Sarah Longwell began asking Republican voters questions because she thought she knew the answer. Donald Trump had won the presidency, but surely there were enough Republicans who wanted somebody else. Find a challenger. Demonstrate the appetite. Give the party a way out. It was a reasonable theory, particularly if the Republican you had consulted most thoroughly was Sarah Longwell.
Then other Republicans began talking. They liked the disruption. They liked the businessman. Some of what appalled her was precisely what appealed to them. The constituency she had imagined was less available than she had hoped. Listening had produced an awkward discovery: her own convictions were a poor substitute for research.
That disappointment became an occupation. Today, Longwell runs a Washington communications firm, publishes The Bulwark, hosts The Focus Group podcast and writes about the electorate. Her September 2026 book, How to Eat an Elephant: One Voter at a Time, carries the unit of work in its subtitle. Even the elephant has to be approached individually.
The Republican who liked an argument
Longwell grew up in central Pennsylvania and studied political science at Kenyon College, graduating in 2002. John McCain was her kind of Republican. Returning to campus in September 2026, she described her undergraduate self as “pretty obnoxious” and recalled enjoying arguments. A college education offered an agreeable supply of people willing to provide the other side.
Her former professor Pamela Camerra-Rowe remembered a student with firm views who also wanted to understand how things worked. That combination matters in Longwell’s later career. Conviction gets her into a fight; curiosity occasionally changes her understanding of the battlefield. Neither quality has required her to become especially demure.
After college came the Intercollegiate Studies Institute in Delaware. She worked on promoting Rick Santorum’s 2005 book, It Takes a Family. Santorum’s opposition to gay marriage made the assignment personally difficult for a young lesbian beginning to live openly. She left with a resolution to be out in her professional life.
Her next apprenticeship was at Berman and Company, where she rose to senior vice president and partner. Public affairs work meant representing clients, mastering an argument and making it stick. The work sharpened skills she would later apply to a political cause of her own. Persuasion had been her trade before Trump became its central subject.
There is a temptation, with political biographies, to make every early job a rehearsal for the principled stand to come. Longwell’s history resists that tidiness. She spent years inside conservative advocacy and corporate communications. Her subsequent break carried weight because she had a professional life there, with colleagues, commitments and a plausible future.
A resignation with consequences
She became the first woman to chair the national board of the Log Cabin Republicans, the organization for LGBT Republicans. In 2016 she helped prevent a Trump endorsement. When the organization endorsed his reelection in 2019, she resigned. The dispute had passed the point where remaining at the table seemed useful.
Longwell had already learned something about inconvenient affiliations. She married her wife in 2013 while continuing to work in conservative circles. “I got comfortable with everyone being mad at you,” she said. The sentence has the economy of a person who has stopped keeping a seating chart of the offended.
Her break developed into a collection of institutions. With Bill Kristol she founded Defending Democracy Together in 2018. Republicans for the Rule of Law and Republican Voters Against Trump gave specific names to specific arguments. The names did some of the introductory work: a voter could see the position before anyone had delivered a speech.
In 2019 she also founded Longwell Partners. Its services include research, advertising, video production and strategic communications. She had moved from the problem of expressing dissent to the practical work of recruiting people, making material and putting an argument before an audience. An objection needs a production schedule if it is going to become a campaign.
The useful discomfort of hearing people out
Her research has ranged across political camps, with particular attention to swing voters and current or former Trump supporters. The people who interest her can hold views that refuse to arrange themselves into a party platform. A person’s previous vote supplies a starting point for a conversation. It does not supply the conversation.
By April 2025, she was describing a schedule of three or four groups a week, with increasingly specific segments within the electorate. Repetition lets a researcher compare reactions across groups and over time. It also means repeatedly encountering people who will not deliver the answer that would make the researcher’s political life easier.
The Focus Group turns some of that work into public listening. Audiences hear voter recordings alongside analysis. Longwell also co-hosts The Next Level with Tim Miller and Jonathan V. Last, and The Secret Podcast with Last. Across those formats, she moves between hearing voters and arguing about what their words should mean for political strategy.
