Real-time verification is like a receptionist with super fast on-the-spot response – but the problem is, different business scenarios require very different performance requirements for “this receptionist”. On the registration page, it should greet with a smile, on the payment page, it should look at the six ways, and on the tax declaration page, it should be serious. This requires us not only to engage in a general strategy, but to “customize the performance script” according to the scene.
Registration/Login Scenario: First Impression Can’t Be “Social Death”
As soon as the user clicks on the registration page, you roll your eyes and say “your name is not good”, and it is not surprising that the user immediately turns around and leaves. Registration and login are the “first impression” scenarios of the product, and the verification design must be friendly, rhythmic, and stepped.
Focus field with default state
The first step in the form experience starts with Focus. When the user opens the registration page, the input cursor does not automatically focus on the first input box, and the embarrassment of “you don’t take the initiative, I don’t start” will appear immediately. Especially for new users who are not so familiar with web page operations, they may scan back and forth on the page, or even think that the page is not loading well, and just exit.
A small interaction detail, such as automatically focusing the cursor on the “username” or “mobile number” input box, is actually like a considerate waiter handing over the menu and saying to the user: “Please fill in here, we are ready.” ”
Let’s talk about the default state. Many forms are full of red warnings as soon as they open: “password cannot be empty”, “please enter a valid email address”, and even “error messages” are lined up under the fields – the problem is, the user has not lost anything! It’s like as soon as a customer enters the door, the clerk rushes up and points to his shoes and says, “You can’t do this!” Instantly persuaded people to retreat.
The correct thing to do is to leave the default state “blank and quiet”. Only in the userActively start typingOr leave the field before gradually triggering the checksum prompt. This is “respect” and “guidance”, not “deterrence” and “strike”.
How to remind people not to hurt with weak passwords
The password field is one of the “emotional minefields” in the form. Many systems verify passwords while using percussive language to force users to change them:
- “Too short”
- “Too weak”
- “It must contain uppercase, lowercase, numbers, special symbols, your dad’s birthday, your mother’s maiden name…”
These prompts sound like the system is “picking and choosing” users. But in reality, most users are not resistant to setting strong passwords, they just areI don’t know what is “strong”。
After 10 years of interaction design, why did I transfer to product manager?
After the real job transfer, I found that many jobs were still beyond my imagination. The work of a product manager is indeed more complicated. Theoretically, the work of a product manager includes all aspects of the product, from market research, user research, data analysis…
View details >
Therefore, instead of using “rejection” language to attack users, it is better to use “encouraging” guidance:
- Provide a real-time password strength bar (e.g., a progressive hint that goes from red→ orange→ green, or “weak → → strong”);
- At the same time, a gentle suggestion was attached: “It’s safer to add special characters, for example!” 、#、&”；
- Encourage users, not punish them for not meeting the standards.
This method is much friendlier than the cold “password does not meet the specifications”, and users are more willing to cooperate with modifications.
Duplicate email/username: Don’t be the “King of Denial”
There is also a place where I often step on the pit, that isDuplicate email or usernameVerify. Imagine a user who has just entered a nickname that he has been thinking about for a long time, and the system coldly gives him a sentence:
“The nickname has been registered.”
Of course he knows that nicknames may be repeated, but your phrase “registered” is said indifferently and bluntly, like walking into a restaurant and being told “there is no place left”, without a single explanation or alternative suggestion, everyone will feel frustrated.
Better yet—give the step down, give the suggestion option. Like what:
- “This username is very popular, try jason_88 or jason892?”
- “The email address has been registered, are you trying to log in or retrieve your password?”
- “We found that you may already have an account, click here to log in quickly.”
It’s like you go to eat ramen and call someone else, but the clerk kindly says, “Wait a moment, we’ll arrange a window seat for you.” “The feeling of being cared for and understood is completely different. The verification mechanism is essentially “limiting”, but the expression determines the user’s psychological experience. The designer’s task is not only to tell users where they went wrong, but also to guide them on how to bypass problems and achieve their goals.
Business/Payment Scenario: You’re right, but I’m more afraid of filling in the wrong card number
In scenarios involving payment or important identity information, such as e-commerce, payment, insurance, and tax filing, form verification is no longer just about “experience optimization”, but a key line of defense against errors, losses, and complaints. You can think of these forms as “high-voltage lines”: entered correctly, the process is smooth; Wrong inputs can lead to payment failures, customer complaints, bill errors, and even financial or compliance risks. Therefore, the goal of this type of verification design is very clear: clear format, operation guidance, error tolerance, all of which are indispensable.
Strong format verification: don’t let users “play” freely
Some fields, freely enter = disaster scene, such as: credit card number, ID number, tax number, invoice letterhead, bank account information. This information is lengthy, has a fixed format, and must be 100% accurate. Once the user enters the wrong digit, the system cannot recognize it. Visually inspecting 20-digit numbers is also prone to errors. So what do we do? The core two words: “standardization” + “burden reduction”.
