Practical guide to number screening service funnel: three-step operation process from number generation to gender screening

When acquiring customers overseas, the biggest fear is not not being reached, but reaching a bunch of invalid numbers—either not registered, or never active after registration, or simply not the target group. The traditional approach is to initiate a full test on all numbers at once. As a result, a large amount of balance is spent on useless numbers, and the cost remains high. Screening service funnel (staged funnel strategy) is precisely to solve this problem: divide the test into three steps: “activation → active → gender (including age)” to narrow down the precise group of people step by step, so that every penny is spent wisely.

This article takes KK-DATA screening platform as an example to break down the specific operations, field interpretations and precautions of each step in detail to help the overseas operation team establish an efficient and reusable screening workflow.

---

What is the filter service funnel?

The number screening service funnel is a staged and progressive number screening strategy. Its core logic is: First confirm whether the number is registered, then determine whether it is active, and finally target gender and age. In this way, you can filter out a large number of invalid numbers in the first stage and avoid wasting the cost of subsequent detection.

Why you need a staging funnel strategy

• Cost Control: The unit price for activation testing is the lowest, while activity testing and gender testing are more expensive. First do the activation, and then do the activity/gender screening of the registration number. The total cost is 30%-60% less than a full test.
• Data Quality: Activity and gender data are only meaningful if the number is registered. Ensure “existence” first, and then evaluate “quality” to avoid invalid analysis of empty numbers.
• Operation Flexibility: Different platforms and different stages can be used in combination, such as only active screening for Telegram and gender detection for WhatsApp.

The difference between the filter number service funnel and the one-time filter number

Dimensions	One-time screen number	Staging funnel screen number
Process	All detection types submitted at the same time	Submitted in stages, activated first and then active/gender
Cost	All numbers are calculated based on the highest unit price	Only paid numbers participate in subsequent testing, and the overall cost is lower
Data reliability	Active/gender fields may be mixed with invalid numbers	Each step of data is based on confirmed valid numbers
Executability	Suitable for small batches, full price and no pressure scenarios	Suitable for large batches (100,000+), cost-sensitive teams

---

Step one: How to efficiently generate numbers to be screened?

The starting point of the funnel is a pool of numbers, rather than importing existing numbers directly. KK-DATA’s Global Number Generation module (located in the left menu of the console) provides three generation methods:

• 240+ countries/regions randomly generated: Select the target country (such as the United States, India, Indonesia), and the system will automatically generate a random number of the real available number range in that country.
• Global number segment generation: Generate numbers in batches based on known number segments (such as the specific prefix of the US +1 number segment), suitable for targeting specific operators or regions.
• Customized CSV import: You already have some numbers (such as collected from exhibitions and forms). After importing, the platform will automatically deduplicate them and use them as the initial pool.

Usage scenarios for global number generation

Let’s say you want to target Indonesian WhatsApp users. In the build module you can:

1. Select country: Indonesia (+62)
2. Select quantity: 100,000 (generated for free)
3. Export the CSV directly as input to the first stage of the funnel.

This model of “first generating for free, then paying for filtering by item” can significantly reduce the initial audition cost.

Tips for importing custom number segments and batch generation

• If you already have a specific number segment (such as a centralized number segment for registered users on an e-commerce platform), directly import the CSV for subsequent detection.
• When you need to generate different number segments in batches, you can generate them multiple times and use the console’s “Data Deduplication Warehouse” to automatically merge them to avoid duplication.
• Important: The generation module itself does not consume the balance, and fees will be deducted only when the screening task is submitted. So the first step is to boldly generate as many numbers as possible.

---

Step 2: How to check the activation status of numbers in batches?

Opening detection is the core part of the funnel. KK-DATA supports activation/effective detection of Telegram, WhatsApp, Line, Zalo, iMessage, RCS, Viber, Facebook, Instagram, Binance, LinkedIn and other platforms. After submission, the system returns whether each number has been registered on the platform.

Multi-platform activation detection process (taking Telegram and WhatsApp as examples)

1. Click “New Detection Task” in the console and select the platform.
2. Upload the number file (CSV/TXT, one number per line), or import directly from the generation module.
3. Select the detection type: only “activated” or “activated + active” (it is recommended to choose only activated to reduce costs).
4. Confirm the estimated cost (visible before submitting the task) and click Submit.
5. Wait for the task to be completed (usually a few minutes to dozens of minutes, depending on the quantity), and the results will be automatically exported or notified through TG.

Field description of activation detection results: tg activation, ws activation, line valid, etc.

Taking Telegram as an example, the resulting CSV contains:

• number: original number
• tg_open: Whether to register (1=registered, 0=not registered)
• tg_active: Activity level (1=active, 0=inactive) If activity detection is selected
• tg_sex: Gender (male/female, only gender detection task)
• tg_age: Age (only gender detection task, see below)

Similar to other platforms, the field prefix is changed to ws_ (WhatsApp), line_ (Line), zalo_ (Zalo), etc. The specific fields are subject to console export.

