Hi Team, I've got a duplicates trouble now. This is the link: Untitled - GoodData Cloud this player only have one record last month and the points was 20.8 see snapshot below. but the GoodData show the result as 145724.80. why is that and how to fix it?

Thinking completed I'm sorry, but the provided knowledge sources do not contain the specific details of your metric calculation or full schema definition to diagnose the exact join multiplier on your dashboard. However, based on the behavior and common modeling patterns in GoodData, an inflated aggregation (such as seeing

145724.80

instead of

20.8

) is typically caused by one of the following: 1. Grain and Fact Duplication / Cartesian Multiplier: If the metric executes a

SUM

over a dataset where rows are multiplied across relationships (for example, joining across a 1:N or M:N relationship to another activity/session table, or joining on a non-unique attribute), the single fact value of

20.8

gets summed repeatedly for each matching row in the joined table [Answers]. In this case,

145724.80 / 20.8 = 7,006

, indicating that the underlying record is likely being joined and multiplied by 7,006 rows. 2. Missing or Incorrect Table Grain: If the table containing the points balance/credited values does not have a primary key/grain properly defined, or if player-level summary attributes/facts are exposed in an aggregated detail table without proper grain definitions, GoodData cannot correctly resolve the aggregation path [Answers]. 3. Joining/Filtering via Non-Unique Fields: Joining or grouping across attributes that are not unique identifiers causes rows to duplicate across matching values [Answers]. Recommended Checks • Check the Metric and Attributes: Verify which attributes are currently in the visualization. If attributes from another fact or detail table (e.g., sessions, transactions, or logs) are present, removing them will show whether the inflation is coming from a fan-out join. • Inspect the Generated SQL: You can inspect the exact query being sent to your database. In the Analyze tab, open the visualization and change

/edit

at the end of the URL to

/debug

to download the

explainAfm.zip

archive containing the generated SQL query [Answers]. • Review LDM Grain and Relationships: Ensure the dataset containing points credited has its primary key/grain defined and that relationships between players and transaction datasets are 1:N rather than causing circular or M:N fan-outs [Answers; Answers]. Handing Off to Support Resolving the exact relationship path and query multiplication requires inspection by a GoodData Support team member. To help the Support team investigate, please add the following details to this thread: • Permission for Support to inspect your workspace (

a73625a4a0c54c408112daeb13296852

). • The exact MAQL definition of the metric displaying the points value. • The names of the datasets involved in this visualization and how they are connected in your Logical Data Model (LDM). • The generated

explainAfm.zip

or query from

/debug

if available. Support Options • Community support (Startup & Professional): You are already in the right place: the GoodData Support team monitors this Slack and a team member will join the conversation when human review is needed, so please keep the discussion and post the requested details in this thread. You can also visit the Community Forums. • 24/7 Support (Enterprise plans): If your account tier includes 24/7 Support or this is a Severity 1 (L1) critical issue, these direct options are available in addition to this thread: ◦ Submit a ticket ◦ Email: support@gooddata.com ◦ Call: +1 (415) 200-0194 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups. 👀 Our Support team monitors all messages and will jump in if needed.

@kapa.ai just using the inbuild sum function.

Thinking completed Thank you for providing the screenshot of the insight. As summarized previously, an inflated aggregation like this—where an inbuilt

SUM(POINTS_CREDITED)

returns

145,724.80

instead of

20.80

(a multiplier of exactly 7,006)—is not caused by the

SUM

function itself, but by a fan-out / Cartesian product in the underlying data model (LDM) [Answers]: 1. Fan-Out via Joined Tables: When

PLAYER ID

,

Date

(

BUS DATE

), and

POINTS CREDITED

are queried together, the generated SQL joins the dataset containing

POINTS CREDITED

with related tables (such as player sessions, transactions, or activity logs). If there are 7,006 rows matching this player and date in a connected table, the engine joins them and sums

20.80

once for every joined row [Answers]. 2. Missing Dataset Grain: If the dataset storing

POINTS CREDITED

does not have its primary key (grain) defined, or if player/points facts are exposed in a table alongside higher-detail records without proper grain separation, GoodData cannot prevent row multiplication during aggregation [Answers]. Human Support Needed Because this issue depends directly on how the datasets, keys, and relationships are structured in your workspace's LDM, resolving the specifics requires review by a GoodData Support team member. The Support team monitors this Slack community and will step into the thread to assist. To help the Support team investigate, please add the following details directly to this thread: • Permission for Support to inspect your workspace (

a73625a4a0c54c408112daeb13296852

). • The name of the dataset where

POINTS CREDITED

originates. • The generated

explainAfm.zip

or SQL query (you can download this by changing

/edit

to

/debug

at the end of the URL while viewing this visualization in Analytical Designer). Support Options • Community support (Startup & Professional): You are already in the right place: the GoodData Support team monitors this Slack and a team member will join the conversation when human review is needed, so please keep the discussion and post the requested details in this thread. You can also visit the Community Forums. • 24/7 Support (Enterprise plans): If your account tier includes 24/7 Support or this is a Severity 1 (L1) critical issue, these direct options are available in addition to this thread: ◦ Submit a ticket ◦ Email: support@gooddata.com ◦ Call: +1 (415) 200-0194 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups. 👀 Our Support team monitors all messages and will jump in if needed.

Thanks, update the LDM and the issue fixed