Learning from Loss: GOTV

Election night’s results were a microcosm of the last year in several ways. Back
In the Iowa Caucuses – you remember those, oh so many months ago – Mitt Romney came in with history, with money, with inevitability. And inevitability was derailed by former Pennsylvania Senator Rick Santorum.

Santorum won by more or less living in Iowa. By going door-to-door, by retail politicking of the variety most loved by Iowans. Thing is, it was little different than what Santorum had done in his initial run for Congress against Doug Walgren. In that campaign, Santorum pretty much knocked on every door in the district. Knocking on doors gets voters up and interested. It’s not that Mitt Romney’s machine for getting out the vote – GOTV in wonk parlance – wasn’t itself formidable. Mitt’s GOTV architecture had been set in stone in Iowa four years earlier. Still, retail politics trumps phone calls every day of the week – and most of all on Election Day.

Today, John Edkhal over at Ace’s expanded on a discussion of the Romney campaign’s solution for getting out the vote: a massive software tracking system code-named ORCA. And between his perspective (as someone who was to be a field user at a polling station, and as a software developer), combined with comments to the article, one thing is clear: ORCA did not work. Because while it worked for some, seamlessly, for others it was a nightmare that failed to account for everything from the use environment (the interiors of libraries and school buildings often have poor cellular reception), documentation (being emailed late, if at all, with grossly disqualifying errors leading to some watchers being evicted), and other problems.

While I applied to participate, I never received follow-up communications. Combined with a hectic work schedule, I played it safe and did what I could at the downstream end.

I did my time over the last week in call centers supporting Mitt Romney, and what I heard on the phone confirmed that the lessons of Iowa weren’t necessarily applied as they should have, or could have been. Call center life is fairly straightforward: first, you dial a phone (or rather, it is dialed for you with a button push). Then you wait. About 70 percent of the time, a call ends at an answering machine. In this case you press a couple of buttons to leave a message, then complete a survey saying “left a message” and repeat; 15 percent of the time there would be a dead line or someone would pick up and hang up. The other 15 percent of the time you’d get a live human being, and talk with them – on election day, it was to ask if they’d voted. Previously, it was when they planned to vote.

Most of that 15 percent had voted. The problem is, most had voted absentee or in early voting five to seven days earlier. Only a handful were going to the polls later that day or had gone earlier in the day.

And was their any upstream connection to ORCA? Most of the numbers dialed for me were local, but there was the occasional out-of-state area code. So, maybe something was making its way out of what was happening elsewhere. One number I called was to a person whose polls had just closed. I got an earful for that. So perhaps there was – though between John Edkall’s piece and Joel Pollak’s expansion at Breitbart, it’s hard to tell for sure.

Here lies the problem of the Romney GOTV machine: phone banks are an effective way to get in people’s ears. But not in their eyes, in front of their faces. And in an era of Billion-dollar campaigns, a tiered approach that increases and starts with direct interaction seems a no-brainer. Read this piece over at the Boston Globe to see how Mumbles Menino got the vote out – though at a lower rate than the rest of the state, it was still effective, because some of the precincts in Ward 14 went 98-0.4 for Obama. A brief excerpt:

The team included political operatives from the Service ­Employees International Union, which also provided a battalion of volunteers. They made a scoreboard on the wall to track each “turf,” a segment of a precinct that could be walked by one or two people. The operation there targeted 65 precincts in Roxbury, Mattapan, and Dorchester, Democratic strongholds packed with sporadic voters. The goal was to knock on the door of every registered voter two or three times.

Entire precincts. Every door. One would think this would be second nature to one from the LDS church. But I digress.

So what should we do in the future to fix this GOTV problem?

The front-loaded primary schedule leaves plenty of time for the eventual winner – and the party apparatus prior to formal nomination – to review primary results to see who voted, and where they live, and to see more importantly who didn’t vote, because that’s who needs motivation – physical interaction – to be engaged to vote in the General Election. These groups vary by size based on when a state lies in the primary season. A Iowa or New Hampshire will have many more voters than a New Mexico, which votes usually after the primary is settled. It is this latter group which needs direct interaction, to (a) determine if the registered voter is still there, and (b) to start encouraging the voting process. Distribution and collection of Absentee Voting forms, where allowed, can and should be done here. But with what?

The base needed to do this outreach starts forming now. Out of election season. The same people who work a phone bank can knock on a block of doors, as people did in Boston. Getting those people organized on our side has to start now, out of season. Erick Erickson over at RedState has long been a proponent of controlling the party from the ground up. This GOTV debacle begs for exactly that kind of intervention.

Then there are the weeks before the election and election day itself. I confess to being one of those that lets all calls without a familiar number go straight to voicemail. And even in a non-competitive state like ours, we would have 4-6 call a night, every night. Door knocking is harder to avoid. It’s also more time intensive. but it’s worth it in getting people to vote.

Did ORCA cost Romney the election by itself? No. But relying on ORCA and phone banks and limiting the number of door knocks did, among other things. Relying on untested software, phone banks, and an unseen yes or no will only net you so many votes. Let’s learn from this and get rolling soon.

There’s another lesson to be learned, for the next post. Another lesson unlearned by Romney that cost in the final two weeks, as I feared it would. But again, that’s for another post.


3 thoughts on “Learning from Loss: GOTV

  1. Very good. I might add that the GOTV lists I had were terrible…I tried the walk list which even listed a couple as being registered at the local city park! There are some serious issues here that need to be dealt with if we are to win in the future!

    • We’re in an era tech-wise where it’s not only straightforward to process street data into a geo-referenced form to identify false information (people registered at parks, non-residential addresses like PO Boxes) and to do what FedEx and UPS have done for years with deliveries to GOTV. It’s a matter of someone recognizing the right role for tech in GOTV and avoiding the wrong (see: ORCA).

      Thanks for the read and the comment, PF!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s