Dynamic mask + automatic formatting assistance
- Automatically add spaces(such as adding a space every four digits) to improve readability and accuracy;
- Limit illegal charactersFor example, the credit card number field prohibits the input of letters and special symbols;
- In-format validation: If the user has not finished losing, he can remind that “the number of bits seems to be insufficient” or “cannot start with 0”, instead of waiting for the user to fill in and then say “format error”, which is frustrating.
Case: Bank card input optimization of a large payment platform
They have introduced a three-in-one solution of “automatic identification of card issuers + real-time format verification + card number error correction prompt”, which not only reduces the input error rate, but also reduces a large number of customer service tickets caused by “card input errors”.The data results showed that the correct filling rate increased by 22% and the user complaint rate decreased by 40%.
Fault-tolerant design: Help users change, not let users come again
Traditional form validation designs are often only responsible for “pointing out errors” and rarely provide themSolution。 For example, if the user enters the last digit of the wrong ID number, the system will say coldly: “Verification failed.” ”
You think this is a “high standard”, but in the eyes of users, it is “impersonal”. A smarter approach is: “The last digit of the ID card should be a number, maybe you accidentally entered it wrong~”
Even more advanced practices are adoptedFuzzy matching + intelligent judgmentto “guess” the user’s possible true intentions, such as:
- the legal area code and date of birth are predicted by the first 17 digits;
- Assist in detecting the consistency between the pinyin name and the ID card;
- Judge the compatibility of OCR scanned content or paste format.
- These seemingly “helpful” details will eventually translate into higher submission rates and fewer error rates.
3. High-risk scenarios: If you make a mistake once, there may be a big problem
When dealing with high-risk operations such as tax declaration, real-name authentication, and provident fund withdrawal, the verification design must not only be accurate, but also understand “emotion management”. Because users are often in a high state of tension in these links, the slightest mistake will be associated with the worst result: “Will I be banned?” “Do I have to pay taxes if I fail to file a tax return?” “Will I be punished if I fill in the wrong information?” If the system only drops a sentence of “verification failed” at this time, the user’s anxiety will only double.
A truly effective verification prompt should form a “closed loop of feedback” – clearly point out the problem, appease users’ emotions, and give specific solutions. For example: “Verification failed, because the photo is not clear, please re-upload your ID document.” This will not affect your current progress. In this way, it not only explains the cause of the problem, but also gives a solution, and also dispels the concerns of users.
Tax declaration, real-name authentication, provident fund withdrawal…… These form scenarios have one common feature:Users are nervous and the system rules are strict.
At this time, the verification design should not only think about “how to stop the error”, but also have triple literacy:
Soothes emotions→ explains clearly→ provides a way out
The core of verification is not just “stopping you”, but “telling you how to walk safely”.
In addition, the multi-step verification process also needs to allow users to “understand the progress”. Imagine this scenario: a user uploads an ID photo, and the system pops up a line of words: “Verification failed.” “In your eyes, this may just be a normal mistake in identification; But in the eyes of users, an alarm has sounded in their minds: “Am I going to be banned?” “Do you have to pay thousands of yuan if you fail to file a tax return?” “Is it because I left the wrong information to be checked by the water meter?” If the last step fails, it should not simply prompt “failure”, but clearly inform that the first two steps have passed, and only need to remedy the current step. Status prompts such as “upload successful, face comparison passed, information recognition failed, please upload again” not only relieves user anxiety, but also establishes a sense of predictability in the system.
In high-pressure scenarios, any vague failure prompt will be supplemented by the automatic brain as “serious consequences”. The correct approach is to explain the “failure description + follow-up guidance” in its entirety:
Verification failed.
Reason: The photo is blurry and the system cannot recognize it.
Recommendation: Please upload a clear photo of your ID card. This error does not affect the current progress and the data is still saved.
This kind of feedback is a complete “psychological closed loop”: telling users where they went wrong, why they were wrong, how to change them, and whether it would have an impact. It can not only reduce misunderstandings, but also reduce anxiety and make users willing to continue operating.
Many processes are now done in steps, such as:
- Real-name authentication: Upload documents → face recognition → information verification
- Provident fund withdrawal: Bind account → information upload→ review and submission
- Tax declaration: identity verification→ information entry→ document generation→ submission platform
The problem is that once a step fails, many systems only show a sentence “Error” without any context. Users create a huge sense of uncertainty:
“Have I uploaded it successfully?”
“Do you have to start all over?”
“Where am I stuck?”
At this time, the “state awareness” mechanism is very critical. Users need to know themselvesWhere have you gone, have you succeeded in the past, and where is the problem now?。 The most basic approach is to displaystep-by-step status prompts,You can even be a little more intimate, adding a “saved, can continue” prompt to let users know that the previous operations will not be in vain.
Finally, don’t forget to provide users with a “way out”. Even if something goes wrong, let them know that the data will not be lost and that they can continue later, or try other ways to correct it. Instead of “you messed up,” they said, “we’ll fix it together.” The essence of the verification mechanism is never to stop people, but to help people.
