Campaigns and elections are hard. There’s always more work than can possibly get done, and people will always want to get it done faster than you think is possible. Most of your colleagues are overworked, underpaid, lacking sleep, eating terrible food, and living at the office. While they may normally be wonderful people, a lot of campaign staffers become the worst versions of themselves during election season: stressed out, irritable, short-tempered, and otherwise unpleasant.
I don’t say this to damper your enthusiasm for working in progressive politics, but you need to understand that going in. In terms of your application, it’s why having the hard skills won’t be enough to actually succeed in this industry. Because election season is such a pressure cooker for everyone, it’s important to assemble a team that can work well together even under those circumstances. This team will have to deliver an incredible quantity of results, and ideally, they’ll do it without wanting to kill each other.
If you know all of the statistics and programming listed above, then you’ll be a competent analyst. You’ll probably even get a job. But without soft skills, you’re never going to be excellent, people won’t like you, and you’re probably not going to be able to make the type of impact that you care about. This section describes some of the qualities you’ll need to succeed in your role and the types of qualities you might be asked about in your interview. You’ll note that many of these qualities aren’t limited to progressive analytics, so they’ll serve you well no matter where you go.
As an analyst, your number one priority should be to do all of your work reliably. That means that you produce excellent work and you do it on time. Million-dollar decisions may be made from your work. Your successes will rarely make the news, but a mistake might dominate national headlines. You want to be the person that your organization just trusts to get things done on time and accurately.
The first step is to understand the parameters of your assignment and to do that early. Whenever you get assigned a task, make sure you understand exactly what you need to do and when it needs to get done by. Here are some questions to ask whoever you assigned:
When does this need to get done by? Is that an internal deadline or a hard deadline? Is that the deadline for a draft or for the reviewed version?1
What outcome do you want and in what format? Are you looking for a table, a graph, a powerpoint presentation, an e-mail, etc.?
When is a good time for you to review my work? 1 day in advance? 2 days? Who else should review my work?
Is there anything in particular to watch out for?
Who is this for? You or another audience?
(Why am I doing this?) You’re not necessarily advised to ask this question directly, but it’s important that you understand why you’re doing what you’re doing. This is how you can check to see if you’re actually accomplishing your objective. A good analyst will do what their boss tells them to do; a great analyst will do what their boss should have told them to do.
Once you figure out what you need to do, you need to figure out how to do it and whether there will be any obstacles. You should ask around on your team about whether anyone has done similar projects in the past, or if there’s anyone with specialized knowledge that is helpful. Try to talk to them and review past projects. Once you’ve got that figured out, make a plan:
How long will things take?
What are each of the steps?
What parts will be trickiest?
Do you have all the data you need? Do you have all the code you need?
Do you need to depend on other people to do things for you?
As you’re making these plans, also make sure to keep your boss in the loop. Especially for the first couple of assignments or really any complicated assignments, reach out to your boss about what you’re working on. Ask him or her to review drafts of your work (just make sure to indicate what is or is not a finished product). You’ll almost certainly be doing something wrong---that’s just a function of growing and learning. Your boss won’t think you’re dumb: conditional on making the mistake anyway, any boss would prefer that it gets corrected sooner rather than later.
Make sure to keep your boss in the loop if timing changes. Often, you’ll be working on one part of a larger project, and if your work isn’t done, then that has downstream consequences. Yes, sometimes your work will be delayed for forces both within and outside your control, but when those times happen, your organization needs to be prepared to deal with it.
Lastly, don’t panic about these things. There will be times when you are panicking. Those are the times when you need to force yourself to take a walk around the block or a one-minute dance break. Take a deep breath, step back from your work, and figure out what’s going wrong. Then, think about what it will take to fix it. More people? More time? More tools? Tell your team what’s happening, and they will be there to help you. You are never alone.
Learn things for yourself
One function of working in politics is that you’re never going to have the training or ramp-up time that you want. Stuff just needs to get done, and no one will have the time to explain to you in depth how to do something. At best, they might walk you through it once. Modal outcome is probably that someone throws a bunch of links and old code at you. Worst case scenario, you’re just left to fend for yourself. Regardless of what happens, try to make the most of it. Being able to teach yourself things is an incredibly useful skill that will serve you well for the rest of your life.
Take notes and ask questions when someone explains something to you. Time is valuable, and most people hate explaining things more than once. Don’t be afraid to ask questions if people have the time. Most people like feeling smart, and they will like your sense of curiosity.
