Saturday, December 14, 2013

Dresner’s Point: Does BI Need a Seducing Effect?

I guess Black Friday, January sales and constant reminders of how many shopping days are left before Christmas weren’t enough enough to entice shoppers to hit the stores early and often enough in the holiday season. Now we even have Pre-Black Friday and Green Monday. These seductive tactics to entice people with something “special” reminded me of a discussion we had at one my Friday #BIWisdom tweetchats.

When I asked my Twitter group of BI enthusiasts what would help to rein in more users, it was apparent that they thought more self-service functionality is needed in BI apps. They agreed that vendors have made great strides in self-service BI (SSBI) but also agreed that vendors need to extend SSBI to advanced analytics.
The weekly #BIWisdom tweetchat is my favorite forum for BI discussions because it brings out varying perspectives among users, vendors and consultants and thus usually bubbles up effective real-world ideas. In the following tweets you can see how they painted decision making for improving the SSBI environment:
Should we “dummy down” advanced analytics for end users?

• No matter how good the information is, if a tool is too complex for end users, they won’t use it. They’ll take the path of least resistance. The more BI practitioners hide the complexity, the more users will think it’s easy.
• Complexity can be hidden behind well-designed semantic layers.
• Users want things simple and want what they’re accustomed to daily: Twitter, Google, Apple. And Twitter encourages over-simplification.

Should we train and educate the users better?

• Some tools require near-zero training to use. Pain comes from not understanding the data.
• Assuring data understanding is tough. Even with better visualization, it’s still a difficult problem to solve.
• I just saw a 3D map that could be rotated in a BI app. But does anyone really need that?
• Should it be a zero training approach or contextual help?
• Training in the abstract is tough, but training with live data is close to zero.
• Users have varying needs. They will care about data and information if it’s relevant to their roles. Training is easier if it’s personalized to their role.

Two of the group tweeted a viewpoint that buttonholed the real issue — “It’s funny how BI is becoming part of consumer apps and the app world is showing the IT world that the focus needs to be on UX / IX design; yet BI vendors struggle with adoption.”

The #BIWisdom group agreed this is a critical distinction, especially since a large part of the population is being raised, or switching to, a ‘mobile-first’ approach. The remaining discussion centered on the vendors’ responsibility for SSBI and BI adoption. As you can see, their opinions varied:

• Tools need to be simple and fast, but it’s not all the responsibility of the vendors.
• BI users aren’t always the right people to ask about UX design. Take me, for example: Give me a big table of numbers, and I’m set.
• Letting analysts design the UI results in a poor design.
• A BICC (BI Center of Competency) should include an expert in UX / UI design.

As the tweetchat wound down, they agreed on two aspects.

First, vendors that don’t focus on UX / IX design and self-service functionality risk becoming irrelevant. “That’s the price of entry,” someone commented. Most agreed that vendors aren’t focusing on design and SSBI in their sales pitch.

Second, you can have the best data or information in the world, but “pretty” matters because pretty and easy beat complete and complicated every day of the week. “Develop an ugly dashboard and see what happens to usage,” tweeted one participant. Someone else added, “Even the placement and color of KPIs impacts usability.”

Bottom line: I recognize the validity of all their comments. But I also recognize that problems are often perception more than reality. I believe the issue around self-service BI is an education issue first and a technology issue second.

I believe organizations need more education (and change management), especially focusing on meaningful BI experiences. And I like the community-based approach to developing solutions and UX / IX design; this can make for more durable and more value-producing software. What could be more seductive than higher value and return on investment?



Click Here to Purchase Your Copy of the 2013 Wisdom of Crowds ® Mobile Computing & Mobile Business Intelligence Market Study

Howard Dresner is president, founder and chief research officer at Dresner Advisory Services, LLC, an independent advisory firm. He is one of the foremost thought leaders in Business Intelligence and Performance Management, having coined the term “Business Intelligence” in 1989. He has published two books on the subject, The Performance Management Revolution — Business Results through Insight and Action, and Profiles in Performance — Business Intelligence Journeys and the Roadmap for Change. He hosts a weekly tweet chat (#BIWisdom) on Twitter each Friday. Prior to Dresner Advisory Services, Howard served as chief strategy officer at Hyperion Solutions and was a research fellow at Gartner, where he led its Business Intelligence research practice for 13 years.

