Data structure of the export

Structure

• The export in the background consists of one or more parts, metrunio-export-YYYY-MM-DD-001.zip and following, with at most about 250,000 rows per part, plus manifest.json. The direct download is a single ZIP with manifest.json inside.
• Per table there is one file json/<table>.json and one file csv/<table>.csv with the same rows. If a table does not fit into one part, json/<table>.0002.json, json/<table>.0003.json and so on follow in the next part.
• manifest.json states format and version (format, version), the time (createdAt), the parts with files, size and SHA-256 (parts), per table the row count, files and description (tables), the tables deliberately left out with their reason (excluded) and the omitted fields (droppedColumns).
• consistent: true means all parts show the same state. false means the export was interrupted and continued in a new state; a row that changed in between may then appear twice or not at all. For a fixed picture, export after closing the account, when nothing changes any more, or start a new export.

Formats and character set

• Character set UTF-8 everywhere, without byte order mark, line ending \n.
• JSON: a list of objects, one object per row, field names as below.
• CSV: semicolon as separator, field names in the first line. Fields containing a semicolon, quotation mark or line break are enclosed in double quotes. An empty field is no value (null).
• If a text in the CSV starts with =, +, -, @ or a tab, it is preceded by an apostrophe so that a spreadsheet does not run it as a formula. The JSON file contains the text unchanged.
• Points in time per ISO 8601 in UTC, such as 2026-10-18T09:15:00.123+00:00; dates as YYYY-MM-DD.
• Numbers with a point as decimal separator. Amounts are in the currency of the row (field currency, ISO 4217), not in cents.
• IDs are UUIDs. A reference such as “Reference to sites.id” points to the ID in the named file; this is how you put websites, journeys, conversions and contacts back together.
• Structured values (JSON, lists) appear as an object or list in the JSON file and as JSON text in the CSV.
• A field with a fixed list of values lists all possible values below.

Retrieval by a new provider with an export key

So that a new provider can fetch your data itself, create an export key under “Exports & AI connections”, card “Take everything with you”, and pass it on. It opens the full export and nothing else: no reports, no read API, no changes. It is available on every plan, including Start and Pro, and also during the 30 days after closing. The read API remains a feature from Growth.

1. 01POST /v1/exports/full-jobs with the header Authorization: Bearer xk_… starts the export in the background or returns the running one.
2. 02GET /v1/exports/full-jobs/:id shows its status. When status is succeeded, parts lists the parts with size and SHA-256.
3. 03GET /v1/exports/full-jobs/:id/parts/:number downloads a part, GET /v1/exports/full-jobs/:id/manifest.json the index.
4. 04For a small account, GET /v1/exports/full.zip also works directly. Above 300,000 rows it answers 409 FULL_EXPORT_TOO_LARGE; then use the route above.

• The parts are available for 7 days, after closing at most until the account is deleted. After that they are deleted and a new export is needed.
• Every retrieval is recorded in your account log. When an export started with an export key is ready, the admins receive an email with the name of the key.
• You can revoke the key at any time under “Take everything with you”, also after closing. Give it a validity that matches the switch.
• People who manage the account (admin) can create it. It returns the same data as the full export in the portal, including the members’ email addresses and the log.

What is deliberately left out

Credentials for other platforms, run control without content of its own and records that belong to us as the provider are not included. Hashes of keys, invitations and unsubscribe links are missing from every table; the row itself remains. Per table:

• tenant_secrets: Credentials for other platforms, encrypted; the new provider connects with its own credentials
• integration_credential_references: References to stored credentials, without content of their own
• oauth_tokens: Access tokens for MCP and the read API; worthless after switching
• portal_login_tokens: Sign-in links with a short lifetime
• ad_oauth_grants: Intermediate state of an ongoing sign-in with an ad platform
• search_console_oauth_states: Intermediate state of an ongoing sign-in with Google
• request_nonces: Replay protection of the collector, hashes with expiry
• event_queue: Processing queue; the content is in raw_events
• processing_attempts: Processing attempts of the worker; the content is in raw_events
• dead_letter_events: References to failed processing; the event is in raw_events
• web_engagement_pending: Intermediate state of an ongoing visit until it ends
• mirror_deletions: Run control of the analytics mirror
• mirror_refreshes: Run control of the analytics mirror
• usage_window_checks: Run control of the usage check
• suppression_scan_state: Run control of the suppression list check
• report_mail_claims: Run control of report delivery
• support_contacts: Belongs to the platform: support enquiries, also from people without an account
• lifecycle_messages: Delivery log of the provider’s emails to the account
• tenant_milestones: The provider’s analysis of the account’s history
• tenant_signup_origins: The provider’s measurement of where the sign-up came from
• paid_revenue: The provider’s revenue ledger for partner commissions
• tenant_deletions: The provider’s deletion log
• legal_holds: The provider’s retention hold

All files and fields

97 files per format, in the order of the export: first the account and its members, then each table by name. Without a meaning of its own, name, format and the table description explain the field.

tenants

The account itself: name, plan, time zone, currency, state and account-level settings. Exactly one row. Files: json/tenants.json and csv/tenants.csv.

users

The members of the account with ID, email address, name and language. Without password or sign-in data. Files: json/users.json and csv/users.csv.

ad_costs

Costs, impressions and clicks per ad and day, as reported by the ad platform. Files: json/ad_costs.json and csv/ad_costs.csv.

ad_groups

Ad groups of the campaigns, with the ID at the platform. Files: json/ad_groups.json and csv/ad_groups.csv.

ads

Ads of the campaigns, with the ID at the platform. Files: json/ads.json and csv/ads.csv.

agency_report_logos

An agency logo for the monthly report, as an image (PNG or JPEG). Files: json/agency_report_logos.json and csv/agency_report_logos.csv.

ai_usage_monthly

Number of questions to the assistant per month and model. A counter without question or answer text. Files: json/ai_usage_monthly.json and csv/ai_usage_monthly.csv.

alert_events

Triggered alerts with the figures that led to them. Files: json/alert_events.json and csv/alert_events.csv.

alert_rules

Alert rules: silence for a number of hours, or a change in a metric compared with the previous week. Files: json/alert_rules.json and csv/alert_rules.csv.

allowed_domains

Domains from which a website accepts events. Files: json/allowed_domains.json and csv/allowed_domains.csv.

anonymous_reach

Reach without consent: counts of page views and sessions without identifiers. Files: json/anonymous_reach.json and csv/anonymous_reach.csv.

api_keys

API keys with name, scope, prefix, expiry and last use. Without the key itself and without its hash. Files: json/api_keys.json and csv/api_keys.csv.

app_links

Connected applications, such as the Shopify app, that measure for a website. Without credentials. Files: json/app_links.json and csv/app_links.csv.

attribution_configurations

Attribution settings: default model, time windows and post-view. Files: json/attribution_configurations.json and csv/attribution_configurations.csv.

attribution_credits

Attributed shares: per conversion, model and touchpoint, the share and the attributed value. Files: json/attribution_credits.json and csv/attribution_credits.csv.

attribution_model_fits

Basis of the data-driven model as calculated from the data. Files: json/attribution_model_fits.json and csv/attribution_model_fits.csv.

attribution_runs

Attribution calculation runs with time and state. Files: json/attribution_runs.json and csv/attribution_runs.csv.

audit_logs

Account log: who changed, created, exported or deleted what, and when. Files: json/audit_logs.json and csv/audit_logs.csv.

billing_line_items

Additional billing items, such as overage and agency surcharge, per billing period. Files: json/billing_line_items.json and csv/billing_line_items.csv.

billing_subscriptions

The subscription at the payment provider: plan, billing interval, state and term. Files: json/billing_subscriptions.json and csv/billing_subscriptions.csv.

