posted on
In 50CAN’s Guide to Building Advocacy Campaigns we introduced the concept of a tactics toolbox. Like tools, there is no one tactic that is best for every situation. The key is to select the right tactics to help you accomplish the job at hand.
Our “Inside the Toolbox” series will take a more in-depth look at some of the tactics you might be interested in when building your next advocacy campaign. In this inaugural post, we explore one of the more powerful—and misunderstood— tools in the advocacy toolbox: microtargeting.
The basics
When building a campaign, selecting a tactic should come only after you have clarified your goals and matched your strategy to your environment. Microtargeting is a tactic you should consider if you have chosen the social movement strategy, which is one of the four top strategies you can select when building an advocacy campaign. This strategy focuses on organizing large numbers of people in support of shared goals.
So, what is microtargeting? Rather than trying to talk to every possible person who might be interested in your cause, microtargeting allows you to identify and target likely supporters to help you more effectively and efficiently build your movement.
If you’re a political junkie, this tactic might sound familiar. The George W. Bush and John Kerry campaigns used microtargeting in the 2004 election, and the tactic quickly spread to statewide and local campaigns. Over the past decade this tactic has made the jump into the world of issue advocacy campaigns.
There are four basic steps in microtargeting: 1) obtain a database, such as a statewide voter file, with information on all the possible people you might want to reach; 2) gather information on your current supporters; 3) identify the unique attributes of these supporters; and 4) build a model with these attributes and apply it across the database to better target future recruitment efforts.
Putting it to use
Our first experience with microtargeting was in 2007 in Connecticut. We had been working in the state for two years, during which time we had engaged in a wide variety of recruitment efforts from door knocking to petition drives outside grocery stores to house parties on the achievement gap to staffing booths at town fairs. It was time-consuming and difficult work, and it often yielded a low rate of return.
After reading a news article on the emerging field of microtargeting, we reached out to one of the leading practitioners, Ben Yuhas. Ben was one of the first people to apply this approach to a statewide election, helping lift Tim Kaine to a surprise victory in the 2005 governor’s race in Virginia. With Ben’s support, we set to work bringing microtargeting to the world of education advocacy.
The first step was to obtain a good database. Luckily, we were able to subscribe to Catalist, a shared-service database founded by Harold Ickes to help progressive groups take advantage of the same tools and data being used by presidential candidates. This gave us information on 3,331,773 people in Connecticut across 660 attributes.
The next step was to compare the 10,000 current ConnCAN supporters in the database that we had recruited through traditional means with a random sample of 10,000 non-ConnCAN supporters living in Connecticut. This was no easy task. If you were to simply try to look at the crosstabs across all the individuals and their attributes it would take more than 7,000 hours and you would have forgotten most of it before you finished.
Using sophisticated statistics (discriminant function analysis), Ben was able to tease out of the 660 variables the ones with the most explanatory power in predicting support for ConnCAN. Five personal traits stood out:
- Voted in political primaries
- Lived in a racially diverse neighborhood
- Obtained high levels of education
- Lived in an educationally diverse neighborhood
- Registered to vote
The model predicted that these politically active, well-educated individuals living in the diverse neighborhoods traditionally underserved by Connecticut’s education system would be the most receptive to supporting our cause. We then applied the model across the entire 3.3 million person Catalist database. By comparing the demographics of the top 10 percent of the prospects to the bottom 10 percent, it was clear how different our potential supporters and opponents were:
When we put the model into use as the guide to our recruiting efforts, we were amazed by the results. Individuals with the top scores were six times more likely to join ConnCAN than individuals selected at random. Further, the top scoring prospects were 100 times more likely to join than the lowest scoring prospects.
Using this model, we were able to double the number of members in just a year by reaching out to people who wanted to join the movement but had never been asked.
You can’t be too local
Microtargeting, like real estate, is all about location, location, location.
When we launched 50CAN in January 2011, one of our first projects was to explore how we could help local leaders use microtargeting in their communities. Could they simply take the Connecticut model we had developed and apply it to the work in their community or would they need to build their own homegrown model of support?
In Rhode Island, the Connecticut-based model worked considerably better than a random outreach effort but not as well as our initial ConnCAN tests. In Minnesota, the Connecticut-based model was even less effective than it was in Rhode Island. To be truly effective, we found, the microtargeting effort needed to be customized to the unique characteristics of the state. As with many things about education advocacy, when it comes to microtargeting, you can’t be too local.
Bigger isn’t always better
“When all you have is a hammer,” as the saying goes, “everything looks like a nail.” It’s all too easy to become so focused on what powerful tools like microtargeting can do that you forget all the things they can’t do. In that way we risk letting our tactics become transformed over time into our goals.
Microtargeting can help you more easily find people who might be interested in joining a movement. But simply adding names to a database accomplishes very little. It’s what you do after you make that initial contact that really matters. If you fail to provide meaningful ways to build relationships, establish two-way communication and create concrete ways for your members to contribute to the overall goal, then your overall microtargeting effort will have been wasted.
At the same time, while it’s amazing to have powerful databases like Catalist, the truth is you don’t have to spend a lot of money to take advantage of the key concepts behind microtargeting. Whether you are tracking your supporters in an expensive database or on a free Google spreadsheet, your success will be based on how well you answer a few key questions:
- Who is regularly engaging in our advocacy work?
- What characteristics do these individuals have in common?
- Where else might we find people with these characteristics in our community?
Write these answers down and then put them to the test. Try reaching out to people and tracking how responsive they are to your message. Once they agree to join the cause, track which ones seem to show up to meetings most often, or volunteer to staff events. Then use this information to further refine your model of who you should be focusing your energy on recruiting next. Its amazing how far you can go when you stop trying to speak to the crowd and start focusing on the people who really want to hear what you have to say.
We hope you have found this look inside the tactics toolbox helpful. Please don’t hesitate to reach out with questions, comments and ideas and join us in the conversation about education advocacy on Twitter and Facebook.
Email: marc.magee@50can.org
Twitter: @marcportermagee
Facebook: 50CAN