Last Thursday was a real treat for the M+R community.

A couple of former colleagues and old friends, Toby Fallsgraff and Marie Ewald, who played key roles on Obama’s digital team (Toby was Obama’s Email Director and Marie was Obama’s Deputy Digital Director) attended our first intimate Speakeasy to share their inside scoop and insights with our clients and us.

What they shared over peanuts, gin, and ginger ale was both surprising and liberating.

These guys had the team (18 staff just on email!), budget, analytic capabilities, and a massive enough list to do pretty much anything you could possibly dream of to get results from an email fundraising and advocacy program. And here’s what they found with all of this at their fingertips:

  • Segmentation based on donor behavior is the only segmentation worth doing. For fundraising at least. But this is kind of a big deal. Even with all that data about who these new sign-ups were, how they voted, where they came from, and what their demographics were, it was donor behavior that mattered far more than anything else. There really wasn’t one example they gave where segmenting by some other variable helped dramatically with fundraising.
  • Even the most experienced email campaigners got stuff wrong. The digital team held a contest for a few weeks to see which staffer could correctly pick the winner of their email tests (eighteen versions of a message, for example!) No one ever got it right consistently. In fact, Toby said that he scored near the bottom at the end of the contest. A good reminder never to assume that what you like is what your supporters will like.
  • Novelty worked… until it didn’t! The team continually tried and retried new tactics, and found new approaches that worked one week stopped working the next. It convinced them to continually re-test what they had previously found to be true.
  • Shorter vs. longer? They tested message length. It mostly didn’t matter, save for the end of the campaign when shorter messages tended to perform better. Content and relevance? They always mattered. [NOTE: Generally, shorter messages worked better, particularly towards the end of the campaign, but it was always worth testing longer.]
  • Microgoals worked for advocacy but not for fundraising. Microgoals are those emails that reference a very specific number of people – like “6 new donors are still needed in Washington, DC”. This tactic worked for their advocacy and mobilization efforts, but failed to make an improvement for fundraising. That said, setting a big goal like number of donations needed in one week often helped.
  • Facebook app made a huge difference for their mobilization efforts. The app allowed the campaign to ask supporters to contact specific people on their friend list based on geography via Facebook. Toby and Marie estimated that millions of additional people were reached this way that weren’t reachable via any other channel.
  • Best performing appeals often had the highest unsubscribe rates. Turns out, evoking passion in supporters worked both ways, but ultimately the campaign decided the positive fundraising results were worth the increased unsubscribes. Even when considering retention, the conversion stats outweighed the downside of losing people.
  • They hardly integrated with snail mail. The online program was mostly a separate entity from the direct mail stream. They used some of the same basic branding and content, but, by and large the channels were optimized to raise the most revenue possible, and that meant not integrating the details.
  • Clickers and visitors to donation pages were strong prospects for resends. They found that people who had started to make a donation, either by clicking or visiting the donation page, responded positively to follow-up emails asking them to go through with making a gift.

Some surprising insights, right? Even acknowledging that the message volume, list size, and time-limited nature of a presidential campaign are different from that of a non-profit organization, I think these insights are pretty interesting, useful, and testable.

And how cool is it to be in a line of work that is this dynamic? Nothing stays the same! Winners become losers! Experts are wrong! Love it – and, personally, I wouldn’t have it any other way.