View real-time prices and estimated fees on the console

The unit price for each platform and each detection type is different. For example: the unit price of Telegram-only activation detection is usually lower than that of WhatsApp activation detection, while the unit price of active detection is higher than activation-only detection. It is recommended to do a low-cost “activation-only” test first and export the registration number before submitting the activity/gender test, which is more cost-effective.

---

Step 3: How to filter activity and gender data?

After activating the test, you will get a batch of registered numbers. Next, for this batch of numbers, submit Activity Test or Gender Test (can be submitted together) to further screen out high-value users.

Actual scenarios and window settings for activity filtering

• Scenario: To promote a short-term event (such as a 7-day flash sale), you should filter Telegram users who have been active in the past 7 days; if it is a continuous interaction with the brand, you can select users who have been active in the past 30 days.
• Operation: Select “Activity” in the detection task and set the time window (such as 7 days, 30 days). The system only determines the activity of registered numbers and does not consume fees for unregistered numbers.
• Note: The unit price of active detection is higher than that of only enabled detection, but active detection of this batch of numbers will only be effective after the first step of filtering.

Interpretation and precautions of age field in gender detection

In addition to returning tg_sex (male/female), the gender test result will also return tg_age (age value, such as 28, 35, 42). This age field comes from platform behavioral data model estimation and is not an accurate age at the ID card level.

Practical Suggestion: You can use the age field as an auxiliary targeting parameter, such as filtering out tg_age BETWEEN 25 AND 35 users to get a group of people “about 30 years old” for the promotion of specific categories such as men’s skin care products and workplace education. But don’t rely on it for real-name authentication or risk control.

---

How to avoid wasting balance due to repeated detection?

KK-DATA’s Data Deduplication Warehouse is a global deduplication mechanism across tasks. When you obtain multiple number files from different channels, the system will automatically compare the numbers that have been detected to avoid further deductions.

• In the “Data Deduplication Warehouse” of the console, you can view all detected number records.
• When submitting a new task, the system will automatically skip numbers that already exist in the warehouse (even if they belong to different tasks).
• Billing Rules: The generation process is free. After the number screening task is completed, fees will be deducted based on the actual number of new numbers detected. Duplicate numbers will not be deducted again.

This function is crucial to the funnel strategy: first create the number to activate the test, and then submit the active test later, the system will not repeatedly deduct the “activation” fee, but will only deduct the new fee according to the new unit price of the “active test”.

---

Filter Service Funnel Best Practices and Checklist

1. Free generation, boldly expand the pool → Create 100,000+ numbers in the generation module and export them for free.
2. Step 1: Enable detection only → Submit the task and obtain the registration number list.
3. Export registration number → Automatically record duplicates in the warehouse.
4. Step 2: Activity Detection → Submit activity detection to the registration number and set an appropriate time window.
5. Step 3: Gender + Age Test → Submit gender test for active numbers and filter by age range.
6. Multi-format export → The final result is exported to CSV/TXT for import into CRM or marketing tools.
7. Turn on task notification → Bind Telegram on the console and receive notification immediately after the task is completed.
8. Balance Management → Check your balance at the console at any time. If it is insufficient, recharge it through USDT (TRC20), with a minimum of 50 USDT.

---

FAQ

Question: Why is it necessary to generate numbers in the first step of the number screening service funnel instead of directly importing existing numbers?

Answer: Generating numbers can quickly obtain a massive number pool (240+ countries randomly/number segment generated), and the generation is free. At the same time, you can customize the target number segment and gradually filter it in conjunction with the subsequent steps of the funnel to avoid consuming balances to detect invalid numbers from the beginning.

Question: What is the difference between enabled detection and active detection? Can we do it together?

Answer: Activation detection only determines whether the number is registered on the platform; activity detection determines whether the registered number has activity within a specified time window (such as 7 days/30 days). It is recommended to perform activation testing first to screen out the registration numbers, and then perform activity testing on the registration numbers to avoid wasting testing costs on invalid numbers. KK-DATA supports separate submission of tasks and merged submission (but merged submission will deduct both fees at one time).

Q: How accurate is the age field in gender detection? Can it be used to filter for specific age groups?

Answer: The age field is estimated based on the platform’s behavioral data model and can be used for group analysis (such as “people around 30 years old”). It is not recommended to be used as a precise ID card-level age. When screening, you can use the age range (such as 25-35) as a reference, but you need to understand its probabilistic nature.

Question: How many numbers can be processed at most in one screening task?

Answer: The maximum number for a single task is about 1 million. If the number pool is larger, it is recommended to submit it in batches, or use the data deduplication warehouse to merge across tasks to avoid repeated detection.

Question: How to check the real-time unit prices of screen sizes on different platforms?

Answer: Please log in to the console (https://app.kkdata.cc/）中创建任务时查看预估费用，详细单价以控制台实时显示为准。官网计费页（https://kkdata.cc/billing/）也有总体说明。

---

Through the installment funnel strategy, you can reduce the cost of screening services by more than half and obtain a more accurate target user list. Try logging into the console now:

👉Log in to the console to start screening numbers Two-way contact customer service: https://t.me/kkdata_robot Official document: https://docs.kkdata.cc/ Official website:https://kkdata.cc/