transaction density may improve dispatch efficiency. The user receives a lower price for a purchase that might not otherwise have occurred.

During an already overloaded lunch peak, the same subsidy can create the opposite result: slower preparation, insufficient courier capacity, late delivery, refunds, and poor ratings. The user saves a small amount but waits much longer. The correct product response may be to reduce the offer, narrow its audience, or route demand toward merchants and areas with available capacity.

The coupon’s value therefore depends not only on who receives it, but also on when, where, and toward which supply it directs demand. The same offer may create incremental value off-peak and amplify service failure at peak. It may help a new merchant overcome cold start while making an already overloaded popular restaurant worse.

Product managers may not decide the company’s total subsidy budget, but they translate strategy into executable rules: eligibility, timing, frequency, thresholds, destination after click, and the conditions under which the offer should be reduced or stopped.

The actual product is not just the pop-up. It is the exposure system that decides who sees it and when.

My repeated visits produced selective exposure: sometimes the coupon appeared; sometimes it did not. The underlying rules could relate to claim status, campaign lifecycle, user state, or frequency controls. The more important product question is whether those rules reserve the discount for users whose behavior is likely to change.

New customers may complete a first order because of the coupon. Dormant customers may return. Frequent users may have ordered anyway, making another subsidy mostly a margin loss. Repeated dismissal is also feedback. If the system fails to lower frequency, short-term impressions can turn into long-term irritation.

A more mature strategy identifies users who may respond to a small incentive and controls exposure: do not repeatedly interrupt someone who already claimed the offer, reduce contact after repeated dismissal, and use a lightweight expiry reminder instead of covering the home screen again.

Precision introduces a fairness problem.

An inactive user may receive a larger coupon than a loyal customer precisely because the inactive user is easier to influence. The platform sees efficient budget allocation; the loyal customer may see a penalty for loyalty. Similar products can effectively have different prices for different users.

Merchants face another fairness issue. Large chains and high-margin merchants can afford deeper discounts and may receive more traffic, while small low-margin merchants struggle to compete. If promotional intensity replaces product quality and service as the main exposure criterion, the marketplace may become increasingly dependent on subsidies.

Geography matters too. The same coupon face value does not create equal value when one area has many eligible merchants and another has almost none, or when local basket sizes make the threshold easy for one user and unrealistic for another.

Fairness does not require identical discounts for everyone. It requires explainable differences, durable benefits for loyal users, meaningful discovery opportunities for smaller merchants, and protection for couriers and frontline workers from bearing only the workload, delay risk, and rating pressure created by demand stimulation.

The platform should not optimize only for the cheapest incremental order. It must also consider whether the distribution of benefits and pressure can sustain trust.

6. What does this screen get right—and is there a better option?

From a conversion perspective, the launch coupon has a clear design.

It captures a moment of relatively high intent, uses color and type size to establish the value quickly, and offers both an immediate action and a lower-commitment claim action. The promise of another reward extends a one-time incentive into a reason to revisit.

The first screen is among the product’s most valuable attention resources. By leading with savings instead of speed, assortment, reliability, or service guarantees, Taobao Instant Commerce makes a strategic choice about brand perception. In the short term, low price can drive browsing and orders. Over time, repeated but controlled offers may build the belief that the service is more affordable and make it more likely to be considered when a delivery need arises.

That strategy also creates four risks.

First is interruption. For sessions selected by the exposure system, the user must process the platform’s promotion before continuing the original task.

Second is comprehension. The discount value is enlarged while thresholds and rules are visually reduced. If the user remembers “24” but can redeem only a fraction of it, expectation and reality diverge.

Third is trust and subsidy dependence. A large occasional discount can create delight, and a reliable discount can build a value position. But if the headline repeatedly proves difficult to use, the promotion becomes noise. If the platform builds no parallel advantage in speed, supply, or fulfillment, users may leave when the subsidy shrinks.

Fourth is operational capacity. If demand created by the coupon exceeds merchant and courier capacity, conversion growth turns into delay, refunds, and negative ratings.

Accessibility matters as well. Small rule text creates difficulty for older users. Strong animation can distract or discomfort. Heavy dependence on red for state communication can reduce clarity for some users with color-vision differences. A high-exposure modal should make terms, actions, and the close control understandable to users with different abilities.

The downstream experience I observed was relatively complete. “Use now” led to eligible merchants and products. At checkout, the system matched available discounts and separated product promotions, store discounts, platform coupons, packaging, and delivery fees. Cases such as ineligible addresses or failed payment also received feedback. The launch offer was therefore connected to selection, checkout, and exception handling instead of operating as an isolated click generator.

But even an effective coupon does not necessarily need to appear first.

If the goal is awareness, a home-screen coupon card creates less interruption. If the goal is to convert hesitation, an offer after the user enters food delivery or browses for a while may be more precise. If the goal is to cross a price threshold, a cart reminder is closer to the decision. If the goal is to reduce checkout anxiety, automatically applying the best coupon may be more direct than asking the user to claim first and search for eligible supply later.

Every placement has a cost. A launch pop-up has maximum visibility but interrupts every selected session. A home card is calmer but may reduce claim rate. A category-level prompt better matches intent but cannot influence users who have not entered the category. Automatic checkout savings reduce effort but lose the ability to motivate browsing.

The experiment should therefore compare not only “pop-up” versus “no pop-up,” but also different moments in the journey. A stronger system may combine them: first-time or dormant users receive a launch offer; frequent users with clear intent enter the home screen directly; people who already claimed are not interrupted again; hesitant users receive a reminder near merchant or cart decisions; checkout automatically applies the best available benefit.

The product goal is not the highest click-through rate. It is the highest combined value across conversion, subsidy cost, supply conditions, fairness, and user interruption.

Conclusion

When I first saw the page, the logic seemed simple: show a coupon at the start and encourage an order. Repeated use revealed another layer. Deciding who should see the coupon—and how often—is itself part of the product strategy.

The screen connects visual attention, transaction behavior, incremental measurement, funding allocation, marketplace supply, targeting, fairness, accessibility, and alternative placements.

That is what makes product teardown valuable. The visible object may be a button, a number, and a coupon. The real system consists of user behavior, marketplace mechanics, and strategic trade-offs.

The final question should not be “How many people clicked?” It should be:

Can Taobao Instant Commerce deliver the right incentive to the right user and supply, at the right moment and frequency, in a way that is explainable, minimally disruptive, operationally supportable, and economically sustainable for users, merchants, couriers, and the platform?

If the answer is yes, the coupon is more than a promotion. It is an effective product strategy.