FAQ, Troubleshooting & Support

This page covers the most common issues encountered when installing and operating Analytics for LS Central, along with step-by-step solutions. Use the quick-links below to jump to a section, or press Ctrl+F (Windows) / Cmd+F (Mac) to search this page — your browser will highlight every match in the text.

ⓘFAQ Common questions answered ⚙Installation & Deployment Setup, Power BI, permissions ⚠Pipeline & Data Processing ADF failures, stored procedures, MERGE errors ⇄Replication Errors LS Central SaaS issues ◕Missing Data Data gaps & empty reports ☰Database Storage & Growth DW growth, staging purge ♻Factory Reset Reset the Data Warehouse ★Update AnalyticsVersion upgrades ✉Contact Support When and how to reach us

ⓘ Get instant help from our AI assistantsAsk Analytics — available on the Analytics for LS Central page. Click the Ask Analytics button in the bottom-right corner. This AI assistant covers all Analytics topics — setup, configuration, reports, and troubleshooting.
Compass — available on every Online Help page. Click the icon in the top-right corner. Compass covers the full breadth of LS Retail topics including LS Central configuration, licensing, and cross-product questions.

Frequently Asked Questions

Answers to the most common questions from partners and customers — covering licensing, setup, Power BI, data, and compatibility.

Licensing & Pricing

Getting Started & Setup

SaaS & Compatibility

Power BI Reports

Installation & Deployment

This section covers errors during the deployment of Analytics, ordered by the stage in the installation process where they typically occur — from initial PowerShell deployment through to Power BI report setup.

Initial Setup Errors (PowerShell Deployment)

Power BI Deployment Errors

Why this happens
Deployment failures typically have one of three root causes: (1) Insufficient Azure permissions — the deploying user must have Key Vault Contributor or equivalent rights; (2) Power BI Pro file-size cap — Pro licenses enforce a hard 1 GB limit that cannot be bypassed without upgrading to PPU or Premium capacity; (3) Outdated Analytics version — always deploy the latest version to avoid known script bugs.

► SaaS topologies For LS Central SaaS, use either Data Director replication (database in Azure) or bc2adls (recommended). Confirm your topology before following deployment instructions.

Pipeline & Data Processing Errors

This section covers all errors related to Azure Data Factory pipeline execution, stored procedure failures, and data processing issues. If a pipeline has failed, start by checking ADF Monitor — hover over the talk-bubble icon in the error column to see the error message.

ADF Pipeline Run Failures (MERGE Statement Errors)

Why this happens
The MERGE statement in Analytics stored procedures assumes each target row matches exactly one source row. When duplicate or multiple-version rows exist in the staging tables — typically caused by Data Director replication updating a row rather than inserting a new one — the MERGE cannot resolve which source row should win and throws SQL error 8672. This is a known limitation of the legacy replication method.

Troubleshooting steps

In ADF Monitor, identify the exact pipeline and activity that failed. Note the stored procedure name in the error.

Check whether two pipeline runs overlapped — this is the most common cause. Run the following query on the Analytics database to detect overlapping runs:

WITH RunLog AS (
							SELECT [ADF Run GUID],
							MIN(PipelineStart) AS startDate,
							MAX(PipelineFinish) AS endDate
							FROM LSInsightAudit
							WHERE [ADF Pipeline Name] = 'All staging Tables'
							GROUP BY [ADF Run GUID]
							),
							OverLap AS (
							SELECT R1.[ADF Run GUID] AS RunID1,
							R2.[ADF Run GUID] AS RunID2,
							R1.startDate AS R1StartDate, R1.endDate AS R1EndDate,
							R2.startDate AS R2StartDate, R2.endDate AS R2EndDate,
							CASE WHEN
							(R1.startDate BETWEEN R2.startDate AND R2.endDate) OR
							(R1.endDate BETWEEN R2.startDate AND R2.endDate) OR
							(R1.startDate < R2.startDate AND R1.endDate > R2.endDate) OR
							(R1.startDate > R2.startDate AND R1.endDate < R2.endDate)
							THEN 'yes' ELSE 'no' END AS OverLapping
							FROM RunLog R1, RunLog R2
							WHERE R1.[ADF Run GUID] <> R2.[ADF Run GUID]
							)
						SELECT * FROM OverLap WHERE OverLapping = 'yes'

If overlapping runs are found, run a Factory Reset to clear the resulting duplicate rows.

