HK Express Membership Engagement PlatformDemo environment, sample data only

Aggregation options for the Japan Explorer mission

The mission rule needs one number: Japan cities flown in the campaign window. The same rule can get it in three ways. Pick a member to compare.

Option 1: Built-in functions in the BRE

Runs in the BRE on every event

count(member.flownSectors,
  #.country == "JP" and
  date(#.flown) >= date(campaign.start) and
  date(#.flown) <= date(campaign.end))

Japan sectors in window

…

No rule fired

Maintained by	BU team, in an expression box
Freshness	Up to date at each event
Data sent per evaluation	…
Evaluation time	…

Open the rule

Option 2: Prepared at source (GCP)

Counted in GCP, sent in the daily sync

gcp.japanCitiesInWindow

// computed in GCP from all UO data,
// arrives with the daily sync

Japan cities in window

…

No rule fired

Maintained by	UO data team for the count, BU team for the rule
Freshness	As of the last daily sync
Data sent per evaluation	…
Evaluation time	…

Open the rule

Option 3: JavaScript function in the BRE

Runs in the BRE on every event

const cities = new Set(sectors
  .filter((s) => s.country === "JP"
    && s.flown >= start && s.flown <= end)
  .map((s) => s.city));
return { count: { japanCities: cities.size } };

Japan cities in window

…

No rule fired

Maintained by	Developer for the code, BU team for the rule
Freshness	Up to date at each event
Data sent per evaluation	…
Evaluation time	…

Open the rule

What each option means to maintain

All three feed the same decision table. The difference is where the count lives and who looks after it.

	1. Built-in functions	2. Prepared at source	3. JavaScript function
Counts different values	No. Count, sum, average, min and max only	Yes, any logic the data team writes	Yes, any logic in code
History needed at runtime	Yes, sent with every event	No, one number	Yes, sent with every event
Reuse in reports and CDP	No, the count lives inside one rule	Yes, the same figure feeds dashboards and segments	No, the count lives inside one rule
Change needs	BU team, test and publish	Data team change in GCP	Developer, code review and test