Qlik Deployment Tips

Congratulations!

You and your company have chosen Qlik as your analytics and business intelligence solution. That was a really good choice.

To reach this decision, you probably reviewed different tools, evaluated their features, read articles, watched tons of videos, and attended several webinars. Maybe you tried it yourself and thought “Yeah, it’s very easy to use and it looks great. My company will love it”. Before the final purchase, the pre-sales team might have developed a proof of concept using sample data. The “wow effect” that occurred in their final presentation got you and your boss really excited. “We will turn our company into a real data driven organization with Qlik. Let’s rock”.

Those early days are fun because all you think about is the potential impact of the platform and all the benefits that the business will get. Now, you have installed the server, activated the license key, connected your user directory and reality is hitting you: What do we do next?

Whether you have a brand new installation or your company has reached a point where it needs a fresh take on analytics, this post will give you some useful tips on how to improve your deployments and create a better BI environment.

Do you have a plan?

Wanting to become a data driven company and actually making it happen are two very different things. As my friend Dalton Ruer says: “It’s not about the software installation, it’s about the cultural transformation”.

In other words, it’s not about technology or your ability to develop works of art with it. It’s all about the human part in your organization and their desire and ability to consume what you build. I’m not trying to be overly dramatic, but your main challenge will usually be user adoption. If nobody uses your apps, how long will your pride in deploying the amazing dashboards last?

From an IT perspective, it’s great to have a modern analytics platform with robust architecture, data governance and plenty of applications available anytime, anywhere. Everyone is watching at the RAM and CPU usage, the license allocation, and the number of sessions. Yeah for DEVOPS. However, this is a technology-centric approach to a very human problem. Despite all these measurements, you don’t have a clear understanding of how successful (or not) your Qlik deployment is for the end users!

That brings me back to my point: when the technology part of the equation is covered, how will you take your organization to the next level?

Have a clear understanding of your targeted audience

You will face different challenges depending on the type of industry and the size of your company. In some cases, your user base will be composed of technology-friendly people and you’ll just have to make sure you meet their business needs. In other situations, you will face subject matter experts that are less engaged with this type of tools. And what about the business leaders? My experience tells me that, frequently, you will have to convince decision makers to spend some of their precious time learning how to use your apps instead of asking others to give them the answers right away.

What’s the data literacy level within your organization? 

Knowing your target audience is as important in data analytics as it is in marketing. In a way, deploying dashboards is like launching a marketing campaign.

Here are few questions that you will need to answer:

• Which are the tools people use on a daily basis?
• How much time do they spend analyzing data?
• Does corporate messaging include data visualizations?
• Do they compare their figures against a benchmark (budget, last year, etc.)?
• Do they have to work with (or against) the data to get answers?
• Do they trust the numbers they see in corporate reports or do they trust the person who manufactures them?

In this regard, your main goal is to ensure that everyone has the skills to understand, use and collaborate with data. Like any new ability, some will pick it up faster than others and a few of them will truly struggle.

What’s the best strategy to achieve your goals?

Well, it depends. The appropriate strategy is always the one that fits your organization and its culture. I will highlight a couple of strategies that I’ve seen at various organizations in different industries, sizes, and maturity levels. Pick the one that feels closest to you and plan around it.

“All-In” poker strategy

This is an “all or nothing” tactic. As the name implies, it is a high-risk, high-reward strategy that is commonly used on projects at the very heart of the business and involve a lot of people across the organization. It usually requires considerable upfront investments to hire experienced consultants or to buy prepackaged solutions because failure is not an option.

Some examples include:

• HR dashboards that will be used by all the managers in the company
• Sales performance applications that cover the whole commercial team (from top management to sales representatives)
• Integral ERP analysis including KPIs from multiple departments

If you choose this path, keep a long-term approach in mind. Remember to start simple and improve while your colleagues get more engaged and excited about it. Don’t wait to build a complete solution for every business case and for every team. Analytics is an iterative process. Embrace it as such.

High performing teams often take an agile development approach and pair the developers with the business units. You can’t expect a robust requirement document from people who have never actually seen their data. One of the best things about Qlik is that it’s data modeling and front-end design are lightning fast. Take advantage of that and conduct rapid BI prototyping sessions where developers access different data sources and visualize the data to allow the business users to determine which pieces will help them answer their questions. The look-and-feel can always be tweaked but knowing that data actually supports the business is vital to success.

It’s also important to forget about the “completion percentage” type of reporting that IT loves so much. You will be constantly making strides towards better analytics and helping users work more efficiently.

Just think back to those lengthy projects you worked on where business users didn’t get involved. We tell ourselves things like “the requirements were poorly translated in the document” or “the business leader was flaky”, but the truth is we were trying to force business users into our ancient IT process that measure activity instead of results.

Once you have deployed your first apps, it time to capitalize on your success and set expectations a little higher for the following projects. Keep an agile mindset to respond to business needs in a timely manner.

In the end, one of your greatest challenges will be to develop skills internally, hone your processes and build a team that will keep assisting and producing high quality projects that always involve the business users.

Start small. Get big.

In this scenario, you might be in a small team trying to complete a project for your department. You don’t want to transform everyone’s life with your Qlik app just yet. You just want a quick but memorable win. The main goal here is to demonstrate your ability to develop and deploy an app that brings high value to end users.

The same iterative, rapid BI prototyping process employed for the “All-in” scenario is also well suited for this environment. Our main goal remains the same: ensure that the platform allow business users to answer all their questions. In this situation, it is vital to engage with users because you probably won’t have the same level of executive support that forces people to cooperate with you.

Ask yourself these questions:

• Do people struggle with tons of flat files sent by email and shared across the company?
• Do they have static reports but constantly ask IT to make changes?
• Do people extract information manually?
• Do they spend more time producing reports than analyzing and making decisions?
• How can they become more efficient in their daily activities?

This situation exists in every company around the globe. You just need to find the right person to support and sponsor the project.

After your first win, chances are that some stakeholders start asking for a larger deployment. Be sure to focus on the business impact and involve your users. It’s all about marketing. People will be curious, perhaps even jealous. Some of your business users will become real champions of the data-driven mindset.

Once again, just market the vision and allow everyone to become a part of it. The three biggest keys for you are quality, reliability, and trust.

Empowering your teams

You’ve done a great job. So good, that you’ve worked your way out of it and into a different role. Now you are working in a central department and your role is to support business lines by providing training and improving their tools.

Basically, your goal in this new “center of excellence” will be to make life easier for developers and data leaders. You will also ensure that everyone is following best practices and complies to your governance schema, so your environment performs well. Your job is to make every project a success but not by developing it directly.

These are some tips that you might find useful:

• Start by finding users that know the business and can handle technology in order to help them develop their skills. Remember that you didn’t get to your position overnight and that they won’t be as comfortable as you with certain topics.
• Build a community. Maintain and collaborate with your user base so they know there’s a data culture surrounding them.
• Keep it fresh. Never hesitate to share new or creative content about Qlik, Data Literacy or Data Visualization techniques.
• Your focus on data quality and defining adequate business rules is more important than ever before.

Empowering the business with the right skillset in a governed environment is, in my opinion, the best way to make the most out of Qlik.

Final thoughts

These strategies are drawn from my personal experience and, obviously, there is so much more to say about them. As you can imagine, the last one is my favorite as it involves the organization as a whole and breaks the barrier between the developers and the business.

Thanks a lot for reading! I’d be really happy to go deeper on this topic or talk about your own experiences, so don’t hesitate to comment or to reach out to me on LinkedIn.

Christophe, with the invaluable help of Dalton Ruer.