Winning companies use data to their strategic advantage throughout the organization. However, for a company to truly become data-driven, they cannot rely on departmental deployments of an analytics platform like Qlik – they need a solution that spans the enterprise.
Analytics8 Managing Consultant, James Carpenter, provides his tips and strategies to move from a departmental to enterprise-wide implementation of Qlik.
Think about climbing a mountain. Before you start climbing into higher elevations with less oxygen and colder temperatures, you must get acclimated. Climbers use a “basecamp” to do this. We also suggest our clients go through a basecamp process to build a base comfort level using Qlik across the enterprise. When moving from a departmental to enterprise implementation, you will have competing interests, because some aspects of the implementation are quicker to implement than others. One department may receive a high-value quick win that builds the foundation for future development, while another department must wait for something more complex like an executive dashboard which requires integration of many data sources. The Basecamp provides an opportunity to assess the business reasons behind using Qlik and to prioritize and organize each departments’ needs and projects.
Here are some elements we like to address with our customers during strategy-building:
The purpose of starting with strategy building is to promote adoption of the tool you are investing in. By addressing these aspects of an implementation up front, there will be a shared vision of what Qlik will offer and be used for.
When implementing a data and analytics solution, you want to align your Business Intelligence (BI) objectives to your corporate objectives. For example, a common BI objective is to create a data solution that is the “single source of truth” so that all decision makers in a company are looking at consistent and accurate data. A corporate objective may be to increase sales by 10%. One way to accomplish both could be to make sales opportunities more visible so decision makers throughout the company can analyze sales effectiveness.
Once you’ve matched BI objectives to corporate objectives and created a list of desired behaviors of your solution, you then assess those projects from 2 perspectives:
With our clients, we often use a tool called a BI Prioritization Matrix which helps them compile and prioritize key BI initiatives to determine the technical feasibility and real business value of each item and how they correlate with overarching corporate objectives.
The BI Prioritization Matrix helps you create a roadmap for your priorities and recognize things you can readily implement.
Once you have your plan in place, you now want to implement your BI platform in a way that is scaled and designed for enterprise use. One of the challenges when working with Qlik is defining a centralized library for master items that can be shared among multiple applications. One method that helps with this is Governed Metrics Service, an extension to the Qlik Sense server environment. It allows you to define your metrics externally, load them into Qlik Sense, and apply them to any/all applications that you choose. This enforces consistency (remember the “single source of truth?”) and reduces administration because you don’t have to replicate complicated master items across applications.
The architecture for this product has metrics stored externally (Excel spreadsheet, a database table, etc); a .QVD is built from this; and the Governed Metrics Power Tool is installed on the server, and that’s the interface you use to apply the metrics library across applications.
Governed Metrics Service
If you build it, then they will come… but can your hardware handle it? Scalability testing needs to be done early in the process to answer uncertainties about more memory or faster processors. The Scalability Toolkit helps you perform regression failover testing with automated testing. The toolkit allows you to set up a hypothetical scenario of typical and atypical usage of the application, and then executes the scenario to produce metrics and outputs for you to analyze. For example, you can assess how increased users affect RAM usage and response times. It’s important to test performance, expose weak points, and make appropriate adjustments before users interface with the system.
You can’t simply rely on data from one department to get the full picture of what’s happening at your company #QlikEnterprise
Qlik Sense comes with the Qlik management console which works well, but there is an opportunity for extending functionality with QMC Utilities, an open source project that assists with Qlik administrations. Here’s a sampling of QMC features that ease the administration of Qlik:
In4BI, available for Qlik Sense and QlikView, is a tool that allows developers to work in the same document in parallel with a formalized checkout process and some added governance functionality. Administrators act as gatekeepers and control the promotion of applications between environments, ensuring that QA has been met and checklists have been completed. Additionally, version numbers are automatically assigned to eliminate confusion among developers.
It’s highly possible that part of your system is already in the cloud – like one of your Qlik components or other source system data (CRM data in Salesforce.com for example) that you upload into your system. Running Qlik in the cloud should be a strong consideration as you move from a departmental to enterprise solution. Hosting components on the Qlik server in the cloud gives you several benefits:
It can be a challenge to deploy Qlik to the enterprise, but these are proven strategies and tools that have helped our customers succeed.
We cover these strategies in more detail in our Qlik Enterprise Webinar. VIEW NOW
To thrive with your data, your people, processes, and technology must all be data-focused. This may sound daunting, but we can help you get there. Sign up to meet with one of our analytics experts who will review your data struggles and help map out steps to achieve data-driven decision making.