Synthetic dataset
Telecom Subscriber Journeys & Changing Usage
Follow synthetic subscriber states, network use, spending, mobility, SIM/device changes and tariffs in one monthly research panel.
Nairobi Pinned published reference · research CSV bytes verified · Version V1
- Sampled subscriptions
- 500
- Telecom events
- 14,556
- Event period
- 2016–2026
- Released format
- CSV (gzip)
This Nairobi event-level reference is separate from the catalogue’s 100-subscriber monthly-panel preview and from custom orders. Its counts, event fields and checks do not certify the monthly panel, Lagos or an order-specific extract.
Overview
Nairobi’s selected research panel contains 100 synthetic subscribers, 5,240 monthly observations and 47 original fields. SubscriberId and CitizenId link subscription, usage, spending, mobility and device histories.
The record concepts include SubscriberId and CitizenId, ObservationAsOfDate and monthly boundaries, HistoricalSubscriberStatus and HistoricalStatusReason, HistoricalTariffPlanId and TariffChangesInMonth, DataUsedMB, VoiceMinutes and SMSCount, AirtimeTopUpLocal and DataBundlePurchaseLocal, MobilityTripCount and MobilityDistanceKm and SIM episodes and device replacements. Compare historical states with usage and spending. Review tariffs, mobility and SIM or device changes. Retain supplied splits, partial months and separate snapshot context.
This dataset is prepared to order. Choose the city, date range and required scope; the accepted order defines the population, final fields, formats and delivery.
- Onboarded city options
- Nairobi (Kenya)
- Custom city requests
- African Major Cities · Additional cities require availability, scope and quote confirmation.
- Data type
- Synthetic
- Connected domains
- Telecommunications, Demographics
- Catalogue structure
- Longitudinal monthly panel
Key features
Record concepts for your order-specific extract.
Telecommunications
Historical subscriber states and recorded usage
Demographics
Age, identity and resident context
Monthly observations
Observation cutoffs, partial months and separate snapshot context in the selected Nairobi panel
Who it is for
- Telecom researchers
- Data scientists
- AI evaluation teams
- Product analysts
Data & Schema
- Proposed study unit
- Subscriber-month
Explore changing subscriber states and monthly use within the selected Nairobi research panel.
Record concepts for your extract
- SubscriberId and CitizenId
- ObservationAsOfDate and monthly boundaries
- HistoricalSubscriberStatus and HistoricalStatusReason
- HistoricalTariffPlanId and TariffChangesInMonth
- DataUsedMB, VoiceMinutes and SMSCount
- AirtimeTopUpLocal and DataBundlePurchaseLocal
- MobilityTripCount and MobilityDistanceKm
- SIM episodes and device replacements
The agreed specification defines the final columns, keys, relationships and observation period for your order.
Pinned published reference · research CSV bytes verified
Nairobi Synthetic Longitudinal Telecom Journeys — V1
This Nairobi event-level reference is separate from the catalogue’s 100-subscriber monthly-panel preview and from custom orders. Its counts, event fields and checks do not certify the monthly panel, Lagos or an order-specific extract.
- Reference release
- NRB-TEL-HF-V1
- Release date
- 2026-10-04
- City
- Nairobi
- Sampled subscriptions (499 synthetic people)
- 500
- Telecom event rows
- 14,556
- Telecom event columns
- 25
- Observation cutoff
- 2026-09-01
- Original export format
- CSV (public repository); CSV.gz (supplied package)
| Configuration | Train | Validation | Test | Total |
|---|---|---|---|---|
| telecom events | 11,216 | 1,560 | 1,780 | 14,556 |
| subscriber profiles | 390 | 50 | 60 | 500 |
Citizen-grouped hash splits, not chronological cohorts; all subscriptions of a synthetic person remain in one split. Event-row counts differ from profile-row counts.
