Smarter Government, GPRA and questions about the Capitals
Delaney Turner 270003RQ8K Delaney.Turner@ca.ibm.com | | Tags:  ibmsoftware business_analytics
0 Comments | 1,394 Visits
What’s the tougher question for the Washington government official these days:
Luckily, in the former at least, there is a ready answer.
Smarter Government was the theme of our 8th annual IBM Government Forum in Washington, DC and in the elegant meeting rooms of the gleaming Ronald Reagan Center nearly 400 representative from more than 100 federal agencies convened to learn how smarter government solutions - powered by IBM Business Analytics - could help them improve outcomes and optimize results.
Lee Chichester, IBM Business Unit Executive for U.S. Federal Business Analytics, hosted the event and guided attendees through a fully stocked agenda of keynote speeches, breakout presentations and panel discussions. My colleague Blythe Howard-Chou provided a super summary of the day’s key learnings. Below, I’ve summarized two of the day’s sessions in more detail.
First on the agenda was Gerard Mooney, IBM General Manager for Public Sector, Global Government and Education, who through a combination of IBM research and customer success stories, explained the three core tenets of Smarter Government. Here are his slides from the session:
Achieving Better Outcomes with Data Driven Decision-Making
Jimmy Wales: Five Wiki Principles
Mooney ceded the stage to Wikipedia founder Jimmy "Jimbo" Wales, whose free, open, collaborative and co-operative Web encyclopedia (now within the top 10 sites on the web in many countries) embodies many of the attributes government agencies themselves are working hard to achieve in the real world.
Wales attributed Wikipedia’s success and continued growth to five core principles espoused and upheld by what is now a global community of thousands of contributors around the world. Those principles are:
Panel discussion: Setting goals and measures
Another highlight was the panel discussion entitled “Identifying Goals and Building Measures” featuring:
Here’s a transcript of the highlights:
Jon Desenberg: We're all on a journey of adopting analytics. What hints or tips to you have to get organizations to adopt this management practice?
Greg Greben: In the past, when we went into an organization and talked with leaders about analytics, our bias was that the barriers were technical and financial – that there was a high cost of entry, that you had to build the infrastructure and build the skills. What we've learned through our surveys and our more recent experience is that the challenges are more organizational in nature. It's the same in government – obviously leadership competencies are very important, but of the dozen or so barriers we see, five of the top six are organizational -the ability to have adequate leadership attention around an information management and analytics campaign, to secure top talent, to build a culture of collaboration – these are key areas to address early on.
Jon Desnberg: The performance culture within government has often been viewed as a “gotcha” culture – there's no upside, only downside. Is that changing? How do we get away from the “gotcha”?
Jonathan Bruel: Well, the “gotcha” culture is a natural one in government because of the focus on accountability and concerns about misspending. And frankly, because we have an army of people in government doing that kind of work. The last count I saw was 13,000 people in IG offices. That's a lot of people looking for mistakes. So there's a ready-made machine to go in that direction.
But there is a change under way and there are a few indications of that – people are much less interested in fault finding in their use of measures. They're much more interested in evidence of progress improvement. They've been very explicit in saying that it's not a problem if they come a cross an agency that fails to meet its goals. It's a problem if they’re not aware that they have a problem or if they don't have a plan to fix it.
The second element is the enactment of the GPRA modernization Act. That was in important milestone because Congress basically recommitted to the entire issue of performance. They essentially pushed the reset button to say “let's take the experience, the measures and the goals of the last 18 years and put them to use.” It was an endorsement of what had happened in the past but a change that was much more positive.
Jon Desenberg: I don't think a lot of people have seen the analytical culture at work in private sector organizations. Greg, how is it used in this culture? Obviously there's a real emphasis on results, but how do they keep people from setting the bar too low?
Greg Greben: Setting the bar low is a trap because it won't allow you the types of organizational commitment and empowerment you need to address the big challenges. In the private sector I see organizations taking on the challenges that directly affect their ability to survive. This involves heavy senior leadership and it does involve financial commitment and technology to overcome the barriers we discussed. But taking on the biggest challenges doesn't mean taking on the biggest risk if you can rally your organization around solving them. Look at the industries that have gone through mergers and acquisitions – you have significant back office, infrastructure, data and analytics challenges to solve. You can't solve them in piece parts. You need to take them on as large-scale transformations.
