Project Plan

Overview

This solution defines how HotelMap can measure event-related accommodation demand and hotel conversion performance with privacy-first data practices.

The objective is to identify attendees who are likely to require accommodation, then compare that population with those who complete a hotel booking.

What We Must Measure

MetricDefinitionPurpose
Addressable marketAttendees likely to need accommodation (for example, non-local or long journey time)Estimate hotel demand opportunity
Hotel capture countAddressable attendees that complete a hotel booking through tracked channelsMeasure conversion into booking outcome
Capture ratehotel_capture_count / addressable_market_countPrimary commercial KPI

Determining Non-Local Attendees

A binary non-local flag can be collected directly or inferred through coarse location signals. Travel time or travel band is more robust than pure distance because it better reflects real effort to attend.

MethodInputPrecisionPrivacy FootprintRecommended Use
Direct declarationAccommodation needed (yes/no) or journey 3+ hours (yes/no)MediumLowPrimary
Coarse origin inferenceZIP/postcode or city-country, mapped to travel-time bandMedium-highMediumSecondary
IP geolocationApproximate network locationLow-mediumHighFallback only

Implementation Options

Recommended
Direct ask + derived travel band

Combine a short registration question with travel-time derivation from coarse location where needed. This balances actionable accuracy with low privacy overhead.

A
Ask directly at registration
Request accommodation need or estimated travel distance in a yes/no format.
B
Infer from coarse location
Collect postcode/ZIP or city and derive travel time band to event location, then discard precise source data quickly.
C
Use IP as fallback only
Apply only when users skip fields and ensure truncation or fast deletion due to higher legal and privacy sensitivity.

Capture Rate Formula

Core KPI for events with accommodation demand: measured against the addressable market, not total registrations.

typescript
const addressableMarketCount = nonLocalAttendeeCount;const hotelCaptureCount = bookedAddressableAttendeeCount;const captureRate =  addressableMarketCount === 0 ? 0 : hotelCaptureCount / addressableMarketCount;
Reporting note
Break down capture rate by event, travel-time band, and ticket type to identify conversion gaps and operational opportunities.

Data Model Decisions

Keep operational data and analytics signals in separate layers so each dataset stays constrained to its purpose.

LayerTypical FieldsRetention Guidance
Operational registration systemsName, contact data, ticket selections, payment and customer support detailsBusiness and legal necessity
Analytics measurement layerAnonymized event key, travel band, addressable flag, booking outcome, timestampShort row-level retention, aggregate long term

Privacy and Legal Framing

The model can generally rely on legitimate interests when data is minimized, transparent, and proportionate. Consent may still be required for non-essential tracking in some implementations.

Data ElementTypical BasisControl Measure
Direct accommodation need fieldLegitimate interestsKeep question narrow and purpose-specific
Coarse origin for travel bandLegitimate interestsDerive quickly and remove precise source values
Cookie or tag-based cross-site attributionConsent (jurisdiction dependent)Gate non-essential tracking and respect opt-out

Recommended Implementation Sequence

1
Finalize the operational definition
Agree threshold and fallback rules for addressable/non-local classification.
2
Instrument registration flow
Capture direct need flags and optional coarse origin with clear disclosure at collection time.
3
Implement booking attribution
Use direct booking add-ons or tagged booking links to connect registration outcomes to hotel bookings.
4
Build KPI dashboard
Track addressable market, capture count, and capture rate with breakdowns by event and distance band.
5
Apply privacy controls
Add retention windows, aggregation thresholds, and role-based access controls to limit identifiable risk.
6
Run legal review
Confirm lawful basis, notices, and consent logic where required by jurisdiction.
7
Pilot then scale
Validate data quality and conversion logic on a small set of events before broad rollout.

Technical Feasibility

Feasibility summary
This solution is technically feasible with common event registration tooling and a lightweight analytics model. Main complexity sits in attribution wiring, data quality checks, and governance setup rather than algorithmic difficulty.