Customer acquisition data export specifications: complete guide to format, field mapping and downstream tool docking

After overseas teams export customer acquisition data from the platform, they often encounter problems such as garbled fields, incompatible formats, and duplicate numbers. These problems may seem minor, but they directly lead to the inability to import the downstream CRM, failure to send EDM, illegal format prompts on social media tools, and even account bans due to repeated contact. This article will explain in detail the customer acquisition data export standards, including CSV/TXT field mapping rules, batch deduplication mechanism, and suggestions for docking with CRM/EDM/social media tools, to help you improve data utilization efficiency and reduce the risk of downstream errors.

What are customer acquisition data export specifications? Why does it directly affect customer acquisition efficiency?

Export specifications refer to the unified agreement on file format, field structure, encoding, deduplication logic, naming rules and other elements when exporting data from the Sieve platform. Lack of standardized export may cause the following problems:

• Duplicate detection wastes balance: The same number appears in multiple tasks in different formats (such as with country code/without country code), and the platform cannot automatically identify it as a duplicate, resulting in multiple deductions.
• Downstream import error: Commas or line breaks are mixed in CSV fields, causing column misalignment during Excel or CRM parsing; inconsistent encoding causes Chinese garbled characters.
• Population targeting failure: Fields such as gender and activity are not correctly mapped, causing the filtering conditions to be misnamed and sent to the wrong user group.

Therefore, specification export is the key hub connecting “screen number” and “reach”.

Common export confusion scenarios: field misalignment, inconsistent encoding, and null value processing

• Comma Conflict: The CSV is separated by commas, but the field value also contains commas (such as the company name “KK Data, Ltd.”), resulting in column misalignment.
• Excel automatic conversion: The number “8613800138000” is recognized as scientific notation or a number, and the last few digits are missing; the date “01/02” is converted to Excel’s internal date format.
• Encoding problem: The system defaults to UTF-8, but when users open Excel with GBK encoding, Chinese characters are displayed as garbled characters.
• Null value processing: When fields such as gender and activity are empty, the export file may remain blank or filled with “N/A”. If the downstream tool does not recognize it, an error will be reported.

Three core dimensions of standardized export: format, field, structure

Dimensions	Key elements	Role
Format	CSV, TXT	Determine whether the file can be read directly by downstream tools
Fields	Number, platform, status, activity, gender, ID	Ensure that each record carries available business information
Structure	Fixed column headers, no extra spaces at the end of lines, field reference rules	Ensure parsing consistency and reduce manual cleaning

It is recommended to confirm these dimensions before each export to avoid duplication of work.

CSV and TXT export formats: What scenarios are each suitable for? How to choose?

Both formats have their advantages, and the choice depends on the nature of the downstream tools.

Comparison items	CSV (Comma-Separated Values)	TXT (plain text)
Structure	Multiple columns, each column has a label	Single column or multiple columns, no fixed label
Applicable scenarios	CRM import, EDM mass sending, advertising platform upload audience	Telegram mass sending, blacklist filtering, script batch processing
Field reserved	Multiple fields such as number, status, gender, etc. can be exported at the same time	Usually only the number is included, or one field per line
Encoding issue	UTF-8 encoding is the most common and needs to be specified when opening Excel	Column-less structure, garbled characters have less impact
Parse complexity	Need to deal with column mapping and escape characters	Simple, rows are records

Selection Suggestions:

• Downstream is a CRM (such as Salesforce, HubSpot) or marketing automation platform → CSV
• Downstream is a script or command line tool → TXT
• Just plain number list → TXT
• Need to export number + gender + activity level at the same time → CSV

Import field mapping rules for customer acquisition data - what does each field represent during docking?

Taking the typical exported fields of KK-DATA as an example, the meaning of each field and the common mapping methods in downstream tools are explained.

