How to choose the TG filtering order? Detailed explanation of the three-layer funnel of activation, activity and gender

In overseas marketing, community operations or private message promotion, TG filtering (Telegram filtering) is the first step in batch screening of target users. But many teams will encounter the same problem: Should they first check whether the number has opened Telegram, or check the activity first, or prioritize gender screening? Choosing the wrong order will not only waste your budget, but you may also fail to obtain high-quality data.

This article focuses on the three core filters in TG Screen Number - activation, activity, and gender to analyze the three-layer funnel strategy. You will learn to select the optimal filtering order based on customer acquisition goals, budget and number pool quality, and use tools (such as KK-DATA) to execute it efficiently. Every step in the article can be implemented, and a practical checklist and frequently asked questions are attached at the end of the article.

Why is the order of TG filtering important?

Let’s take an extreme example first: you take 1 million numbers and run activity detection directly, but 40% of them are not registered with Telegram at all. As a result, the activity detection fee was also deducted for these 400,000 invalid numbers, which was actually a waste of money. On the contrary, if you first use activation detection to eliminate invalid numbers, and then perform active detection on the remaining 600,000, the cost can be saved a lot.

The core idea of ​​Three-layer funnel is: put the lowest-cost filter at the front, gradually reduce the number pool, and allow high unit price detection to only act on valid numbers. At the same time, the order of the funnel also affects the quality of the final results - for example, if you only want to find highly active female users, doing gender first and then activity will be more efficient than the other way around. Therefore, tg filtering is not only a technical action, but also a multiple-choice question of cost and accuracy.

Understand the three basic filtering fields: activation, active, gender

On screening platforms such as KK-DATA, TG screening usually provides the following three detection types (fields are subject to console export):

Activation detection: Whether the number has been registered with Telegram

“tg activated” means that the mobile phone number has registered a Telegram account. This is the first threshold for Telegram filtering. Only numbers that pass the activation test are meaningful for subsequent operations. The unit price of this test is lower than that of activity and gender, making it suitable as a first-tier funnel.

Activity detection: Does the user have any recent usage behavior?

The “tg active” detection will return the number’s active status within the specified time window. The platform usually allows you to select an active window, such as the last 7 days, the last 30 days, the last 60 days, etc. The shorter the window, the more active the filtered numbers are, but the smaller the number. If your marketing goal is short-term reach (such as event invitations), it is recommended to use the last 7 days; if it is long-term group cultivation, the last 30 days is more appropriate.

Gender/Age Identification: Data Basis for Crowd Targeting

Telegram’s gender data is not officially provided, but is inferred through algorithmic models. The results include gender (male/female) and age fields, which can be used to filter for males, females, or users of a specific age group (for example, people around 30 years old). Note: This is not ID card level accuracy, but it is enough for crowd targeting. If you need to promote “men around 30 years old”, you can use the gender + age field combination to filter.

Three-layer funnel strategy: filtering order selection under different goals

The following three funnel sequences cover common Telegram filtering scenarios. You can choose according to actual needs, or adjust the middle layer.

Funnel 1: Open → Active → Gender (the most economical and the most accurate)

Applicable scenarios: Limited budget, high accuracy requirements, and ultimately reaching highly active and gender-oriented numbers.

Execution Steps:

1. First run the “tg activation” test to screen out the registered numbers.
2. Submit the “tg active” test for the activated number and select the appropriate window.
3. Submit the “tg gender” test for the active and passed numbers to get the final targeting list.

Advantages: Each layer only processes the numbers passed in the previous step, and the invalid overhead is minimal. Disadvantages: It needs to be divided into multiple tasks, and the total time is slightly longer.

Funnel 2: Open → Gender → Active (people orientation is prioritized)

Applicable scenarios: You need to first ensure that the number is the target gender/age, and then find active users from it. For example, skin care products promotion only wants to find women, and game promotion only wants men.

Execution Steps:

1. Enable detection first and remove unregistered numbers.
2. Conduct gender/age detection on valid numbers to screen out the target group.
3. Finally, submit the target group to activity detection and limit the activity window.

Advantages: Gender-specific intervention occurs earlier to avoid wasting the cost of active testing for non-target groups. Disadvantages: The unit price of gender detection is usually higher than activation, but lower than active, and the overall cost is still within control.

Funnel Three: Active → Activated → Gender (high activity priority, suitable for old number pools)

Applicable scenario: You already have a batch of historical Telegram numbers (for example, exported from old groups), which are most likely registered and the gender is known, and you just want to clear the activity again. Or if the number pool is of high quality (activation rate > 80%), you can directly activate it first.

Execution Steps:

1. Directly submit activity detection and screen out high activity numbers.
2. Check the activation of the active number (although it is usually registered, it can be confirmed).
3. Finally do a gender test (optional).

Advantages: Save an intermediate duplicate submission and get the active list quickly. Disadvantages: If the activation rate of the original number pool is low, many unregistered numbers will also be deducted the activity detection fee, which is costly. Therefore, this sequence is only recommended for high-quality veteran account pools.

---

Comparison table of three funnels

Funnel sequence	Recommended scenarios	Cost control	Accuracy
Open → Active → Gender	Limited budget, general customer acquisition	★★★★★	★★★★★
Open → Gender → Active	Gender orientation priority (female/male)	★★★★☆	★★★★☆
Active→Activate→Gender	Second cleaning of old number pool	★★★☆☆	★★★☆☆

How to set the filtering order according to the amount of data?

The number of numbers also affects strategy selection.

