Skip to main content

Guide: track a brand's health over time

Goal: a smooth, multi-metric time series for a single brand — the kind of view you would put on a brand-health dashboard.

Steps​

  1. Find your region, sector_id and brand_id (taxonomies).
  2. Pick the metrics that represent "health" for you (e.g. index, buzz, consider).
  3. Choose a rolling window and a moving average to reduce daily noise.

Request​

{
  "meta": { "version": "v1" },
  "data": {
    "queries": [
      {
        "id": "brand-health",
        "entity": { "region": "us", "sector_id": 1, "brand_id": 1007 },
        "period": {
          "start_date": { "days": 180 },
          "end_date":   { "days": 1 }
        },
        "metrics_score_types": {
          "index": "net_score",
          "buzz": "net_score",
          "consider": "positives"
        },
        "moving_average": 7,
        "scoring": "total",
        "filters": []
      }
    ]
  }
}

POST this to /v1/analyses/execute.

Reading the result​

The single query returns one block whose data cube is indexed values[metric][date][perspective]. Plot the score perspective per metric over date. Watch last_period_is_complete — if false, treat the final point as provisional.

Variations​

  • Swap scoring to aware to track health among people who know the brand.
  • Add a benchmark by including a second query for the sector median — see Benchmark against a sector.
  • Use resample instead of a moving average to report weekly/monthly points.