Published paths use CSV; the supplied local package uses gzip wrappers with matching decompressed research-data bytes.
telecom events reference schema
SyntheticSubscriberId (many events to one sampled subscription) Published paths use CSV; the supplied local package uses gzip wrappers with matching decompressed research-data bytes.
These are verified original CSV columns. Logical SQL types are stated only where explicitly declared in SOURCE_QUERY.sql; other fields are text serialization with no supplied complete logical type. Snapshot context belongs to 2026-09-01.
Selected fields · 12 of 25 columns
Full field definitions, types and descriptions 25 columns
- Observed reference row key
- SyntheticSubscriberId + EventOrder
Study, context and observation fields
Technical identifiers and evidence references
subscriber profiles reference schema
Snapshot* values apply to 2026-09-01, not to every historical event date Published paths use CSV; the supplied local package uses gzip wrappers with matching decompressed research-data bytes.
These are verified original CSV columns. Logical SQL types are stated only where explicitly declared in SOURCE_QUERY.sql; other fields are text serialization with no supplied complete logical type. Snapshot context belongs to 2026-09-01.
Selected fields · 12 of 26 columns
Full field definitions, types and descriptions 26 columns
- Observed reference row key
- SyntheticSubscriberId
Study, context and observation fields
Technical identifiers and evidence references
Reference data files
data/telecom_events/{split}.csvdata/subscriber_profiles/{split}.csv
Selected synthetic research panel
This is the selected synthetic sample, not the full city population or a certified paid extract.
Proposed preview columns
Selected fields · 12 of 47 columns
All proposed preview columns 47 columns
These names describe the illustrative preview. Final fields and definitions are agreed for your order.
Study, context and observation fields
Technical identifiers and evidence references
Use Cases
Published reference applications
- Subscription lifecycle and event-order analysis
- Linked telecom, device, mobility and economy journey exploration
- Citizen-grouped benchmark evaluation using the supplied ResearchSplit
- Event-based analysis of monthly usage periods with documented text parsing
Custom study design
Compare subscribers at consistent observation cutoffs. Retain supplied splits, separate snapshots from history and flag the open September 2026 month. This selected synthetic sample supports neither causal nor population estimates.
This Nairobi event-level reference is separate from the catalogue’s 100-subscriber monthly-panel preview and from custom orders. Its counts, event fields and checks do not certify the monthly panel, Lagos or an order-specific extract.
Proposed questions for your custom order
- How do subscriber lifecycle state, network usage, recharge and bundle spending, mobility, SIM/device replacement and tariff changes co-evolve through time?
- How do recorded changes in subscriber status and tariff relate to subsequent monthly usage, recharge and bundle spending in the selected Nairobi panel?
Proposed research and analytical applications
- Longitudinal research
- Usage and spending analysis
- Subscriber-state comparison
- Temporal model evaluation
Study and modeling guidance
- Published split unit
- Keep citizen-grouped ResearchSplit assignments. The 500 profiles use 390/50/60 train/validation/test rows; their events use 11,216/1,560/1,780 rows. These are not temporal cohorts.
- Historical event context
- Snapshot* values describe 2026-09-01 and must not be projected backwards as historical predictors. POINT events and MONTH usage periods have different temporal grains.
- Usage and money
- Monthly data MB, voice minutes and SMS survive in EventLabel text, not dedicated typed usage columns or EventDetailJson. AmountLocal is a separate event-specific amount; tariff fee metadata is not transaction spending.
- Separate monthly panel and custom scope
- The existing 100-subscriber monthly panel is a different artifact. A custom forecasting panel requires agreed monthly aggregation, event cutoffs, targets, follow-up and evaluation; this event package does not certify that panel.
Methodology & Quality
- Data generation
- FCTE synthetic city engine
- Data classification
- Synthetic
FCTE generation methods and published city/domain checks provide source context. The order defines the final extract and the release-specific checks, documentation and limitations supplied with it.
Reference validation
These results describe the frozen Nairobi reference release NRB-TEL-HF-V1. Validation for a prepared custom extract is defined with that order.