Check the staging tables for duplicate rows. The following script scans all staging tables automatically and reports any duplicates found:

DECLARE @column_list NVARCHAR(MAX), @sql NVARCHAR(MAX),
							@table_name SYSNAME
							DECLARE CUR CURSOR FOR
							SELECT TABLE_NAME FROM INFORMATION_SCHEMA.TABLES
							WHERE TABLE_TYPE = 'BASE TABLE' AND TABLE_SCHEMA = 'dbo'
							AND TABLE_NAME LIKE 'stg$%'
							OPEN CUR
							FETCH NEXT FROM CUR INTO @table_name
							WHILE @@FETCH_STATUS = 0
							BEGIN
							SELECT @column_list = ''
							SELECT @column_list = @column_list + '[' + C.COLUMN_NAME + '], '
							FROM INFORMATION_SCHEMA.COLUMNS C
							WHERE C.TABLE_NAME = @table_name AND C.TABLE_SCHEMA = 'dbo'
							SELECT @column_list = SUBSTRING(@column_list, 1, LEN(@column_list) - 1)
							SELECT @sql = 'SELECT ''' + @table_name + ''' AS TABLE_NAME, '
							+ @column_list + ' FROM [dbo].[' + @table_name
							+ '] GROUP BY ' + @column_list + ' HAVING COUNT(*) > 1'
							EXEC sp_executesql @sql
							FETCH NEXT FROM CUR INTO @table_name
							END
							CLOSE CUR
					DEALLOCATE CUR

Also check for duplicates in dimension tables:

DECLARE @column_list NVARCHAR(MAX), @sql NVARCHAR(MAX),
							@table_name SYSNAME
							DECLARE CUR CURSOR FOR
							SELECT TABLE_NAME FROM INFORMATION_SCHEMA.TABLES
							WHERE TABLE_TYPE = 'BASE TABLE' AND TABLE_SCHEMA = 'DW'
							AND TABLE_NAME LIKE 'd%'
							OPEN CUR
							FETCH NEXT FROM CUR INTO @table_name
							WHILE @@FETCH_STATUS = 0
							BEGIN
							SELECT @column_list = ''
							SELECT @column_list = @column_list + '[' + C.COLUMN_NAME + '], '
							FROM INFORMATION_SCHEMA.COLUMNS C
							WHERE C.TABLE_NAME = @table_name AND C.TABLE_SCHEMA = 'DW'
							AND COLUMN_NAME NOT LIKE 'SK_%'
							SELECT @column_list = SUBSTRING(@column_list, 1, LEN(@column_list) - 1)
							SELECT @sql = 'SELECT ''' + @table_name + ''' AS TABLE_NAME, '
							+ @column_list + ' FROM [DW].[' + @table_name
							+ '] GROUP BY ' + @column_list + ' HAVING COUNT(*) > 1'
							EXEC sp_executesql @sql
							FETCH NEXT FROM CUR INTO @table_name
							END
							CLOSE CUR
					DEALLOCATE CUR

If duplicates are confirmed, run a Factory Reset (Analytics Setup > Factory Reset) to clear staging tables and reload from source. This may take several hours on large datasets.
After the Factory Reset, re-run the pipeline manually and confirm all activities complete successfully.
If the error persists, consider increasing the Azure SQL service tier — insufficient tempdb space (S2 = 14 GB only) can cause this error. Upgrade to S3 or above.
If you are on an older version of Analytics, this error may already be fixed in a newer release. See the Fix availability section below and upgrade to the latest version.
If none of the above resolves the issue, it may be a bug in the stored procedure caused by a data pattern not covered in testing. Please contact support with the stored procedure name, the output of the duplicate-check scripts above, and the ADF overlap check results.
Confirm resolution once two consecutive daily pipeline runs complete without error.

ⓘ Long-term recommendation If you are using the legacy Data Director replication method, this issue may recur. We recommend migrating to bc2adls (Azure Data Lake Storage) replication, which is the current supported approach and avoids the row-versioning issue entirely.

Connection and Runtime Errors

Replication — LS Central SaaS-Specific Errors

The SaaS replication process can sometimes produce errors. Even if a replication job is started from LS Central, it can fail during run in the Data Director. We recommend monitoring the first few runs of new replication jobs using the Configuration tool and job monitor.

Missing Data & Report Issues

If Power BI reports show partial or no data after setup or after pipeline runs, this section covers the most common causes and solutions.

No Data or Missing Data in Power BI Reports

Why this happens
Missing data is almost always caused by a pipeline failure upstream — the Power BI report itself is fine, but data was never loaded into the DW or the semantic model was not refreshed. Secondary causes include: incorrect date filters hiding data; the Finance Report requiring a Chart of Accounts mapping in LS Central before data appears; or transaction data falling outside the DW retention window after a Factory Reset.

For the full missing-data troubleshooting process, see Troubleshooting Data in Analytics.

Sales amounts in Analytics don't match LS Central figures

Discrepancies between Analytics and LS Central figures are usually caused by one of the following:

Pipeline ran mid-day — Analytics captures a point-in-time snapshot. If the pipeline ran overnight and you are comparing to end-of-day LS Central figures, data from that day may be partially loaded.
Company filter — confirm you are comparing the same company in both systems. Multi-company setups can cause confusion.
Date/time zone differences — check how transactions are timestamped in LS Central vs the date filters used in Analytics reports.
Posted vs. unposted transactions — Analytics loads only posted transactions. Unposted entries in LS Central will not appear.
Negative cost values — if the discrepancy is in margin calculations, check for negative CostAmountLCY values in the fact table, which can affect margin figures in Power BI.

