The Grass Is Greener: Analytics Department Highlight
What really goes on in those other campaign departments you’re always hearing about? You probably have a basic idea, but we want to take a closer look with you so you can truly understand how the work of other departments on your campaign affects what you do as an organizer and vice versa.
Every department plays a role in supporting organizers like you, so let’s see what’s going on behind the scenes.
This issue we’re learning more about analytics departments with Andrew Claster.
What does a campaign analytics department do?
An analytics department on a political campaign exists to help give decision-makers the tools and information they need to make better decisions. These tools can include a spreadsheet that ranks media markets or voter targets, a caucus simulator, a targeted list of voters, a predictive model, a map displaying useful information, and much more.
To do this, analytics teams analyze all of the internal and external data that a campaign has access to, including: the voter file, other internal and external lists, volunteer data, fundraising data, paid media data, polling data, voter contact results, etc.
That’s A LOT of data.
A presidential campaign will generate literally billions of data points. Because of the huge amount of data, the analytics department needs to:
- Prioritize which data and analytics projects are most likely to deliver highest return on investment (ROI). ROI on a political campaign is the votes per dollar per person and hour.
- Conduct that analysis of the projects’ ROI accurately and quickly at lowest cost.
- Communicate the results of that analysis in a way that makes sense to decision-makers.
- Translate analysis into recommendations that are reasonable and can be implemented.
- Measure results.
When Andrew worked in the analytics department for President Obama’s reelection campaign, they viewed themselves as an internal consulting group. “We met with every campaign department – paid media, fundraising, field, communications, operations, political. We asked them: What do you know already? What don’t you know that you need to know in order to do your job better? What can we give you that will help?”
“Then we figured out how we could give each department the tools they needed. Creating the best tools was an iterative process, but by the end of the campaign, we were able to support every department in the most effective way.”
What parts of the work of an analytics department are most relevant to organizers?
- Vote goals: Analysts help determine vote goals. What is our baseline? How do we get to victory using voter registration, persuasion and GOTV?
- Modeling: Analytics teams create models to help decide which voters you target for voter registration, persuasion, and GOTV.
- Mapping: Okay, so now that you know who your targets are, where are they? Analytics teams map your targets and help assign turf to reach them.
- Resource Allocation: How many field organizers, volunteer leaders, and volunteers does this campaign need? How do we assign them?
- Campaign Techniques: Data and analytics help determine which campaign techniques are most effective for registration, persuasion, and GOTV. Spoiler alert: it’s almost never yard signs.
How does an organizer’s work affect an analytics department’s work?
Your work as an organizer affects an analytics department’s work in two main ways: execution and data.
- Execution: Nothing the analytics department does matters unless the volunteers and organizers in the field use it. Andrew describes it this way. “I used to have an orchestra teacher who would wave his baton in the air to demonstrate to the audience that he can’t make a sound unless he has an orchestra full of musicians who can play. The analytics team isn’t exactly like an orchestra conductor – maybe more like the guy who tunes the piano. You can be the best piano tuner in the world. If nobody plays the instrument, nothing happens.”
- Data: The data collected by volunteers and field organizers is among the most valuable data the analytics team has. Your data tells analysts who was canvassed, who they support, how likely they are to vote, what their most important issue is, and much more.
The data collected from door knocks and phone calls placed by volunteers is critically important. For instance, it’s a major input into model scores. In addition, organizers provide both a gut-check and a test-bed for the conclusions and recommendations that analysts develop. If analysts have made a mistake, organizers are often the first to show them that something is not right.
Well, there you have it. Now we know what an analytics department does, how their work affects your work, and how your work affects theirs. (Though I’m still not exactly sure what they do on their computers to complete all this, and I’m pretty sure I’ll ever know. But I do know they love Microsoft Excel.)
If you’re interested in possibly working on an analytics team in the future, here are some basic steps you can take to prepare yourself:
- Learn MS Excel very, very well. This will help you organize as well! VLOOKUPs, pivot tables, and text columns are extremely versatile.
- Learn SQL.
- Consider taking a couple statistic courses (after Election Day, obviously).
- Learn a good statistics program. R and Stata are the most commonly used options in politics, and SPSS is common among pollsters.