The supplied Nairobi 500-subscription event reference only; not the separate catalogue monthly panel or a full eligible population
Publisher reference gates
10/10 PASS
Publisher report for the supplied Nairobi event package
Packaged-row checks
0 violations
Ten boundary, membership, split and amount checks; holder age uses fallback 18, not an unavailable actual ops.Config threshold
Manifest data assets
9/9 verified
Compressed data and companion hashes and row counts; not all package documentation files
Event row-key duplicates
0
SyntheticSubscriberId + EventOrder in all 14,556 event rows
Pinned research CSV byte parity
6/6 verified
All subscriber-profile and telecom-event split files match the decompressed supplied files at the linked October 4 revision; not complete-package or current-main parity.
Reference release modeling notes
- Published split unit
- Keep citizen-grouped ResearchSplit assignments. The 500 profiles use 390/50/60 train/validation/test rows; their events use 11,216/1,560/1,780 rows. These are not temporal cohorts.
- Historical event context
- Snapshot* values describe 2026-09-01 and must not be projected backwards as historical predictors. POINT events and MONTH usage periods have different temporal grains.
- Usage and money
- Monthly data MB, voice minutes and SMS survive in EventLabel text, not dedicated typed usage columns or EventDetailJson. AmountLocal is a separate event-specific amount; tariff fee metadata is not transaction spending.
- Separate monthly panel and custom scope
- The existing 100-subscriber monthly panel is a different artifact. A custom forecasting panel requires agreed monthly aggregation, event cutoffs, targets, follow-up and evaluation; this event package does not certify that panel.
Reference boundaries
- Not observed operator/consumer data or a population estimate
- No real phone numbers, MSISDN or citizen core PhoneNumber are included
- Current snapshot fields must not be projected backwards as historical event features
- Monthly usage aggregates are periods, not point events on the first day
- AmountLocal meanings depend on event type; tariff fee metadata is not transaction spending
- EventDetailJson omitted after 1,353 truncated mobility JSON payloads; detailed usage/device/tariff payload fields are not dedicated columns in the 25-column public events config. Some details survive in EventLabel text, including monthly data MB, voice minutes and SMS; do not confuse these labels with separately typed columns or complete original JSON payloads.
- Featured companion is deliberately excluded from the research sample
- CC-BY-NC-4.0 public-reference rights and separately agreed commercial/custom-order rights must remain distinct
- The holder-age check uses SQL fallback age 18; the actual configured threshold is unavailable in the package.
- Deterministic source-sample selection was not reproduced without the full eligible population.
- The page links the October 4 published revision: six research CSV files match the decompressed supplied package. This verifies research-data contents, not identical gzip wrappers, README, manifest or the complete package.
- The newer October 6 public revision contains different records and 14,320 events. Its manifest has six stale file digests; it is not represented by this page’s 14,556-event reference facts.
Known limitations
- This selected synthetic sample is not population representative.
- Monthly totals are not individual event histories.
- The source panel checks and sample manifest do not certify the full telecom domain or a paid extract.
- Synthetic records describe generated scenarios; they do not establish observed population prevalence or real-world causal effects.
Delivery & License
- Catalogue base price
- $599 USD
- Price basis
- Base price per selected city
- Availability
- Prepared after order
The displayed catalogue amount is a base price per selected city. Your requested dates, population, fields, formats and usage rights determine the agreed scope and final quote. Applicable published discounts and taxes are confirmed with that quote.
Your prepared package
Your order confirms the data files, documentation, evidence, examples and release controls included in the prepared package.
- Agreed data extract
- Schema and field documentation
- Release-specific evidence and limitations
- Agreed loading or analysis examples
- Licence, manifest, checksums and delivery note
Package layout
Illustrative filenames. Your agreed order lists the final files and formats. Certification, benchmark splits and model examples are included only when specified.