• Small batch (less than 5,000 items): It is recommended to directly use “activation + activity + gender” for one-time full detection. The single cost is estimated to be low, and running multiple times in layers will be a waste of time. KK-DATA supports up to about 1 million items in a single task. The estimated cost will be displayed before the task is submitted, making it convenient for small-amount testing.
• Large batches (over 50,000): The layered funnel is highly recommended. Taking 500,000 numbers as an example, if activation is done first, half of the invalid numbers can be eliminated (assuming a 50% registration rate), then subsequent activity/gender detection will only deduct fees for 250,000 numbers, saving about half of the cost.

KK-DATA also has a built-in data deduplication warehouse, which automatically removes duplicate numbers across tasks and avoids repeated deductions for the same number. Before submitting a task, the system will automatically compare historical tasks, and no further charges will be deducted for numbers that have been detected.

Practical checklist: Confirm these 5 items before submitting the TG screening task

1. Confirm number source: Is it randomly generated, imported into CSV, or exported from other platforms? Make sure the number is in the correct format (with country code, such as +86…).
2. Choose the correct detection type: Clarify the fields you need - activated, active (specified window), gender (including age field).
3. Estimated fee: Check the estimated fee deduction on the KK-DATA console before submission to confirm that the balance is sufficient.
4. Enable data deduplication: If you have run similar tasks before, the system will automatically deduplicate the data to avoid repeated deductions.
5. Prepare export format: Select CSV or TXT and confirm the required fields (such as phone number, tgid, active status, gender, etc.).

In addition, it is recommended to clean the “checked number pool” regularly or use KK-DATA’s deduplication warehouse for long-term maintenance. For example: last week you detected 100,000 numbers that were activated but not active. You can directly export these numbers this week and submit them for activity detection. The deduplication warehouse will automatically skip the detected numbers.

Differences in screen number order between different platforms (Telegram vs WhatsApp vs Line)

The three-tier funnel strategy is not unique to Telegram, but each platform’s active definition and field accuracy are different and require fine-tuning.

Features of Telegram filtering

• High Activity: Telegram users are accustomed to being online on multiple devices, and activity detection coverage is wide. Supports frequent batch testing, suitable for the “Activate→Active→Gender” funnel.
• Relatively rich gender data: The model has a high gender recognition accuracy for users aged 20-40, but the age field is only for trend reference.

WhatsApp filter order suggestions

WhatsApp’s activity determination relies on the recent “last online” time. Some platforms even require a number to be online within 48 hours to be considered active. It is recommended to activate first → be active first**, because the activation rate of WhatsApp is usually lower than that of Telegram, and it saves money to eliminate invalid accounts first. Gender detection is available on WhatsApp, but accuracy varies by region.

If your target is Line (Japan, Taiwan, and Thailand markets), Line activation and validity testing are common entrances. Gender recognition performs well in Southeast Asia. It is recommended to “open → gender → active” in the order. The same applies to Zalo (Vietnam market), you can refer to Line logic.

FAQ

**Q: Why is it more cost-effective to turn on filtering first than activate filtering first? ** A: The unit price for activated testing is usually lower than active testing. Activating first can quickly eliminate unregistered numbers, so that activity detection can only be applied to valid numbers and avoid invalid numbers from being deducted. The specific price difference is subject to the real-time price of the console.

**Q: After selecting the wrong filtering order, can the remaining numbers be resubmitted? ** Answer: Yes. Just resubmit the activity detection task with the last export result (such as “numbers with valid subscription but not detected as active”). KK-DATA’s data deduplication warehouse will automatically remove duplicate numbers to be inspected, and no duplicate charges will be deducted.

**Q: What is the difference between selecting 7 days and 30 days for Tg active window? ** Answer: The shorter the window (such as 7 days), the more active the filtered numbers will be, but the number will be significantly reduced; the longer the window (such as 30 days), the more numbers can be obtained but the activity may decrease. It is recommended to choose according to the marketing cycle - 7 days for short-term activities and 30 days for long-term group cultivation.

**Q: When detecting multiple platforms (such as Telegram+WhatsApp) at the same time, how to arrange the filtering order? ** Answer: It is recommended to submit activation tests for each platform separately in the same number pool, screen out the valid numbers for each platform, and then submit activity/gender tests separately according to the target platform (such as Tg or WhatsApp). The order of multi-platform joint screening can refer to the single-platform funnel, but it should be noted that the activity determination rules of each platform are different.

**Q: Does small batch testing (such as 500 items) also need to follow the three-tier funnel sequence? ** Answer: For small batch testing, it is recommended to directly use the “activation + activity + gender” one-time test. The cost of a single task is estimated to be low, and there is no need to run multiple times in layers to save balance. The cost-effective advantage of the layered funnel is only obvious when the batch size is large (such as more than 50,000).

---

Summary and action suggestions

There is no absolute optimal order for TG filtering. Everything depends on your customer acquisition goals, budget and number pool quality. Remember the core principle of the three-layer funnel: Use low unit price filters to remove invalid data first, and let high unit price detection only execute on valid numbers. If it is a new number pool, the first choice is “Activate → Active → Gender”; if gender orientation is a priority, bring gender to the second level; if the existing number pool is of high quality, you can run active first.

You can now log in to the console to generate a number pool for free or try out TG screening numbers. There is no need to subscribe to a package. You are billed on a per-item basis. You pay for what you use. The minimum recharge of USDT (TRC20) is about 50 USDT. Start your first Telegram filtering task.

👉Log in to the console to start screening numbers Two-way contact customer service https://t.me/kkdata_robot For detailed usage documentation, please refer to https://docs.kkdata.cc/