Dresner’s Point: How to Move the Needle in User Adoption

The topic of how to increase user adoption of business intelligence technologies generated a lot of opinions in a recent Friday #BIWisdom tweetchat.

The tweets started with opinions about what hampers user adoption. Some tweeted that the tangling of BI into the Big Data space causes confusion, it’s a mistake to try to get end users to adopt “analyst tools” and we must stop force-feeding generic analysis tools to end users even if they are “simple” to use. Someone tweeted that some organizations make the mistake of working only with power users who may have a vested interest in not expanding adoption to others. Many agreed that trying to change a user’s adoption of and interaction with BI tools on a daily basis is difficult; one tweet compared it to trying to change someone’s religion or belief system.

The conversation turned when a #BIWisdom tribe member stated that “BI encompasses people, process and technology; but most implementations focus on the technology and forget the people and process.” Others then chimed in with:

• BI should just be intuitive and just a part of other apps; users don’t want to think about doing BI
• Predictive analytics adds a lot of value; but it needs to be a part of users’ current desktop app, not a separate app
• Analytics should be placed inside business processes and link to actions rather than making users go to another system
• BI solutions should not be approached from a one-size-fits-all mindset

A participant tweeted that “Vendors need to push toward true self-service functionalities and super-easy user interfaces; self-service lies not in making generic tools simpler to use but in custom interfaces for specific tasks.” This tweet sparked comments comparing BI apps to mobile apps, as apps on smart phones and apps are very task oriented, enticing and focus on the end user. Mobile apps have the characteristics that “good BI” should have: agile, visual, interactive and intuitive. Unfortunately this seldom occurs.
Some tweeted that perhaps we should apply the mobile app model to business intelligence. Another participant added that “Every hour of training needed for BI technology cuts user adoption by at least 50 percent — compare that to the quick adoption of iPads and other tablets, which need no manual or training course.”

In the end, the group concluded that there is not enough focus on incentives. They agreed that bad analysis often goes unpunished and good analysis often goes unrewarded. “People who use BI technology well should be made into organizational heroes in internal communications and should be rewarded,” tweeted a tribe member. Organizations should praise people who demonstrate transparency and accountability as both are essential in an effective BI environment.

Along that line, someone added that organizations should also “identify employees who are against the necessary change management necessary to build a corporate culture that embraces fact-based analysis and accountability and then communicate to them the value/benefits of leveraging BI. That aligns with my own observation over the years that, without top-down strategic change management, an organization will not widely adopt BI.

Bottom line: A lot has changed, and the way we use technology today is vastly different from 20 years ago or even a decade ago. But one thing has not changed: technology adoption must be driven top down. If the CEO and other CxOs don’t believe in the technology, nobody else will either. My observance is that CxOs must embrace the technology; when they believe it is critical to their personal success, everyone else will get on board too. As soon as there is a hint that maybe the technology is not strategically important, people will treat it with denial and neglect.

There are some things that can be done to “move the needle” in BI user adoption, but nothing is as impactful as getting the CEO on board.



Click Here to Purchase Your Copy of the 2013 Wisdom of Crowds ® Mobile Computing & Mobile Business Intelligence Market Study

Howard Dresner is president, founder and chief research officer at Dresner Advisory Services, LLC, an independent advisory firm. He is one of the foremost thought leaders in Business Intelligence and Performance Management, having coined the term “Business Intelligence” in 1989. He has published two books on the subject, The Performance Management Revolution — Business Results through Insight and Action, and Profiles in Performance — Business Intelligence Journeys and the Roadmap for Change. He hosts a weekly tweet chat (#BIWisdom) on Twitter each Friday. Prior to Dresner Advisory Services, Howard served as chief strategy officer at Hyperion Solutions and was a research fellow at Gartner, where he led its Business Intelligence research practice for 13 years.