campaign_costs

Advertising costs per campaign and day, with impressions, clicks and the conversions the platform itself reports. Files: json/campaign_costs.json and csv/campaign_costs.csv.

campaigns

Campaigns per channel, with the ID at the platform and run time. Files: json/campaigns.json and csv/campaigns.csv.

channels

Channels of the account, such as Google Ads, Meta or email. Files: json/channels.json and csv/channels.csv.

client_report_recipients

Recipients of the monthly report that an agency enters per client. Files: json/client_report_recipients.json and csv/client_report_recipients.csv.

client_report_recipients_off

Since when the monthly report recipient list has been switched off for an agency. Files: json/client_report_recipients_off.json and csv/client_report_recipients_off.csv.

client_report_settings

Monthly report settings per client account: interval, language, content. Files: json/client_report_settings.json and csv/client_report_settings.csv.

client_report_suppressions

Unsubscriptions from the monthly report, only as a hash of the address. Files: json/client_report_suppressions.json and csv/client_report_suppressions.csv.

client_takeover_invitations

Invitations from an agency to take over an existing account as a client account. Files: json/client_takeover_invitations.json and csv/client_takeover_invitations.csv.

closure_refund_claims

Refunds from an account closure that are still open at the payment provider or are refunded manually, with amount and attempts. Files: json/closure_refund_claims.json and csv/closure_refund_claims.csv.

cohort_snapshot_months

Saved cohorts: values per month since the first purchase. Files: json/cohort_snapshot_months.json and csv/cohort_snapshot_months.csv.

cohort_snapshot_periods

Saved cohorts: the periods that were calculated. Files: json/cohort_snapshot_periods.json and csv/cohort_snapshot_periods.csv.

cohort_snapshot_scopes

Saved cohorts: scope of the calculation, such as website and channel. Files: json/cohort_snapshot_scopes.json and csv/cohort_snapshot_scopes.csv.

cohort_snapshot_states

Saved cohorts: calculation state per scope. Files: json/cohort_snapshot_states.json and csv/cohort_snapshot_states.csv.

consent_configurations

Consent settings per website: categories, behaviour before the decision, banner text version. Files: json/consent_configurations.json and csv/consent_configurations.csv.

consent_states

The most recently reported consent state per browser. Files: json/consent_states.json and csv/consent_states.csv.

contract_confirmations

Copies of the confirmations sent when the contract was concluded. Files: json/contract_confirmations.json and csv/contract_confirmations.csv.

conversion_definitions

Goals: which events count as a conversion, with trigger and value. Files: json/conversion_definitions.json and csv/conversion_definitions.csv.

conversion_forwards

Conversions sent to ad platforms, per destination with state and attempts. Files: json/conversion_forwards.json and csv/conversion_forwards.csv.

conversion_items

Line items of an order: product, quantity, price, cost. Files: json/conversion_items.json and csv/conversion_items.csv.

conversions

Conversions: orders, enquiries and other goals with value, refund, currency and attribution. Files: json/conversions.json and csv/conversions.csv.

cost_connectors

Connections for fetching advertising costs automatically. Without credentials. Files: json/cost_connectors.json and csv/cost_connectors.csv.

cost_sheets

Spreadsheets with advertising costs that are read regularly, and the state of the last run. Files: json/cost_sheets.json and csv/cost_sheets.csv.

creatives

Creatives of the ads, with the ID at the platform. Files: json/creatives.json and csv/creatives.csv.

data_quality_issues

Data quality findings, such as missing consent or duplicate orders. Files: json/data_quality_issues.json and csv/data_quality_issues.csv.

diagnostic_runs

Setup check runs with their results. Files: json/diagnostic_runs.json and csv/diagnostic_runs.csv.

dismissed_notices

Portal notices that a person has dismissed. Files: json/dismissed_notices.json and csv/dismissed_notices.csv.

