In the world where user preferences and behaviors are constantly evolving, It’s crucial for companies to make informed decisions to remain relevant and successful. One of the most effective ways to achieve this is through research, which provides insights into user needs, pain points, and expectations. However, research is not just about gathering data — it’s about asking the right questions and testing the right assumptions. This is where the power of researching and well-formed hypotheses comes into play.
Principles for Selecting Research Methods
To select the best research method, it’s important to define the research goals and hypotheses, as well as the specifics of the product and user behavior. Here are three main principles for choosing a method:
1. Define Research Goals — this could involve testing a user preference hypothesis, identifying interface issues, or assessing new functionality. Depending on the hypothesis, either qualitative or quantitative methods may be more appropriate. For significant changes, both types may be needed: qualitative to gauge user reactions to new ideas, and quantitative to validate the results.
2. Segment Users — consider differences in behavior, experience, and motivation among user groups. For instance, preferences may vary between new and returning users.
3. Consider Resource Constraints — methods differ in terms of effort and cost. Quick checks work well with surveys and online tests, while in-depth behavioral analysis requires interviews and observations.
Common UX/UI Research Mistakes and Solutions: A Practical Guide with an Analysis Template
Errors in research planning or execution can lead to biased data and inaccurate conclusions. Here are some of the most frequent ones:
1. Bias in Research Questions
Description: Bias can appear in the form of leading questions, which steer users towards certain answers. For example, questions like “How much do you like this feature?” or “Do you think this interface improves your experience?” imply positive responses, potentially limiting honest feedback.
Solution: Ask questions neutrally. Instead of implying an answer, ask: “How would you describe your experience using this feature?” or “What difficulties did you encounter with the interface?”
2. Non-Representative Sample of Participants
Description: A sample that doesn’t reflect the diversity of the target audience may yield limited and unrepresentative results. For example, interviewing only new users might overlook the experiences and needs of returning users.
Solution: Create user segments for each key category (new, returning, active, inactive users, etc.) and include representatives from each group. This provides a more complete understanding of the user experience.
Engage diverse participants to uncover interaction nuances. For example, seek feedback from both experienced and new users to understand differences in interface perception.
3. Lack of Quantitative Data to Support Findings
Description: Relying solely on qualitative data (e.g., in-depth interviews or focus groups) may lead to subjective conclusions. While qualitative methods are valuable for exploring motivations and perceptions, they may not always reflect general patterns.
Solution: For broader applicability, combine qualitative insights with quantitative data, such as analytics, survey results, or A/B testing.
How to formulate hypothesis
A hypothesis serves as the foundation for any research, especially in user-centric B2C applications. It acts as a guiding light, steering the research team towards specific areas of focus. In the context of product development and UX/UI improvements, a well-crafted hypothesis helps to identify which elements of the user experience are most likely to affect key metrics, such as user retention, conversion rates, and overall satisfaction. Without a clear hypothesis, research can quickly become unfocused and result in insights that are difficult to act upon.
Hypothesis must be more than just a vague assumption. It should be clear, measurable, and testable, providing concrete criteria that can be observed and analyzed during the testing phase.
Here is a step-by-step approach to hypothesis formulation
1. Avoid using vague terms like "difficult" and "easy"
Terms like "difficult" and "easy" are often subjective, making it hard to verify the hypothesis. It’s better to specify how the difficulty or ease will manifest.
Example:
✕ "The user will easily find the 'Help' section."
✓ "The user will find the 'Help' section within 30 seconds."
2. Use specific parameters to measure success
It’s important to set measurable parameters such as time, number of steps, or other quantifiable indicators so the hypothesis can be tested.
Example:
✕ "The user will register quickly."
✓ "The user will complete the registration process in 2 minutes."
3. Don’t make the hypothesis too general
Avoid phrases like "users don’t understand," which cover too broad a scope. Instead, focus on a specific part of the interface or functionality.
Example:
✕ "Users don’t understand how the service works."
✓ "Users don’t understand that they need to enter a Tax ID to start."
4. Avoid unnecessary "taste" judgments
Hypotheses about users' preferences (e.g., "users will find the perfect product") are hard to test and assess objectively.
Example:
✕ "The user will find the right dress."
✓ "The user will be able to filter products by size and color."
5. Formulate only one assumption in the hypothesis
Complex hypotheses with multiple assumptions can be confusing and complicate result interpretation.
Example:
✕ "New users won’t find the Help section, and old users will already know where it is."
✓ "New users won’t find the Help section."
✓ "Old users will find the Help section."
6. Ensure the hypothesis is testable
Make sure the hypothesis is directly related to testable aspects of the interface, not abstract aspects of the user experience.
Example:
✕ "Users will be happy with the filters."
✓ "Users will successfully filter results by price."
7. Be aware of testing method limitations
Acknowledge that user testing may not fully replicate real-world conditions (e.g., if interactions with the product might happen in multitasking environments or with distractions).
Example:
✕ Test a navigation app only in controlled environments.
✓ Conduct testing in-store if your product is used in a shopping process.
8. Formulate the hypothesis to reflect real user experiences
Break down the hypothesis into different stages of the user journey and their interactions with the product.
