Donald Trump’s 2016 surprise victory shocked millions of Americans into action, amid real confusion and uncertainty about how best to respond. Experts, including the leaders of many Democratic-leaning organizations were caught flat-footed, while newcomers flooded in trying their own strategies and tactics to build resistance. People pulled guidance from sources near and far, seeking confident voices who offered diagnoses and direction.

While we may (or may not) be done with the pandemic phase of COVID, we are definitely beyond any novel or acute phase of the political emergency of 2017-20 and into the endemic phase of troubles for our democracy. Today Democrats face an uphill battle in the midterms, in many cases running against opponents whose agenda seems ever more extreme on issues from abortion rights to false claims of supposed electoral fraud. Action doesn’t just feel urgent—it is urgent. But it’s high time for that action to be more strategic, which must include taking an unsentimental look at results achieved so far and making changes as needed. What does the evidence of the last five years say about the techniques and tactics on offer? Which ones are going to be most impactful over the long haul, given that the threats to an inclusive democracy did not, it turns out, end with Donald Trump’s exit from the Oval Office?

If you are reading this, you are probably one of the millions of concerned individuals who over the past five years has given your time, talent, and treasure to try to support small-d democracy and capital-D Democrats. You may well have worked as a volunteer on local, regional, and national candidates’ campaigns, or as part of campaigns organized by various national activist organizations. 

This note seeks to help you place the tactics those campaigns prioritize in context, understanding why what is most urgent from their perspective may not be right for you. The “data-driven” or “evidence-based” voter mobilization strategies that campaigns channel volunteers into as they try to win elections should not be either the starting point nor the centerpiece of your own thinking about how to make political change. 

When campaign professionals evaluate tactics, they measure return-on-investment in terms of impact on votes cast in one election, with the gold-standard impact data being that generated by randomized controlled trials. The findings they rely on are useful within their own parameters, but it’s crucial to recognize that those studies: 

• assume anonymous and interchangeable actors; 

• look only at impact within one election cycle; 

• place no value on participants’ learning or relationship-building; and 

• presume that volunteers’ time is free and has no better or higher use. 

It’s natural that campaign practices reflect these assumptions and limits. Campaigns have finite resources, a short time horizon, a pre-set geographic territory, and staff who often have no prior connection to or knowledge of their target communities. Given all of these, they calculate likely “cost per vote” and choose tactics to follow.