Jon Desenberg: On that same point, we get a lot of questions about target setting. Specifically, stretch targets. Is it always appropriate to set stretch targets? What kind of culture do we need to hit them?
Jonathan Bruel: Stretch targets are necessary because in some cases they’re the only way you're going to get significant program improvements and some countries do some interesting reporting. The Bush administration had its scorecard with red, yellow and green reports. But if you go to Singapore, for example, they also use this method but have a category above in blue for stretch goals. You get extra credit for meeting these goals but you're not penalized if you fall short.
That's something we need to adopt here. It's a real culture change because if you look at the performance plans of some of the agencies, they're actually setting their internal targets at 80 to 85 percent of what they achieved before. There's a term - “timid targets” - to describe where the goal is set at a very safe level. But that kind of result is not going to impress the organizational leadership or lead to any kind of significant improvement.
Jon Desenberg: You've both touched on leadership. What are the competencies needed for an analytically oriented future?
Greg Greben: Adopting analytics is a significant undertaking. So you want leaders with ambition, who aren't afraid to challenge the status quo, who aren't comfortable in ticking off low-hanging fruit and setting targets that are easy to achieve. You need to look for people who think bigger than that. We're also seeing leaders who understand that it's inadequate to make key decisions on intuition alone. They want the data. They want the advanced analytics that help them understand the tradeoffs and alternatives that exist. They make that a key part of what they do.Another element is empowerment – not just for themselves, but for the people on their team. They empower their people to make decisions with the information they have. Empowerment isn't just something that pockets of people do. We're looking for a more pervasive set of analytical capabilities.
Jonathan Bruel: Look at the GPRA. You've got an explicit assignment for deputy secretaries to act as chief operating officers. That’s a set of competencies that has not been recognized or attended to as much as in the past. But it's not for the faint of heart. It's a huge set of responsibilities and it requires someone with the experience and appetite to manage a large public organization. It's not something that can be delegated away. The deputies really need to give their full attention to this assignment. The second aspect of the Act is the performance improvement officer, which calls for someone with much more technical and bureaucratic skills to operate those activities. You also have goal leaders, who need to possess extraordinary subject matter expertise. At the analytics level, you need people with the skills we don't recruit sufficiently – mathematicians, operations researchers – people who can really command the numbers and produce the insights that are called for here.The data alone won't give you the answers. You need to probe and make connections to understand the causalities. Again, that's a competency that's not exhibited enough in government these days. We need to go after this with much more imagination and energy.
Jon Desneberg: It's common to hear “there are too many measures.” Currently in the US there are 126 high-priority government-wide metrics. The GPRA aims to have 100. The UK only uses 30 and other studies show that you shouldn't use more than 12. What's the right number?
Jonathan Bruel: The quick answer is that there's no right answer. You're right that under Tony Blair, the UK focused on 30. But the UK is smaller and it's a parliamentary government that delivers services much more directly than here. So there are a lot of differences. If you look at the various administrations over the last 20 years, it looks like the fewer goals, the better. Vice-president Gore's program had 1,600 different goals; under president Bush we had five.
What's most important is to align your metrics with the leadership you're serving. If it's Secretary [Ken] Salazar going out with a new strategic plan for the Interior, the number of goals and the goals themselves need to align with his understanding of the problem. You need to meet his numbers. I don't think the precise number of metrics is as important has having a set that the leadership understands and is comfortable owning. The goals have to make sense to them and work for them if they're going to work at all. Otherwise it’s just a paper exercise that gets discarded.
Jon Desenberg: I think I’d be remiss if i didn’t mention predictive analytics – looking into the future is probably the toughest thing that we’re trying to do. What are the lessons learned in taking analytics from looking backwards to looking forwards?
Jonathan Bruel: Predictive analytics become relevant, if not absolutely necessary, when you get into the budgetary and fiscal problems we face. Anyone who can tackle the kind of policy and program problems that we’re facing with any kind predictive insights will be invaluable. Agencies are going to be incredibly challenged. The ability to come to the table with predictive analytics that can help people understand what their caseloads or operating environment is going to be like is going to be an invaluable asset to their cabinet secretary, leadership and committees on the hill. Otherwise we’re going to continue to fly blind with budgets that are just crude adjustments to the funding policies that we had in the past. We’re going to need much more insight and imagination to solve these problems.