SD_RESEARCH_DATASET/ ├── data/ │ ├── subscriber_months.parquet │ └── subscriber_context.parquet ├── documentation/ │ ├── README.md │ ├── DATA_DICTIONARY.csv │ ├── SCHEMA.md │ └── METHODOLOGY.md ├── evidence/ │ ├── CERTIFICATION.json │ ├── VALIDATION_REPORT.pdf │ └── KNOWN_LIMITATIONS.md ├── MANIFEST.json ├── CHECKSUMS.sha256 └── LICENSE.pdf
Delivery process
Confirm the cities, date range, records, formats, rights and price in writing. The order-specific extract is then prepared and supplied under the agreed payment and delivery conditions.
One secure download package for the agreed release and licensed entity.
Delivery timing, access expiry, formats and prerequisites are confirmed in the accepted order.
Support, refreshes, future releases, hosting and APIs are included only when expressly ordered.
Data Licensing Agreement · 2026-10
A non-exclusive, non-transferable licence for the named customer within the accepted scope and duration.
- Permitted use
- Internal research, development, testing and analytics within the accepted order.
- Model training
- Model training, external applications, commercial distribution and access by other legal entities must be expressly addressed in the order.
- Redistribution
- Raw or substantially reconstructable data cannot be resold, published or redistributed without separate written authorization.
- Attribution
- Preserve release provenance and applicable third-party notices. Published outputs must identify the data as synthetic.
Reference release license
The 2026-10-04 Nairobi reference uses CC-BY-NC-4.0. Existing rights granted for that named release remain available. Custom orders use the licensing terms confirmed for the agreed scope.
Supplied public research reference under CC-BY-NC-4.0; commercial and custom-order rights require separate agreement.
New website dataset materials follow the publisher’s CC-BY-NC-4.0 policy. Previously released material retains its original license.
FAQ
Is this real personal data?
No. The records are synthetic and do not represent actual identifiable people or reproduce an individual's historical experience.
What does the catalogue price cover?
The displayed amount is a base price per selected city. The final scope and quote are agreed for your requested dates, population, fields, formats and rights. Adding a dataset to the cart starts the existing checkout inquiry; it does not purchase an unrestricted custom extract.
What can I inspect before ordering?
Inspect the selected synthetic sample and its stated schema, period and limitations. Sample counts and fields apply to that sample only.
Is the dataset prepared for my requested dates?
Yes. Regular and Research datasets are prepared after an order, with the city, date range, population and other required scope agreed before preparation.
Can records be joined across domains?
The profile explains the intended relationships. The order-specific schema documents the supported keys, effective dates and joins for your agreed extract.
Can I use the dataset for AI or machine learning?
Review the study-specific inputs, outcomes and information cutoffs. The pinned Nairobi reference uses CC-BY-NC-4.0. All six research CSV files match the supplied package after decompression; newer main contains different records, and full-package byte parity is not claimed. Custom model training and external applications must be expressly addressed in the accepted order.
Can I use the data commercially?
Commercial use follows the accepted dataset-specific scope and licence. A sample download, cart selection, inquiry or consent checkbox does not activate a commercial licence or complete payment.
Can I request another city, period or format?
Yes. Use Data Request to describe the required city, population, dates, record concepts and formats. Onboarded city options for this product: Nairobi. Other African major cities can be requested; availability, scope and the quote must be confirmed.
What documentation and evidence are included?
The accepted order confirms the schema, methodology, evidence, examples and release controls supplied with your prepared extract.
How do I receive the dataset?
After scope, terms and payment conditions are agreed, Fidelra Africa prepares the order-specific extract and supplies it through a secure download package. Timing and download-access expiry are defined in the order.
Does the public reference licence cover my custom order?
The supplied Nairobi reference uses CC-BY-NC-4.0 under the reference licence scope stated in this profile. A separately prepared custom extract follows the accepted order and the Data Licensing Agreement; reference counts and date coverage do not define that order.
Need different data?
Tell us the entities, fields, cities and time span you need.