If It’s Broke, Fix It

At about the 15-minute mark, the sensible thing would have been to go home and try again later, but by then perverse curiosity was calling the shots.

Can this
get any stupider?

In this case, *this* was us trying to cast a vote in the first hour of the first day of the Advance Poll in our local community centre. You know, before they got busy. As it turned out, a few others in our neighbourhood had the same idea: There were unexpectedly high numbers of Advance voters, especially for one polling station among the 5 or 6 that our centre was serving. Our polling station, as it turned out.

Was *this* a finely tuned system being overwhelmed by too many pesky voters? Not so much. Let’s look at the things that went wrong.

New-to-their-jobs staff. Confirming identity, checking-off names of eligible voters, marking/recording ballots for later scrutineer-ing, handing out said ballots, and managing the flow of people into and out of the voting booth: these tasks are all important to election integrity and have to be done a certain way. I’m guessing the job neither pays well, nor offers many hours. As a result, election to election there is little continuity in the worker pool: Training newbies will be key. The good news? The training was done. The bad news? The reason I know it was done is because I saw it being done in front of me, as I stood in line.

Unsuited-to-their-jobs staff. Checking photo ID against the List of Electors is not conceptually tricky, but to do it efficiently you have to know the order of the alphabet by heart: knowledge our poll worker was a little shaky on. It also helps to know how to quickly check whether a given name can possibly be on the print-out page you’re looking at. (Hint – Before laboriously reading every line, check the name at the top and at the bottom.) (Potential problem – This method requires that you know the alphabet.) Anyway, our poll worker did not know or was not able to avail themselves of this hack. I had ample time to watch a worker at another polling station manage an admittedly much smaller flow of voters: They were never backed up more than 4 deep, and they moved steadily when they were in that line, with each elector processed in about one minute. By contrast, our line stalled completely, several times, for 10 minutes at a time. I’m not kidding.

Insufficient staff, full stop. Each table had just one person assigned to it: to check ID, find the name on the List of Electors, scratch out that name, fold the ballot, initial the ballot, and record the ballot number by hand. If anyone showed up with valid ID but no entry on the List of Electors, the worker had to enter their name and ID by hand. When one polling station backed up, as ours did immediately upon them opening the doors, there were no surge staff available.

Too few sergeants. Do people-as-individuals know their jobs? Have these people-as-a-group worked together before? Is the work coming at an expected and manageable pace? When all of these conditions are met, there is little need for supervision: not zero, necessarily, because things can still go wrong, but not much, either. When none of these conditions are met, there is a big need for supervision; for “extra” people who aren’t doing the hands-on work themselves, but who are watching to see where things are going wrong, and who are able to take action to fix it:

• Adding resources (if available) – as one example of creating availability, combining two (or three!) unbusy polling stations into one, and reassigning their staff to the overloaded one
• Reassigning overwhelmed workers to less-demanding tasks at their initial polling station (if possible)
• Reassigning overwhelmed workers to less-busy polling stations (if possible)
• Getting help from a central office (if available)
• Advising would-be voters of their likely wait times before they join the line (no caveats here, this is always possible, desirable, and even essential)

This Advance Poll likely needed several supervisors: It had one. They were also responsible for doing the on-the-job, not-quite-in-time training that I saw in action, and for eating their breakfast at work.

After 45 minutes of standing in a line that was barely moving, we saw two people join the one guy at our table. With three of them on the case, things started to move: not quickly, exactly, but any movement was welcome. I have no idea where they came from: what function they had to shut down to free them up.

Was it just bad luck that our Advance Poll was uncharacteristically busy? Sure. And it *was* uncharacteristic: Reports from across Canada told the same story of delays.

Was it just bad luck that the seemingly least competent person in the room was facing the most people clamouring to vote? Yeah, likely.

Was it just bad luck that this choke-point continued for more than 10 minutes? No. That was a matter of choice:

• Before the fact: staffing and training choices
• On the ground: supervising choices

In other years, voting at an Advance Poll has taken no more than 10 minutes, start to finish, with most of that spent walking from the car into the poll and back again. This year, it took an hour and change. Is that too high a price to pay to live in a democracy? Certainly not. But . . . two things.

First, as noted above, a small amount of timely communication on probable wait times would have gone a long way and cost them/us nothing in extra staff. This tweet came out on Saturday and its relentlessly cheery, we-didn’t-do-anything-wrong, it-was-an-ambush (see the video clip in the sidebar) style of communication doesn’t, in fact, encourage my (presumed) patience.

Second, as our politicians are fond of saying, let this performance or lack thereof be a teachable moment for all of us, even if not for the, you know, actual performers. Things don’t always go as expected: it’s good to be ready for the unexpected. Thinking ahead costs us nothing.