email_campaign_stats

Figures reported by an email platform per sent campaign. Files: json/email_campaign_stats.json and csv/email_campaign_stats.csv.

email_platform_connections

Connections to email platforms such as Klaviyo or Brevo. Without credentials. Files: json/email_platform_connections.json and csv/email_platform_connections.csv.

experiment_groups

Groups of an experiment, such as test and control. Files: json/experiment_groups.json and csv/experiment_groups.csv.

experiment_periods

Periods per group of an experiment. Files: json/experiment_periods.json and csv/experiment_periods.csv.

experiments

Experiments on the measured effect of campaigns. Files: json/experiments.json and csv/experiments.csv.

export_jobs

Exports of the account: who fetched which export when, including this full export. Files: json/export_jobs.json and csv/export_jobs.csv.

identities

Recognised visitors or customers per website, with first and last contact. Files: json/identities.json and csv/identities.csv.

identity_links

Attributes by which an identity is recognised, such as an anonymous ID or the hash of an email address. Files: json/identity_links.json and csv/identity_links.csv.

import_jobs

Imports of costs and offline conversions with state and result. Files: json/import_jobs.json and csv/import_jobs.csv.

integrations

Installation per website: method (snippet, tag manager, plugin) and state. Files: json/integrations.json and csv/integrations.csv.

invitations

Open and accepted invitations to the account. Without the invitation link. Files: json/invitations.json and csv/invitations.csv.

journeys

Customer journeys: the contacts of an identity up to the conversion, with start and last activity. Files: json/journeys.json and csv/journeys.csv.

measurement_domains

Custom measurement domains of the websites with DNS check state and certificate. Files: json/measurement_domains.json and csv/measurement_domains.csv.

memberships

Who is a member of the account and with which role. Files: json/memberships.json and csv/memberships.csv.

oauth_grants

Approvals for AI tools via OAuth: who allowed which tool to read. Without tokens. Files: json/oauth_grants.json and csv/oauth_grants.csv.

onboarding_states

State of the setup in the portal. Files: json/onboarding_states.json and csv/onboarding_states.csv.

order_bridge_candidates

Matching data of the pixel bridge between server order and browser, with a short retention period. Files: json/order_bridge_candidates.json and csv/order_bridge_candidates.csv.

outbound_deliveries

Deliveries to your own destinations such as webhooks, Slack or Teams, with state and attempts. Files: json/outbound_deliveries.json and csv/outbound_deliveries.csv.

outbound_endpoints

Your own destinations for outgoing messages: webhooks, Slack, Teams. Without signing secret. Files: json/outbound_endpoints.json and csv/outbound_endpoints.csv.

outbound_recipient_trail

Trail of which destination a message went to, for access and erasure requests. Files: json/outbound_recipient_trail.json and csv/outbound_recipient_trail.csv.

payment_connections

Connections to Stripe or PayPal as a source of purchases. Without credentials. Files: json/payment_connections.json and csv/payment_connections.csv.

payment_events

Incoming log of payment providers, without order numbers or identifiers. Files: json/payment_events.json and csv/payment_events.csv.

payment_rejected_orders

Rejected payment purchases, only as a hash with time. Files: json/payment_rejected_orders.json and csv/payment_rejected_orders.csv.

placements

Placements on which ads were shown. Files: json/placements.json and csv/placements.csv.

privacy_requests

Data subject requests such as access and erasure, with state. Files: json/privacy_requests.json and csv/privacy_requests.csv.

processing_agreements

Data processing agreements concluded, with version, company and time. Files: json/processing_agreements.json and csv/processing_agreements.csv.

pushback_destinations

Destinations for returning conversions to ad platforms. Without credentials. Files: json/pushback_destinations.json and csv/pushback_destinations.csv.

raw_events

Raw events as they were accepted: page views, clicks, orders, enquiries and more, within the retention period. Files: json/raw_events.json and csv/raw_events.csv.