Skip to main content

Documentation Index

Fetch the complete documentation index at: https://honeydew.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

Recipes

Each recipe pairs a modeling pattern with a step-by-step implementation - showing how to combine entities, attributes, metrics, and parameters to solve real business questions in Honeydew.

What a recipe contains

Most recipes include some or all of these sections:
  • When would you use this? - the business question the recipe answers
  • Prerequisites - setup needed before you start
  • Step-by-step implementation - the objects to create, with full YAML
  • Sample output - results you can verify against
  • Key design notes - design decisions and trade-offs

Dataset: tasty_bytes

All recipes use the tasty_bytes workspace - a food-truck order dataset. The key entities are:
EntityTypeDescription
order_detailFactLine-item orders - revenue per item sold
order_headerFactOrder-level - customer, date, city, truck
menuDimensionMenu items - category and subcategory
dateTime SpineCalendar spine used by time-based metrics

Available recipes

Filtered metrics

Metric for a specific category

Isolate revenue for a specific product category using a filtered metric.

Category share of total

Compute a specific category’s share of total revenue using ratio metrics.

Time-based metrics

Active customers & stickiness

Measure customer engagement with daily, weekly, and monthly active user counts.

Period-over-period growth

Compare revenue to the same period last year and compute the growth rate.

Fixed and dynamic grouping

Average order value

Average spend per order, computed correctly regardless of how the data is sliced.

Share above a threshold

Share of orders above a threshold, using either a fixed amount or a data-driven percentile.

Average daily revenue

Average revenue per day using dynamic grouping that adapts to the user’s slicing context.

BI parameters

BI parameters

Route a single metric to different calculations based on a BI user’s filter selection.