Core required fields: number, platform, status

Field name	Meaning	Mapping example
phone	Complete number with country code (such as 8613800138000)	Directly as the phone field of CRM
platform	The platform to which the number belongs (such as Telegram, WhatsApp, Line)	Tools for diverting traffic to different channels
status	Test result: valid or invalid	Filter condition: only import records of valid

Value: These three fields determine “whether the number is available and through which channel.” When exporting, ensure that the phone field formats are uniform (such as all E.164 formats) to avoid downstream sending failures due to missing prefixes.

Advanced filtering fields: activity, gender, age, ID (tgid/wsid/uid)

Field name	Meaning	Function
activeness	Activity, such as 3天内活跃, 7天内活跃, 30天内活跃	Used to target “recently active” users and reach them first
gender	Gender identification results, such as 男, 女, 未知	Used for gender-targeted marketing
age	The age field in the gender detection result can be interpreted as about 30 years old, etc.	Used for crowd profiling and screening for specific age groups
tgid	Telegram user unique identifier	Skip number verification and map directly to registered users
wsid	WhatsApp user ID	Same as above, reduce invalid sending rate
uid	Line/Zalo user ID	Used for accurate push or re-detection

Mapping Suggestion: If the downstream tool supports conditional filtering, you can convert activeness into a logical condition (such as “only import numbers whose activity is marked within 3 days”); if not, you can use a script to generate the “active days” numeric field and then map it. It is recommended to check ID fields such as tgid and wsid when exporting. Importing them into the mass sending tool can skip number verification and greatly reduce the invalid sending rate.

How to ensure data uniqueness when exporting in batches? How does the deduplication mechanism work?

Duplicate numbers not only waste the detection balance, but also easily lead to repeated downstream contacts, causing user complaints and even account bans.

How the duplicate warehouse works: In KK-DATA, the deduplication warehouse is a cross-task number pool. When you submit a new task, the platform will automatically compare the existing number records in the warehouse:

• If the number has been detected and the status has not changed → No further detection will be performed and no fee will be deducted.
• If the number is new → Check normally and charge.
• If the number was previously detected as invalid, but the status changes to valid when submitted this time → re-detect and charge the fee.

When exporting, you can choose to “keep the latest detection results” or “retain only one after deduplication”. It is recommended to select to keep only one record after deduplication, and give priority to retaining records with status valid.

3 common errors and solutions when exporting data to connect with downstream tools

1. Number prefix does not match

• Performance: After importing the CRM, it displays “The number format is invalid”, or the Telegram group sending tool prompts “The number is not registered.”
• Cause: The exported number is in E.164 format (with +), but the downstream requirement does not have + or does not have an international prefix.
• Workaround:
  1. Confirm the number format requirements of the downstream tool before exporting.
  2. If it has been exported, you can use script or text replacement function to remove + (such as sed 's/^+//'), or add the missing 00 prefix.

2. The date field was tampered with by Excel

• Performance: “2024-03-15” in the activity field is automatically converted to “45121” or “2024/3/15”.
• Cause: Excel automatically recognizes text as date and converts the format.
• Workaround:
  1. Set the target column format to “Text” before importing into Excel.
  2. If it has been tampered with, use the “Split Column” function in Excel to change the columns to text format and then import again.
  3. It is recommended to use an editor that supports UTF-8 (such as VS Code, Notepad++) to view CSV instead of Excel.

3. The CSV field contains carriage returns and line feeds, causing parsing to fail.

• Performance: One row of records is split into multiple rows, causing the CRM to report an error “number of columns does not match”.
• Cause: The field value contains \r\n internally (for example, the company name contains multiple lines of text), but the CSV is not wrapped in double quotes.
• Workaround:
  1. Make sure to use the CSV format with double quotes (standard RFC 4180) when exporting.
  2. If it has been exported, use a text editor to find the unwrapped multi-line fields and add double quotes manually.
  3. Simple solution: Use a script to replace newlines in the field with spaces.