“I sit and listen to 10 people so that I can figure out how to persuade 10,000.”
Sarah Longwell, on Possible
The distinction between understanding an answer and endorsing it is central to the job. A moderator who turns every objection into a debate may win an exchange while losing the information. For a listener at home, restraint can look maddening. For someone planning a persuasion campaign, the uncorrected answer can expose the obstacle the campaign actually has to address.
There are limits. A small discussion cannot tell you how an entire country will vote. Group dynamics can shape what people say, and an analyst can hear what she hopes to hear. Longwell has acknowledged both problems. Her own original misreading of fellow Republicans supplies a useful warning against treating personal familiarity as electoral knowledge.
Permission to leave the tribe
Republican Voters Against Trump put ordinary voters in front of the camera to explain their change of mind. The premise was social as well as political: people might listen to somebody whose earlier loyalties resembled their own. A former Trump voter could offer a kind of recognition that a familiar Democratic spokesperson could struggle to provide.
The strategy asks a practical question about belonging. If voting Republican has been part of someone’s identity for decades, changing a vote can feel larger than choosing between candidates. Hearing another Republican explain the same decision can make the departure less solitary. Longwell has described building a small community within the larger political tribe.
Her 2024 TED talk took this problem to a wider audience. Its subject was why people vote against their party, and the role of community in making that possible. The researcher's challenge and the campaigner's challenge meet here: establish enough trust to hear an honest answer, then find a messenger who can earn a second hearing.
The breadth of her desired coalition can sound almost mischievous. In a 2024 conversation about democracy, she posed the task of spanning Elizabeth Warren to Liz Cheney. That is a considerable amount of ideological furniture to fit into one room. The common purpose, in her account, is protecting democratic institutions despite disagreements about what government should do.
A home for people between parties
The Bulwark gave that political estrangement an editorial address. Longwell helped found the publication in 2019, working with figures including Kristol, Last and original podcast host Charlie Sykes. Its audience could share opposition to Trump without sharing a complete political program. For readers unsettled by their former affiliations, the publication offered people to argue alongside.
It also became a business. An August 2026 account put the previous year’s revenue above $20 million and paid subscriptions above 140,000. Those figures describe the publication, rather than Longwell’s personal wealth. They suggest that an audience for political companionship can support more than a modest newsletter.
One early subscription calculation captures the difference between Last and Longwell. He anticipated perhaps 7,000 paid subscribers within six months; the publication attracted 12,000 in roughly two. He credited her with seeing the potential sooner. Their recurring on-air disagreements have acquired a commercial benefit: people pay to spend more time with the conversation.
The publication added video versions of podcasts on YouTube in 2023. Giving the conversations faces opened another route for people to discover them. The expansion fits Longwell’s work as a communicator: once you have something to say, you still have to ask where the intended audience spends its time.
Back in the room
On September 16, 2026, Longwell returned to Kenyon’s Rosse Hall with her new book and sat down with Camerra-Rowe. The former student who enjoyed an argument was now explaining what years of listening had taught her. Hundreds attended. The setting placed a long professional detour back inside one of the classrooms that had helped launch it.
The book is 336 pages, with a foreword by Jonathan V. Last. It brings together her political journey and her case for a broader movement against authoritarian populism. She wants communications infrastructure, persuasive leaders and a coalition capable of enduring beyond one election. Those are her proposals, with the electorate still entitled to complicate every one of them.
By 2026, she was publicly describing herself as a former Republican. Her advice to Democrats remained demanding: talk directly about what matters to voters, including their economic circumstances and sense of safety. A person who has left one party has not necessarily signed away the right to irritate the other.
Her work now runs between a communications firm, a publication, podcasts and a book. All require speaking. The experience that gave them their direction was the less flattering one of hearing her theory fail. Longwell went looking for agreement and found a reason to keep asking questions. The next voter still gets to answer.