I’ve spent most of my career working in government performance and innovation – but I’ve found we don’t all mean the same thing when we say “innovation.” For many public leaders, innovation means moving things online, going paperless, turning government into an app.

On the one hand, I get it: We’ve all heard the horror stories of maintaining COBOL-based systems from the 70’s. We do have an imperative to make government services more convenient for citizens.

And yet: When I run process improvement training for government leaders, I always ask them: “How many of you have been a part of a big IT implementation?” Every hand goes up.

And then I ask: “And how many of those IT implementations solved your problems?” Every hand goes back down.

The truth is, there’s no such thing as an IT problem. IT is a potential solution.

The trouble is, few government leaders take the time to understand the problems in their underlying business process before implementing IT. Applying a solution without understanding what problem you’re trying to solve is a surefire way to make that problem worse.

In my first role out of grad school, I provided support to a medical clinic who made a transition to digital radiology – that is, moving x-rays from physical film to digital. This was a carefully planned months-long transition involving new equipment, new software, and new integrations into the electronic health record.

About a week after the launch, there was a crisis in radiology: The new system had no way to accommodate “re-reads.” A re-read, it was explained to me, was needed when there was a miscommunication between the clinician and the radiologist. The doctor wants to know if the patient’s finger is broken and orders an x-ray of the hand, and the radiologist writes a report that says that the patient’s wrist is fine.

In the old system, the doctor took the physical film, put a sticky note on it, and sent it back downstairs to radiology for a re-read. The new system did not have a way to digitally “return” the radiology report – so clinic leadership were left scrambling, working urgently with the vendor to customize the software to create this digital equivalent of the old sticky note.

But in all this, no one asked the fundamental question: Why doesn’t the radiologist understand what the clinician needs from the x-ray? The clinic was so focused on eliminating physical film that they’d failed to understand what it would mean to have a well-functioning radiology department. As a result, instead of fixing the error, we were paying an IT vendor to accommodate the error in our system. Permanently.

In reality, IT can’t fix a bad process. More often, IT makes a bad process worse, hard-coding your convoluted steps in a way that’s time-consuming and expensive to change. Slapping a pretty app on top of your complaint management system without actually managing complaints better doesn’t resolve your problem; it just hides it. Not to mention the challenge of completing a full digital conversion while adequately addressing equity concerns: it’s much easier to translate a paper form than a website or an app, and many of our vulnerable populations may be slower to adopt online options.

The irony is that buying IT is expensive, while cleaning up our business process is relatively cheap. Yet our governments are willing to throw millions of dollars at major IT implementations, while allocating little or nothing for basic problem-solving and process streamlining.

Of course, there is a place for technology investment. (Seriously, those COBOL-based systems need to go.) But to make a good IT investment, you must understand clearly what problems you’re trying to solve, why those problems are happening, and whether the root causes of those problems can be best addressed by IT or something else.

A couple years ago, I worked with a county that wanted to upgrade the system its clerks used to manage agendas for the Board of Supervisors. But instead of jumping straight to IT, this County asked key questions first about the business process: What does it look like to have an excellent agenda management process? How are we performing now, and where do we want to be? What obstacles are preventing us today from being where we want to be?

We spent several months working with their managers and front-line staff to map processes, identify root causes, and brainstorm solutions. Two key results from that work:

• Board minutes were being published in 6-8 weeks. We identified how to publish them in one day, with less overall effort by Clerk staff and no IT changes. Delays weren’t about their software, but rather how the clerk staff balanced and prioritized their workload.
• We eliminated 85% of the paper being used at Board meetings and hours of effort printing, collating, and mailing agenda packets, again with no IT change. It turned out that Supervisors and their staff already had access to digital packets, but the old paper process had never been phased out.

If the county had immediately implemented a new system, they would not have eliminated paper or sped minute publication, because they were not IT problems.

At the same time, we also illuminated the issues that did lend themselves to an IT solution – for example, creating a transparent way for a department to know where its agenda item was in the approval process workflow. We were able to craft a strong set of IT requirements that drove the eventual vendor selection and resulted in a new system that actually helped them work more efficiently.

Here’s best practices for government leaders to move into the future intelligently:

• Don’t mistake innovation with IT. Innovation is not synonymous with “buy new IT.” Innovation is about applying creative problem solving to public problems.
• Invest in process improvement before investing in software. Take 10-20% of your IT budget and invest it in process streamlining before procuring software. You’ll find that half of your problems can be solved before implementation, and you’ll more than make your money back by better tailoring your IT product to what you actually need.
• When they’re necessary, build IT solicitations with a clear understanding of the problems you want it to solve.  “Our system uses a DOS prompt” is not good enough. What are the obstacles to your staff’s daily work or to providing quality services to your customers? Let those obstacles drive the requirements that you set in your procurement. IT vendors are fond of calling their products “solutions.” Make them answer the question, a solution to what?

Your citizens don’t want another app on their phone or a web portal. They want a government that’s responsive to them. Let’s figure out how to give them that.

Ryan Hunter is co-founder of Partners in Public Innovation, whose mission is to transform public services by empowering staff to create a culture of continuous improvement. Before founding PPI, he directed San Francisco’s citywide program in lean process improvement. He has trained over 1,000 public servants to effect change in their daily work. He lives in Sacramento, California.