What You Touch You Don’t Feel

I am fascinated with the term “creative destruction.” Joseph Schumpeter, the economist associated the concept, described it as the way industrial progress destroys the processes that preceded it. But to me, it sums up my approach to project management, for better or worse.

On the worse end of things, I have a tendency to start projects (such as blogs or websites) with a lot of enthusiasm, then tear them down as soon as I am dissatisfied with the direction they are going. But on the better end of things, pulling something apart is the only way I can fully understand how it works. And sometimes, by breaking up the old systems, I can build something more efficient and thoughtful in its place.

When I began to manage my office’s data collection a few years ago, I took a hard look at the software we were using in the collection process. It became very clear that there were a few problems with not only the software as we had set it up, but the process itself. I decided the best way to move forward was to stop using the software and restart the process from scratch.

Obviously, I had to work hard to convince my office’s leadership that this was the right approach, especially because I was creating a massive workload for myself. But I thought that by ditching the old process, we could get a better sense of how our posts collect and manage their data, and we could rebuild from there.

My first step was to ask posts to email me their data in the format that they used themselves. This turned out to be a really interesting endeavor. While we were dictating what we wanted to collect, most posts either already had their own processes in place or had taken their cues from our original requests.

Using all of these reports, I built a stop-gap process involving Google Sheets. It was built around each fiscal year and organized by country-specific tabs, divided by space, and broken down by month. For various reasons, this stop-gap turned out to be our long-term method of data collection. It’s not ideal, because a series of spreadsheets doesn’t really equal a database, but through creative use of Google Sheets formulas like IMPORTRANGE, I built something that was a lot more useful than the elaborate software system that it replaced.

Yes, I am bragging. Just a little bit.

The other aspect to our data collection process that I changed was a bit more simple. We used to grumble in our office that our data was frequently incomplete. When we did training sessions or meetings with staff, we would stress the importance of submitting their reports. I remember one session in which a library director asked if we could set up reminders to input their data. Our response was to remind staff to set up their own reminders. This seemed wholly unsatisfactory, even at the time.

It occurred to me that we should be requesting the data we want from everyone instead of expecting everyone to just send it to us. So after that initial data call, I just started to email everyone reminders on a regular basis. And pretty soon we ended up with as complete a dataset as we could expect. Sometimes if you just ask nicely, you can get what you want.

Yes, I am bragging again. I am proud that I was able to take my destructive nature and turn it into something constructive and ultimately pretty useful. That said, it is time to move into something a bit more sophisticated, because I manage a lot of spreadsheets now. But I may still send out those reminders personally. I kind of like having that connection with everyone from around the world.

But Researchers Never Found All the Pieces Yet

I have spent the last couple of years at my job wrangling data: visitor numbers, electronic resource fulltext downloads, activity reports, things like that. All of the numbers and reports and the like are stashed in Google Sheets and Google Docs, in SharePoint and in OneNote, and they are shared constantly through email and Slack. And we can put them all together to tell compelling stories about the work we do.

But I know that this isn’t quite enough to really capture the whole story. There are better data points and better ways to collect and manage those data points and better ways to evaluate them.

(I am giving some of my spreadsheets the short shrift when I put it that way. They are beautiful spreadsheets and like Big Daddy Kane they get the job done.)

I’ve thought a lot about how we can improve our data collection and management over the past couple of years. I’m not thinking about little tweaks here: I’ve done those little tweaks already. I want to make significant changes, and I’ve put a lot of thought and done a lot of work to figure out how to do so.

My office is presently implementing a new two-to-three year strategic plan. As the plan has fallen into place, a lot of the work I had done has been incorporated into it. Even better, my colleagues have come up with new ideas that either complement or improve on mine. Fresh sets of eyes bring fresh perspectives, and people who haven’t lived with the day-to-day tasks of data management can help those who do see the forest through all of the trees.

We recently had a meeting with folks from outside our office to talk about our data plan. They will be helping us put it into motion. One of the social scientists we met made a point that summed up how I’ve felt the past couple of years: How do we move from data management to data evaluation? We all agree that we’ve come up with a framework to make that leap.

I love it when a plan comes together. Now if we can just continue our work without further interruptions, everything is going to be great.

Strongest Signal That I’ve Seen

I’ve added a privacy policy page to my site. I am not going to pretend that I fully understand the European Union’s General Data Protection Regulation (GDPR), mostly because I haven’t bothered to read it. Not to say I am completely oblivious: I’ve mostly read the emails I’ve received from every website I’ve ever submitted my email address to, which has been hilarious, especially when I got emails from websites I had forgotten about.

Technically speaking, the GDPR shouldn’t affect this site, but I made sure to include a privacy policy on our Eurovision blog, especially because we use Google Analytics and AdSense on our site. WordPress had provided a template, but to be frank, I found it a little bit too bloated for my liking.

Inspired by a tweet from Fiona Bradley, I wrote up a brief description of how our site uses cookies and what Google and Automattic (WordPress’ developer) does with them. I know I have a tendency to be verbose so I tried to keep it as simple as possible, then linked out to the Silicon Valley jargon repositories for more information.

That done, I could sit back and pop some popcorn to enjoy while reading about non-European websites caught off guard when the regulation went into effect on May 25.

We Want Qualitative Information

I am quite desperate for a way to manage and analyze qualitative data.

My main job task at work is managing quantitative data. If it’s a number we can plug into a spreadsheet, I know how to collect it. There is plenty of room for improvement and, to be honest, I am cursing the fact that I didn’t pay more attention in my Access class in grad school. But generally speaking, I have a good handle on the types of quantitative data we collect, the flaws and the needed improvements to our processes and to our datasets, and the ways we can use that data to tell our story to whoever asks.

But we have access to all sorts of qualitative data as well. For example:

  • Reports from foreign service officers;
  • Cables from posts;
  • Reports submitted to an internal reporting system;
  • Posts in the community forum on our website;
  • Posts in our Facebook group;
  • Newsletters and other activity reports that are either sent to us directly or shared via our email lists.

All of this is spread out over a variety of disconnected locations. We have troves of information stashed in mattresses all over our house and seemingly no good way to tie it all together.

So that is my holy quest: to research and compile ideas for managing qualitative data and figure out how best to implement those ideas. I am told that it better not just result in a word cloud.