Use Google. Give me Google, and I can move the world.
Use friends and colleagues. Ask them questions, even if they work in other organizations or other industries.
Teach others. If you want to really strengthen your skills, find opportunities to teach your colleagues or interns. It’s the best way to find out whether or not you actually know something.
Learn non-technical skills. Ask to go to meetings with other parts of the organizations or outside entities. The worst thing that happens is someone says no.2 Go talk to pollsters or field staff. Understand how the things you create are actually being used.
Challenge yourself. If you already know how to do a certain function, Try doing it in another language. Look for new, better ways of doing something and then implement those changes at scale. Though, fair warning, if you really want to go down this route, you’ll have to practice on your own. An average analyst will just do their job and learn on the job. An excellent one will learn outside of it as well.
Optimize on skills that are both interesting and useful. Your boss or other mentors can help you figure out what’s most important to learn. Don’t be afraid to abandon things that you hate. If you get too good at something, you might be obligated to do it forever.
I’m referring to two types of communications here: technical communication and non-technical communication. Technical communication is when you talk to other technical people (note: your boss may not be one of these). Non-technical communication is when you talk to a field organizer or a client or otherwise people who have no idea what it is you did or why. These two audiences will require different forms of communication.
When you’re talking to technical folks, you should focus a little more on process because their primary objective is probably to check or audit your work. What did you do? What assumptions did you make? What alternatives did you consider but ultimately rule out?3 Anticipate questions and objections, and address them before they are raised. Your charts should be clear with titles and axes, your code should be commented, etc.
When you’re talking to non-technical folks, focus more on results than process. How does this piece of information relate to actionable outcomes? How does it change the program they would otherwise have run? How does it make their life easier? Avoid jargon. Don’t be condescending. Be a translator. Visualizations and metaphors may be helpful.
No matter who you’re talking to, be concise. Start with the results of your work and then, as necessary, go through process. You should always know why you’re communicating something. Verbosity does not land well on campaigns, and it’s often synonymous with condescension.
And for the love of god, do not mansplain things. Given how hard it is for women (especially women who are now in senior roles) to enter and succeed in this industry, the average woman you encounter is likely in the top half of skill and knowledge in the industry. Don’t assume that you somehow need to be extra clear just because someone has a high-pitched voice or looks young or is wearing a skirt; they probably know exactly what they’re doing.
This almost goes without saying, but be nice. Listen to other people. Don’t interrupt. Keep an open mind. Be humble. Be patient. Be kind.
You may think you are the smartest person in the room. That might even be true. But that doesn’t mean you’re the most important person in the room, or that anyone is obligated to listen to you, especially if you’re being patronizing or confusing. (Sadly, the two are not quite separable at times.)
Don’t get me wrong--your work is important and helps other people do their work and make decisions. However, there are many other factors that go into decision making and you may not always be a party to them, especially when you’re just starting your career. If things don’t go your way, don’t immediately panic or get upset. There are probably things going on that you’re not aware of.
Finally, be inclusive. You’re part of a community now. If someone helped you with something, make sure to thank them and say their name out loud when you deliver results. If someone would be really awesome to be brought into a conversation, try to bring them in. And if you can help someone advance in their goals, do it.
The role of analyst is to provide ground truth. In practice, that means your job is to provide the best estimate of the truth that you can and to be explicit about the uncertainty and assumptions that go into that estimate. In order to do your job well, you have to do the right thing even when it’s hard. There’s something fundamentally different about the analyst’s jobs versus a communications or policy person on a campaign: you can’t change your tools or methods to support what someone else wants to think.
Everyone working in analytics feels the temptation to say what other people want to hear, but that’s different from what is right and from what is true. You need to have a sense of personal integrity because you will be tempted to believe that what your campaign is doing works and it will be easy to manipulate the data or results to fit that narrative. But your job is not to support narratives. Your job is to help people determine which narrative is actually the most plausible. In order to do that, you need to provide the truth.
1. Seriously, if you’re working on something and you can’t immediately tell someone when it’s due, then something has gone very wrong. ↩
2. Although, to be perfectly honest, you do want to make sure you establish the reputation of being competent first. Otherwise, you’ll just seem arrogant and presumptuous. ↩
3. A quick note on humblebragging: I’m not generally a fan, but it is important to make sure people know how awesome you are. One way to do that is, in the process of describing your work, talk briefly about things that were particularly tricky and show how you figured it out. ↩