Aggregation methods for chart metrics

Why aggregation matters

Raw data often contains multiple entries (e.g. multiple properties, transactions, or listings). Aggregation helps you answer questions like:

• What is the total transaction volume?

• What is the average property price?

• What is the highest sale price in a region?

The wrong aggregation can lead to misleading charts, so always match the method to your editorial goal.

Available aggregation methods

1. Sum (default)

What it does:
Adds up all values in the selected field.

When to use it:
Use sum when you want a total value.

Real estate examples:

• Total transaction volume in a city

• Total number of square meters sold

• Total commission generated

Good use case:
“Total sales volume in Barcelona in 2024”

Watch out:
Do not use sum for metrics like price per property—it will inflate the result.

2. Average

What it does:
Calculates the mean (average) of all values.

When to use it:
Use average to show a typical value.

Real estate examples:

• Average property price

• Average price per square meter

• Average days on market

Good use case:
“Average price per m² in Lisbon”

Watch out:
Averages can be skewed by very high or very low values (e.g. luxury listings).

3. Min

What it does:
Returns the smallest value in the dataset.

When to use it:
Use min to highlight the lowest value.

Real estate examples:

• Lowest property price in a region

• Shortest time on market

Good use case:
“Entry-level property price in Berlin”

4. Max

What it does:
Returns the highest value in the dataset.

When to use it:
Use max to highlight the top end of the market.

Real estate examples:

• Highest sale price

• Most expensive listing

• Largest property size

Good use case:
“Top luxury property price in Dubai”

5. Count

What it does:
Counts the number of records. Works with numeric and text fields.

When to use it:
Use count to show volume or quantity.

Real estate examples:

• Number of properties listed

• Number of transactions

• Number of cities represented (even if stored as text)

Good use case:
“Number of listings per property type”

Key advantage:
Works even if your field is not numeric (e.g. property IDs or city names).

6. Countdistinct

What it does:
Counts the number of unique values.

When to use it:
Use countdistinct to avoid duplicates and measure variety or coverage.

Real estate examples:

• Number of unique cities with listings

• Number of unique agents involved

• Number of distinct property types

Good use case:
“Number of unique markets Engel & Völkers operates in”

Watch out:
If multiple entries share the same value, they are counted only once.

Choosing the right aggregation

Use this quick guide when deciding:

• Use sum → for totals

• Use average → for typical values

• Use min/max → for range boundaries

• Use count → for quantity of records

• Use countdistinct → for unique values

Good examples vs. bad examples

Good examples

• Sum of transaction volume to show total market size

• Average price per m² to compare regions

• Count of listings to show supply

• Countdistinct of cities to show market coverage

Bad examples

• Sum of property prices to represent market pricing

• Average of IDs or non-meaningful numeric fields

• Count instead of countdistinct when duplicates exist

• Max used when the goal is to show a typical value

How filters affect aggregation

Aggregation methods are always applied after filters. This means:

• Filters define which data is included

• Aggregation defines how that data is summarized

Example:

• Metric: Property price

• Aggregation: Average

1. Without filters → average price across all properties

2. With filter “Property type = Villa” → average price of villas only

3. With filter “City = Paris” → average price in Paris only

Even though the aggregation method stays the same, the result changes significantly.

Aggration methods & filters: good examples vs. bad examples

Good examples

• Applying “City = Barcelona” and using sum to show total sales volume in that city

• Using average price with a filter for “Apartments” to show a specific segment

• Clearly labeling charts (e.g. “Average price of villas in Marbella”)

Bad examples

• Comparing a filtered chart (e.g. villas only) with an unfiltered chart

• Using average without realizing luxury properties skew the result within a filtered set

• Not reflecting filters in the chart title or description

Aggration methods & filters: best practice for editors

• Always check active filters before interpreting a chart

• Reflect filters in your chart title or surrounding content

• Be cautious when combining multiple filters + averages

• When comparing charts, ensure they use consistent filters

Best practices for editors

• Always start with your question: what do you want the chart to show?

• Check if your field contains duplicates before choosing count or countdistinct

• Be careful with averages in luxury markets (they can be misleading)

• Label your charts clearly (e.g. “Average price” vs. “Total volume”)

• When in doubt, validate your result with a quick manual check

• Filters change the dataset before aggregation is applied. This means the same aggregation method (e.g. sum or average) can produce very different results depending on active filters.

Choosing the correct aggregation ensures your charts are accurate, trustworthy, and useful for your audience.