Article summary
Learn how to prepare a Viber phone-number TXT file, interpret active days, MID and offline-time fields, and turn the Excel output into useful list segments.
A Viber activity check is useful only when its output fields are interpreted in context. “Active days” is not a universal score for customer value, and an offline-time value does not prove that somebody wants to receive a message. The result is best treated as a list-hygiene signal that can be joined back to an authorized customer or lead dataset.
In the AIPUSH Viber activity task, the result contains the submitted phone number together with active days, MID and user offline time. This guide explains what those columns can support, what they cannot establish, and how to turn a raw export into a defensible segmentation worksheet.
The four output fields and their practical roles
| Excel field | How to use it | Do not interpret it as |
|---|---|---|
| Phone number | Join key for matching the result to the submitted list | A new identity or proof of ownership |
| Active days | A relative recency indicator for sorting or cohort comparison | A guarantee that the user will respond |
| MID | A Viber-related identifier returned for record handling | Permission to contact the person |
| User offline time | An additional timing signal for quality review and segmentation | A precise prediction of future online behavior |
Keep the original column names in the master file. If analysts create business-friendly labels such as “recent,” “warm” or “dormant,” store those in separate derived columns so the source result remains auditable.
Prepare the TXT input without losing the source relationship
AIPUSH accepts a TXT file for the Viber task. Put one phone number on each line and normalize the list before upload. Remove decorative punctuation, make the country code explicit, preserve leading digits correctly and deduplicate exact repeats. Do not place names, emails or CRM notes in the TXT file.
Before exporting from the CRM, generate a stable internal row ID. Keep that ID in the local source table rather than sending it in the TXT. After the result arrives, the phone number becomes the bridge back to the original row. This approach reduces the personal data submitted while protecting the analyst from losing the acquisition source or consent record.
Run a small sample before committing the full list
Use a pilot containing records from several known cohorts: recent customers, older leads, multiple countries and a few deliberately duplicated formats. The purpose is not to hunt for a perfect “active days” cutoff. It is to verify that normalization, task selection, output columns and join logic behave as expected.
Record the submitted row count, unique normalized count and returned row count. If they differ, investigate formatting, blank lines, duplicates and join-key transformation before making marketing decisions. A small controlled sample exposes these issues far more cheaply than a full-file run.
A worked example: turn fields into review cohorts
| Illustrative row | Active days | Offline-time observation | Reasonable next step |
|---|---|---|---|
| A | Low | Recent relative to the sample | Place in a priority review cohort, then confirm consent and campaign relevance |
| B | Medium | Older than the recent-customer median | Keep separate and test with conservative frequency |
| C | High | Longer offline interval | Do not discard automatically; compare source, geography and last customer event |
| D | Missing or anomalous | Unavailable | Send to a data-quality queue rather than forcing a score |
These are hypothetical interpretations, not platform guarantees. The numerical meaning should be calibrated from the actual export and your own business outcomes. A team selling a weekly service may define recency differently from a company with a twelve-month replacement cycle.
Design thresholds from your own cohorts
Start with distributions rather than arbitrary rules. Calculate the median and percentiles of active days for each acquisition source or customer group. Then compare those bands with known outcomes such as reply, purchase, service ticket or valid opt-out. A threshold becomes useful when it distinguishes behavior in your data—not when it merely produces a round number.
A simple analysis plan is:
- Separate customers, opted-in prospects and legacy records.
- Plot active-day ranges within each group.
- Check whether geography or source produces different baselines.
- Create a holdout group before changing campaign treatment.
- Review both positive outcomes and complaints, blocks or opt-outs.
Reconcile the Excel result with the source file
After download, do not immediately overwrite the CRM. Import the Excel workbook into a staging table and normalize the phone-number key using the same rule used before upload. Join it to the preserved source extract, then classify unmatched, duplicate and conflicting rows.
Three counts should be visible in the reconciliation sheet: input unique numbers, output unique numbers and successfully joined source records. Add a reason column for exceptions. This makes the activity check repeatable and allows another analyst to understand why a record entered a segment.
When activity checking is the wrong task
If the only question is whether a number is registered on Viber, use the Viber registration-status task instead of inferring registration from activity fields. If the project needs demographic and profile-related fields, the Viber gender-and-age task returns a different set of columns, including nickname, gender, age, avatar-related values, skin tone, MID and timing fields.
Task choice matters because the selected service determines the Excel schema. Ordering the broadest dataset “just in case” increases handling responsibility and may not answer the original question. Define the decision first, then request only the fields needed for it.
Privacy and permission controls belong outside the score
Activity data does not establish consent. Keep the source, collection date, permission state and opt-out status in the company’s own system. Exclude records that lack a lawful or policy-compliant basis for the intended use, regardless of how recent the activity signal appears.
Restrict access to both the upload and the result, set a retention period, and avoid distributing full workbooks through informal chat channels. Derived segments should include only the columns required by the operating team.
A repeatable Viber activity-check workflow
- State the business decision the activity result will inform.
- Export authorized phone numbers and retain source IDs locally.
- Normalize and deduplicate the numbers into a one-number-per-line TXT file.
- Choose the Viber activity task rather than a different field service.
- Download the Excel output and preserve an untouched master copy.
- Reconcile phone numbers to the source dataset in a staging table.
- Build cohort-specific bands and validate them against real outcomes.
- Document exceptions, permission filters and retention dates.
The value of the check comes from disciplined interpretation. Active days, MID and offline time can sharpen a data-quality decision, but they should remain observable fields—not be turned into unsupported claims about intent or identity.