To compare amounts between the prestaging and fact tables directly, use this query template (adjust the company prefix, date range, and table GUID for your environment):

-- Prestaging table (adjust company prefix and GUID)
					SELECT [Store No_], COUNT(*),
					SUM([Net Amount]) AS [Net Amount],
					SUM([Vat Amount]) AS [Vat Amount]
					FROM [CompanyPrefix$LSC Trans_ Sales Entry$your-guid-here]
					WHERE [Date] >= '2024-01-01' AND [Date] < '2024-02-01'
					GROUP BY [Store No_]

					-- FactSalesPosted
					SELECT [Posting Date], dL.ReportingStore AS [Store No_],
					SUM([NetSalesAmountLCY]) AS [Net Amount],
					SUM([VATAmountLCY]) AS [Vat Amount]
					FROM DW.FactSalesPosted FSP
					LEFT JOIN DW.dLocation dL ON dL.SK_Location = FSP.SK_Location
					WHERE [Posting Date] >= '2024-01-01' AND [Posting Date] < '2024-02-01'
					GROUP BY ReportingStore, [Posting Date]
				ORDER BY [Posting Date]

If the totals differ, the gap indicates where data was lost in the pipeline — either during staging, the MERGE into fact tables, or due to a filter in the stored procedure.

Database Storage & Growth

Data Warehouse Database Growing Rapidly

Why this happens
When using the legacy Data Director replication method, updated rows in LS Central generate a new staging table row rather than updating the existing one. Old versions accumulate indefinitely and are never purged. This is a known architectural limitation of Data Director replication.

► Recommended action Migration to bc2adls is the recommended long-term path for all customers on Data Director. It permanently eliminates this storage growth issue and the MERGE statement errors in ADF Pipeline Run Failures. For customers remaining on Data Director, upgrading to version 2026.3 or later and enabling the built-in data retention policy provides effective growth control.

Reset Data Warehouse (Factory Reset)

The Factory Reset pipeline in Azure Data Factory recreates the whole data warehouse by dropping all staging tables, truncating dimension and fact tables, updating the query base, and re-fetching all data from LS Central.

► Important Make sure no other pipelines are running while a Factory Reset is in progress. Concurrent pipelines will cause duplicate rows and further errors.

The Factory Reset is recommended in these scenarios:

Duplicate rows in staging tables causing errors in fact table procedures — most likely due to two pipelines running simultaneously.
Data imported with old timestamps — causes gaps in fact tables since they only process staging tables based on the newest timestamp from the previous run.
After extending Analytics — always run a Factory Reset after extensions to reset the query base, staging tables, and dimension/fact tables.

Update Analytics

All releases contain an update package for the previous version, enabling easy updates without a full re-setup.

ⓘ Compatibility The update package is only compatible with the immediately previous version. If you are running an older version, you must update through each intermediate version in sequence.

If there are new staging tables in the release, run the Populate Querybase pipeline in ADF before refreshing data:

Parameter	Value
`PopulateQueryBase`	TRUE — always set to TRUE when running this pipeline.
`GetLSCentralMetadata`	Set to TRUE only if new tables have been added to LS Central that need to be added as staging tables in the Analytics database.

Once Populate Querybase has finished, trigger the Scheduled Run pipeline to refresh the data.

If You Need to Contact Support

We want to help you resolve issues as quickly as possible. Many common issues can be solved using the resources on this page or our AI assistants. If you have worked through the relevant troubleshooting steps and the issue persists, we are here to help.

Self-service resources

Resource	How it helps
This troubleshooting page	Step-by-step solutions for the most common Analytics issues, covering installation, pipelines, replication, Power BI, and data issues.
Ask Analytics	AI assistant covering all Analytics topics. Available on the Analytics for LS Central page — click the Ask Analytics button in the bottom-right corner of that page.
Compass	LS Retail-wide AI assistant for cross-product questions, LS Central configuration, and licensing. Available on every Online Help page — click the icon in the top-right corner.
Analytics Online Help	Complete documentation including deployment guides, configuration guides, and report-specific help.

Submitting a support request

If the issue originates from an LS Retail product and cannot be resolved through the resources above, submit a ticket through the partner support portal:

What to include in your request

Including the right information from the start helps us resolve your issue much faster:

Information	Details
Issue description	What happened, and what you expected to happen instead.
Steps to reproduce	Numbered steps that reliably reproduce the error.
Screenshots and error messages	Full-screen screenshots and the complete error text — this is often the most helpful piece of information.
Analytics version	The exact Analytics version number. Found in the setup wizard or the Analytics database version table.
LS Central version	The exact LS Central version running in the customer environment.
Environment topology	SaaS or on-premise? Replication method: Data Director or bc2adls? Hybrid setup?
ADF pipeline logs	For pipeline errors: the failed pipeline run details from ADF Monitor, including activity-level error output.
SQL error details	For database errors: the full SQL error message including error number, severity level, and stored procedure name.

ⓘ Tip The more complete your initial submission is, the faster we can help. Providing all the information above means we can start investigating right away.