Published June 10
Hal Malchow died on March 21, 2024, on his own terms in Zürich, with champagne, Frango mint chocolates and one last urgent message. He was 72, a Hall of Fame political consultant — yes, there is a Hall of Fame for political consultants — and arguably the most consequential political targeting mind of his generation.
He pioneered microtargeting when consumer databases were still a novelty, set fundraising records for the Democratic National Committee that stood for years and built MSHC Partners into the country's largest voter contact firm. His work was chronicled in Sasha Issenberg's The Victory Lab as the secret science behind winning campaigns. In many ways, Hal was that science.
I have a personal connection to his legacy. I worked at Blueprint Interactive, which grew out of the digital department at MSHC, and Hal’s partners, Rich Schlackman, Trish Hoppey, and Chris Cooper, were early mentors who showed me how to be a political consultant. I had the privilege of helping Hal build HalMalchow.com. Working in that orbit, you absorbed something essential: follow the evidence, even when it’s uncomfortable.
Hal's final evidence was deeply uncomfortable.
In Reinventing Political Advertising, the book he finished before scheduling his death, Hal delivered a verdict on the industry he'd helped build: Political advertising, he argued, had historically produced small or zero measurable effects on general election voters. Although the latest Gallup numbers show that 45% of Americans identify as independent, in practice over 75% of those independents lean either Democrat or Republican. When almost 90 percent of Americans are pulling the straight-party lever, spending nearly all of our money on candidate persuasion ads isn't science. It's inertia.
But Hal didn't stop at the critique. He pointed toward someone who can be persuaded: the uninformed voter.
Reaching the uninformed voter
In one experiment Hal cited, voters were divided into two groups: those who knew which party controlled Congress and those who didn't. Three mailers were sent to both. The informed voters moved less than one percentage point. The uninformed voters moved 19points. Same message. Nineteen times the effect.
His conclusion was unambiguous: We are targeting the wrong people. The voters who can actually be moved by advertising aren't the ones campaigns are fighting over. They're the ones being largely ignored.
That insight lands differently now, more than two years after his death, because the data infrastructure to act on it exists, and the tools to deploy it have never been more powerful.
The voter file remains a foundation for planning, but voter files are a lagging indicator. Campaigns are over-indexing on them instead of using the digital signals available right now, in real-time through the bidstream, billions of times a day.
Every time you load a page and an ad appears, it means a silent auction just happened. You didn't see it. It took less than a second. Your device broadcasts a signal: Here is a person, here is where they are, here is what they're reading. Do you want to show them an ad?
The signals campaigns are missing
After the Dobbs decision in June 2022, there was a sudden surge in political interest from audiences that had rarely engaged with political media before. Women consuming lifestyle and parenting content suddenly developed an urgent interest in an issue that drove many of them to the polls that November. The voter file didn't catch that signal as it was rearing its head. But the bidstream had it; the behavioral fingerprint was right there in the data. These are exactly the kinds of voters Hal described: people who weren’t paying attention until suddenly they were.
The same logic applies to future events.
In the months since the war in Iran began, gas prices have risen dramatically. Every fill-up is a small political event. Imagine an ad highlighting your opposition to the war, targeted to frequent GasBuddy users, people actively hunting for the cheapest gas in their area. The voter file can't tell you who's opening GasBuddy three times a week. The bidstream can. That's a behavioral signal of economic pain, a clear opening for persuasion, that traditional targeting methods would miss entirely.
A new model for persuasion
Presidential cycles require a different application of the same logic. In a presidential election, the likely voter is already saturated; they're consuming news and political content, and Hal’s data suggests just 1% are moved by political advertising.
The truly persuadable voter is often someone who has only recently begun paying attention and hasn’t yet hardened into a partisan identity. The bidstream provides the data to find them, and you can go further by layering in your automated content recognition (ACR) datafrom smart TVs and streaming devices to identify voters who are already seeing your ads. Now you have two powerful exclusion audiences: people consuming political content who are already decided, and people already inside your ad universe. What remains is a refined universe of genuinely persuadable voters who aren't yet reached and aren't yet hardened.
Bidstream-derived demographics, media consumption data, psychographic data and affinity segments all give campaigns additional signal layers that go beyond voter files, and each one opens a new axis for experimentation. Successful campaigns will be the ones treating every buy as a test, layering new data signals, seeing what the data returns in real time and adjusting. That spirit of experimentation is exactly what Hal was calling for.
Hal Malchow spent his career building the tools that defined political targeting — and then spent his final months telling anyone who would listen that those tools were no longer enough. That takes a particular kind of intellectual courage.
Before he passed away, I texted Hal thanking him for everything he’s done, letting him know that he was a hero of mine and that I vowed to continue his mission.
He wrote back, “Yes, continue to fight!”
Two years later, the work he pointed us toward still needs to be finished.
This op-ed represents the views and opinions of the author and not of The Current, a division of The Trade Desk, or The Trade Desk. The appearance of the op-ed on The Current does not constitute an endorsement by The Current or The Trade Desk.