Example:
✕ "The user will understand where the discounts section is."
✓ "The user will find the discounts section on the homepage."
9. Pay attention to interface details
Hypotheses related to large changes or entire functional blocks may be too complex. Focus on individual elements of the interface, like buttons or pop-ups.
Example:
✕ "The user will complete the purchase."
✓ "The user will click on the 'Complete Purchase' button after adding an item to the cart."
10. Use both qualitative and quantitative data to test the hypothesis
When formulating a hypothesis, specify what type of data will be used for testing (e.g., surveys, interviews, analytics, etc.).
Example:
✕ "The user will easily complete registration."
✓ "The user will complete registration 90% of the time."
11. Define success criteria for the hypothesis in advance
To ensure the hypothesis is testable, agree on what constitutes success beforehand. This might involve specific user actions or achieving certain metrics.
Example:
✕ "The user won’t face difficulties during checkout."
✓ "The user will complete the checkout process without errors in 5 steps."
12. Use insights from previous research to refine hypotheses
If you've conducted tests on similar topics before, use the data obtained to refine hypotheses and make more accurate assumptions.
Example:
✕ "Users will easily complete registration."
✓ "Users will click the 'Next' button 80% of the time after viewing the first screen."
13. Avoid negative formulations if possible
Negative hypotheses often cause confusion. It’s better to phrase the hypothesis positively to make it easier to determine success or failure.
Example:
✕ "The user won’t be able to find the right product."
✓ "The user will successfully filter products by brand."
14. Define the audience for hypothesis testing
Ensure that the hypothesis reflects the interests of the target audience. This will help interpret the test results correctly.
Example:
✕ "Users will easily choose a subscription plan."
✓ "Users aged 25 to 35 will easily choose a subscription plan."
15. Test the hypothesis on different devices and screen sizes
If your product is used on multiple devices, make sure the hypothesis considers all possible variations.
Example:
✕ "The user will find the product quickly."
✓ "The user will quickly find the product they’re looking for on a mobile device."
16. Avoid overly complex hypotheses
Hypotheses with too many variables are hard to test. It’s better to break them into separate hypotheses.
Example:
✕ "The user will find and add the product to the cart and then proceed to checkout."
✓ "The user will find the 'Add to Cart' button on the homepage."
✓ "The user will find the 'Proceed to Checkout' button in the cart."
17. Ensure the information is accessible to users
If the hypothesis is about providing information to users, make sure that information is easy to access and understand.
Example:
✕ "The user will find shipping information."
✓ "The user will find shipping information on the product page."
18. Use real-life examples if possible
Real scenarios help make hypotheses more understandable and testable.
Example:
✕ "The user will find information about discounts."
✓ "The user will find information about discounts on the product page."
19. Consider the context of product use
The product may be used in various contexts (e.g., while on the go, in an office, etc.). Make sure the hypothesis accounts for these contexts.
Example:
✕ "The user will find information about nearby service points."
✓ "The user will find information about nearby service points while traveling."
20. Focus on real user tasks
Ensure the hypothesis addresses real user needs and tasks, not abstract or hypothetical ones.
Example:
✕ "The user will find the product."
✓ "The user will find the necessary product category within 30 seconds."
Case study
In the beginning, we emphasized the necessity of conducting research for developing a B2C product. From a product-oriented perspective, the logical question is how research affects metrics. Improving metrics such as user retention, active user base (MAU), and satisfaction (NPS) directly depends on identifying and addressing interface flaws.
Let’s examine this impact with examples using a food delivery mobile app.
Case 1: In-depth Interviews for Conversion Improvement
- Problem: Low conversion rate after adding items to the cart.
- Solution: In-depth user interviews revealed that many users didn’t complete their orders due to a cumbersome payment process. Users reported frustration at having to re-enter credit card details at every step. Based on these findings, the product team redesigned the payment flow, added a card-saving feature, and implemented a simplified interface with clear steps.
- Result: After these changes, the conversion rate increased by 15%, leading to a higher overall order volume.
Case 2: A/B Testing for Retention Increase
- Problem: High drop-off rate during product selection.
- Solution: The team conducted A/B testing, comparing the current menu page with a new, visually enhanced version that allowed users to view larger images of items on plates and read customer reviews. Additionally, special offers and recommendations based on previous orders were added.
- Result: The new version increased user retention by 20%, as users spent more time in the app and returned more frequently for repeat orders.
Case 3: Observation for Improved User Journey
- Problem: Low opt-in rate for communication permissions.
- Solution: Observing user actions showed that many didn’t opt in for push notifications and other forms of communication because they didn’t understand their purpose. The research revealed dissatisfaction due to a lack of explanations. After adding brief descriptions about why certain information was requested and incorporating visual prompts, the user opt-in rate rose by 10%.
- Result: These changes increased registration completion by 10%, boosting retention and the share of active app users.
Final thoughts
In B2C products, UX/UI research often becomes a cornerstone of successful development. Understanding user behavior and needs lays the groundwork for designing a valuable, user-friendly, and in-demand product. By applying a combination of these methods, teams can gain deeper insights into user behavior and implement improvements that positively affect key product metrics such as retention, conversion, and NPS.