The Path to Being a Learning Organization
I’ve been thinking about an essay Learning Organizations by Richard Sedlmayr on Substack on behalf of the Agency Fund. I’d never heard the term; maybe he made it up. He refers to them as “entirely new institutions that unify research and action under a single roof” Research and Action. In fact he calls for the “cultivation of an institutional ecosystem linking research and action” and it really got me thinking of the path Turnout Nation has taken.
After the 2016 election I looked at the math of that election. The public discussion seemed to focus on “why the voters voted the way they did” But the concept of there being “voters” seemed odd. There are people who voted every election, and some who voted every even year November, or vote from time to time. But “voters” is in no way a static group. What really matters in terms of electoral outcomes as well as the simple state of democracy is who isn’t voting, why and how do we help them vote?
I set to talking to very many people who hadn’t voted. Almost all felt they should have voted but were not very well informed. They all agreed that if somebody they knew and trusted had helped them and if that person would know if they voted or not….then they would have voted.
Solvable problem, I thought. The plan then was to build a way to help volunteers to get friends to vote and know if they had voted or not by linking to the voter file and that meant technology. We spent years and lots of money on an app so we could scale.
But does this really work? Everybody wanted us to prove it. But how? They were not very specific about that part. We called many many possible entities and asked what we needed to show, what type of trial or proof, at what scale? Finally we found a professor, Don Green at Columbia, who told us. We did that, a Randomized Control Trial; a small one but rock solid miraculous results, “order of magnitude” results. Nobody cared. We got some money but not much, and the worst thing about the money was that it was money designed for “interventions” related to voter turnout, and 99% of the time that means stranger-to-stranger contact, some methodological tweak to batch texting, phone banking, sending mail, or knocking on doors. These are things that are done by HQ to the voters to see what percent of them will be motivated by whatever it is. The voters are the data and not much else. Not surprisingly the turnout effect of these methods are extremely modest, but the methods are cheap and can be done by HQ in a quick and predictable way.
But an organized effort to get volunteers to get some friends to vote takes time. You are dealing with humans. Data points sure. But the fact that they are real live people is a bit more front in center than with mail, text, or call recipients.
We did a number of trials after that but all of them where done with funds that came very shortly before elections and to be honest most of our time and attention was focused on fundraising and making payroll. It was impossible to build trust, which is an important first step to recruitment. And no donor wanted to give us funds for the period after an election, because then, for them, it was over, that wasn’t the voter turnout period. But of course it wasn’t over for the voters.
And that is one of the big problems with voter turnout efforts and all efforts. They end once they get the data.
Which is why the idea of research and action under one roof is cool. Without thinking about it, that’s what we have done. Turnout Nation is now basically a zero budget organization focused on proving that our Captain Method works. For the 2024 election we hope to prove that it can be done at a giant scale in a high turnout national election and that people will notice. Along the way we will get very very many people to vote in two key states, Nevada and North Carolina. We learned tons from loosing our funding. First we learned how liberating it is to stop fundraising. Second the framing of “how giant a trial can we conduct for free?” turned out to be the key question we needed to ask in order to get us going in the right direction. Third and last, if the trial we are working on now works it will get lots of attention. And if it doesn’t, then the problem is the system not us.
In the six years we have been working, we have found that donors are interested in funding learning in theory but that they have extremely narrow definitions of what constitutes learning; definitions based on how things have always been done. What’s interesting about the Agency Fund, is that they are focused on human agency, people taking responsibility for themselves, knowing their own power, the exact opposite of what most voter turnout interventions are based on. And they way they look at their own work seems to foster that. Our experience with donors is that it is difficult to find ones that encourage or even support real agency among their grantees. And that probably says something about how they feel about the people they hope to do something useful for.
