Typical organizations suffer from a collection of problems with their sales analytics initiatives. These include:
SSAS:SQL Server Analysis Services (SSAS) is a commonly used tool for sales analytics because it’s able to create multidimensional data structures, aka “cubes,” from various data sources. It’s challenging to use for many reasons, including performance, the complexity of cube design, and useability with a wide variety of BI and analytics tools.
Tool Space Fragmentation: Over time, it’s common to have a mix of legacy BI tools, relational databases, and open-source big data tooling. This mix causes a range of problems, including inconsistent features/functionality, different levels of expertise across different analysis, and data that’s tied to one tool or another and can’t be used elsewhere.
Multiple Data Sources: Organizations end up with multiple data warehouse and data lake systems. These store millions of records and are often filled with duplicate data, incomplete data, or outdated data.
Fragmented Identity and Access Management: Multiple tools and data sources result in a patchwork of access control. Employees have to submit tickets to do their jobs, causing delays and frustration.
Data Volume Growth: As organizations grow, their datasets grow as well. It’s not unusual for a large retailer to experience over 100% YoY data growth. Not only is this growth hard to manage, data sets become too large to analyze in time to drive business initiatives unless corners are cut and sales analytics are restricted to easy problems.
Infrastructure Provisioning: On-premise infrastructure takes too long to procure, deploy, and setup, delaying any improvements in sales analytics an organization wants to make.
These challenges end up resulting in two fundamental problems.
- Analysis becomes less and less granular over time. It becomes harder and harder to conduct deep analysis, add additional questions, or even find the data you need.
- Analysis can’t be done quickly enough. The organization needs velocity, and sales analytics has to be agile. For example, analytics have to be quick enough to signal, in minutes, when a feature deploy on their website results in lower conversion rates.