Hypotheses for the decline
The new order form differs significantly from the monolithic version, making it unfamiliar to users.
The current order form is long and structured as long single-page layout.
The current order form places the reprice feature at the very end, disrupting the established user pattern of checking the price after setting point B for comparison with competitors or assessing price acceptability.
Why reverting to the old form wasn’t possible
New order form for Delivery
inDrive transitioned from a monolithic to a microservice architecture, rolling out a new design alongside the codebase update. The order form in the courier vertical underwent even more significant changes. The new design included not only UI updates (new components) and an improved UX but also the introduction of a couple of new market-normal features, resulting in a completely different user experience. As a result, the V20 metric for order conversion significantly dropped by an average of 16% across all cities.
Rolling back to the previous order form was not an option, as it lacked essential new features and was built on the monolithic architecture that had already been phased out.
One of the main reasons for the drop in conversion was moving the price block to the second step, placing it at the very bottom. However, placing the price on the first step wasn’t feasible, as both the final and recommended prices depend on certain preceding fields.
Context
Version that drops metrics
A new design for the order form was introduced, enabling scalability for new features, adaptation for each country where inDrive operates, and an increase in order conversion.
IOS | Android, 2024
1
I organized a brainstorming session with the courier design team, where each member focused on a specific concept. After refining our ideas, we scheduled UX testing to evaluate them.
The order form needs an update to be clear and familiar to users. It should be visually lightweight, with optional non-essential fields, and UX-friendly, allowing completion in just a few clicks. The form should also be sales-oriented, highlighting the USP and enabling price comparison without unnecessary steps. Additionally, key fields for accurate matching and order fulfillment should be integrated in a user-friendly format.
We developed three different concepts, each with its own goals and advantages:
—A replica of the legacy form from the monolithic architecture, but with the price field moved to the beginning.
— A single-page form optimized for cognitive ease.
— A revolutionary step-by-step form replacing the previous single-step version.
—A replica of the legacy form from the monolithic architecture, but with the price field moved to the beginning.
— A single-page form optimized for cognitive ease.
— A revolutionary step-by-step form replacing the previous single-step version.
Designing the new form
USP
focused
focused
Cognitively
unloaded
unloaded
Matching
based
based
Our approaches
2
3
The percentage of users who started the prototype and successfully created an order.
The proportion of users who began filling in points A and B for a city ride but then switched to the courier section.
The share of users who attempted to create an order without filling in order details and encountered an error.
I organized a UX research study to determine which of our three proposed solutions best addressed the problem. This was a quantitative UX test conducted with real users of our app. They received a push notification and were asked to complete a task.
User Task
"Imagine you want to send flowers to a loved one using inDrive’s courier service. You need to deliver flowers from Dostyk Avenue, 82 to Abay Avenue, 2. You’d like both pickup and drop-off to be at the entrance. What would you do in this case?"
The second option won by a small margin, as it balanced a reasonable completion speed with a high rate of successful order creation. Notably, the third option had the fastest completion time, but this was partly due to the intentional absence of an error when users skipped the “Order Details” field.
While the specific questions and events tracked varied for each version, the core focus areas included:
UX research and tests in 🇰🇿
UX Test results
We significantly compensated for the drop in V20 (View to Order) conversion, achieving a +7% increase compared to the control group due to product changes.
A technical deep dive revealed that part of the conversion drop was caused by bugs related to the transition to microservices.
Results varied greatly across countries, mainly because the new supply type separation feature made certain order categories undeliverable in some cities.
The conversion rate for order creation differs across regions.
Analytics on required fields (e.g., recipient phone number, package description) and field completion sequences suggest strong regional variations.
A one-size-fits-all form with a fixed set of required fields across all regions lacks the flexibility to optimize metrics and maximize conversion in each specific market.
We launched an A/B test with the second option and got mixed results:
A/B results
Form builder
Solution: Order Form Builder
We developed a form builder that allows:
- Customizing required vs. optional fields.
- Controlling the visibility of fields.
- Adjusting the order of fields in the form.
Through a series of improvements—including the new order form and the form builder—we fully mitigated the conversion drop and even achieved a +2.3% increase in V20 after launching the builder.
Additionally, we gained valuable insights into regional courier-specific behaviors and form completion patterns, revealing that user interactions with the form often begin outside the courier section of the app.
Final Results
The design was based on the previously tested Concept #2 but with fixed spacing and field organization for consistency