How to verify the quality of exported customer acquisition data? A simple checklist

Before importing data into downstream tools, take 5 minutes to perform the following verifications:

• Is the total number of records consistent with expectations (for example, the task result shows “10,000 valid numbers”, the number of rows in the export file should be 10,001 including headers).
• Whether the number of fields is complete (open the CSV file with a text editor and confirm that there is a value after each delimiter).
• Whether the number format is correct (all in E.164 format, no random Chinese characters or letters).
• Is the proportion of null values ​​acceptable (if 80% of the activeness fields are empty, it may need to be retested).
• Whether the duplication rate is less than 5% (can be quickly checked through the “Delete Duplicates” function in Excel).

From export to implementation: a best practice process for customer acquisition data docking

1. Generate → Use the global number generation function (240+ countries/regions) or CSV import of custom number segments to generate a target number pool. This step is free.
2. Screening → Submit the screening task and select the platform and type to be tested (activated, active, gender, etc.). Estimate the cost before submitting the task. If the balance is insufficient, you need to recharge first (USDT TRC20, the minimum is about 50 USDT).
3. Export → Select CSV or TXT format in the console and check the required fields (number, status, activity, gender, tgid/wsid, etc.). It is recommended to use file names with timestamps, such as export_20250220_valid.csv.
4. Deduplication → Use the “Deduplication Warehouse” to clean up duplicate numbers before exporting to reduce the risk of repeated downstream contacts.
5. Validation → Execute the above checklist to check field mapping and null value ratio. If necessary, first extract 10% of the data for test import.
6. Import → Adjust the CSV encoding according to the requirements of downstream tools (UTF-8 without BOM is the most common), and map the field names to the corresponding downstream fields one by one. If the downstream does not support “active days” text filtering, you can first generate a numeric field and then map it.

Additional Tips:

• Keep an export record as an original backup to facilitate subsequent calibration.
• Clean the “duplication warehouse” regularly to keep the number pool healthy and avoid expired numbers affecting detection accuracy.
• For Customer Acquisition Data that is large in size (such as more than 500,000 items), it is recommended to export it in batches to avoid Excel loading lag. The maximum number of single tasks is about 1 million, and if it exceeds the limit, it can be submitted in batches.

FAQ

**Q: What should I do if the numbers are displayed in scientific notation when exporting CSV? ** A: Set the column format to “Text” before opening in Excel, or import using TXT format. If it has been exported, it can be pasted in a text editor and transposed. Some tools support “preceding the number with a single quote” as is.

**Q: How do the exported activity fields (such as “active within 3 days” and “active within 7 days”) correspond to downstream filtering? ** A: It is recommended to convert this field into a logical condition in downstream tools, such as “Only import numbers whose activity is marked within 3 days.” If the downstream does not support text filtering, you can first use a script to generate the “active days” numeric field and then map it.

**Q: How many pieces of data can be exported in one task? Is there a size limit for exported files? ** Answer: A single task supports about 1 million numbers. The size of the exported file is affected by the format. It is recommended to export in batches when the number is greater than 500,000 to avoid lags when opening Excel. See the console export tips for details.

**Q: What are the “tgid” and “wsid” in the export file? How to use them to reach more accurately? ** Answer: tgid is the Telegram user’s unique identifier, and wsid is the WhatsApp user identifier. Importing these IDs into the mass sending tool can skip number verification and directly map to registered users, reducing the invalid sending rate. It is recommended to check the corresponding fields when exporting.

**Q: Can the exported customer acquisition data be used multiple times? Will the platform deduct fees repeatedly? ** Answer: Numbers that have been detected and exported will not be deducted repeatedly. If the exported data is submitted for testing again (for example, to supplement gender testing), KK-DATA’s deduplication warehouse will automatically identify duplicate numbers and only deduct fees for newly added or status-changed records. See the console billing page for specific rules.

After reading this standard guide, you may want to try the standard customer acquisition data export process immediately. Now log in to the console to create a screening task, or contact two-way customer service to get export templates and docking suggestions.

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