Around 6 months ago, something strange happened to me on Instagram. Zohran popped up in my feed for the first time. And instead of the usual sense of impending doom I get from Insta, I binged his reels into the small hours with a smile on my face, and the feeling that maybe everything’s going to turn out ok after all. 

Skip ahead if you are already familiar with Zohran Mamdani:

TL;DR, he’s the Democratic nominee for mayor of New York City, with an impressive list of endorsements and effective “vibes campaigning” that blends politics with culturally savvy media strategy.

So this is about …

Zohran’s strategy

How does he get 4.1 million Instagram followers (with up to 700k likes on some videos) while thousands of other earnest, nice young men with what seem like very similar ideas, rarely get past a few thousand views? 

Yup, of course he’s charismatic, authentic, authoritative, credible, likeable, all the buzzwords.

And yup, of course his stuff is really well produced. Massive kudos to Melted Solids, the agency behind his social. 

But what’s as important as all those elements combined is how he leverages audience insights. More specifically, how he uses data to understand his audience, to understand their pains - the issues they’re struggling with - and to offer them the solutions they need. 

New Yorkers' big pain is being priced out of their own city. Zohran talks about other issues with remarkable authenticity too (notably Palestine), and he frames himself as the good guy underdog against sleazy opponents (Cuomo and Adams). But his core message is about making life better for his audience - about solving their problems.

This video says it all.

Lessons from behavioural science

"Pain point" marketing is a basic concept in business: People are highly focused on their own problems, and they’re motivated to engage with solutions. 

Most products are about solving problems for consumers. We buy new stuff to make life easier, smoother, more comfortable. 

And we buy stuff to try to solve the problem of loneliness, isolation and status. That’s aspirational marketing (which is kind of another category.) 

Look at any commercial through this lens, and you’ll see how it broadly fits into one or both of these two categories.

Why is advertising like this? Because it works. 

Advertising is behavioral science applied to the real world: it’s the study of which inputs (messages) produce which outputs (behavioural responses).

If it didn’t drive sales, brands wouldn’t do it.

Politics as product

Zohran’s agency, Melted Solids, are former product marketers who wanted to do something more with their lives than sell people stuff they don’t need. Good on them.

They’re applying behavioural science / advertising smarts to lift their client’s brand awareness and affinity. 

Zohrans’ playbook - connecting with his audience’s pains, framing himself as being on their side against a common enemy, talking their language, echoing their style - isn’t unique. 

Here’s the AfD’s Alice Weidel talking about the corruption, bureaucracy and the anti-democratic elitism of her opponents. 

Here’s Tommy Robinson’s, the  British far-right activist and anti-Islam campaigner, best known as the founder of the English Defence League - pain and anger and easy solutions. 

And here’s Trump’s jaw dropping Truth Social.

The same playbook - use pathos to tell stories that resonate with your audience - has been used since Plato.

The difference now is that algorithms segment automatically and effectively, to serve people stories they want to engage with. That’s squeezing out the old, less-media-savvy establishment. A quick look at the engagement data shows their stories aren't working.

Products solve problems. So should politics. But for a large section of the population in the US and Europe, politics as usual isn’t solving their problems. All they’re seeing is wage stagnation, price hikes, services going downhill. 

Polling of swing voters across Europe and the States echoes the perception that the establishment parties are just more of the same - it’s not working, so screw it, let’s roll the dice. 

Where to get audience data

Here’s a simple example of how to begin incorporating audience insights into messaging strategy.

I’m doing a fundraising project at the moment for an NGO that sends aid to Ukraine. So the first place I began when I started developing messaging is with the audience data on their social media. 

You can find this in Meta Adsmanager / Audience insights.

And then turning data into insights just takes a bit of logic, empathy and life experience. 

My first idea for this NGO was to try to connect with their audience’s fear of global instability, and try to motivate donations by connecting Putin’s expansionism with their own kids’ futures.

Looking at the data though, the audience is mainly into tabloid news, domestic causes like Breast Cancer, discount shops and comedians on the less cerebral end of the spectrum. 

It’s a less educated, working class audience that’s probably less focused on international issues. And looking at the engagement on the NGOs posts up until now, that hypothesis is born out in the data there too. 

So I moved away from the global narrative to other hypotheses: impact (what we’re doing is actually working), localism (people just like you are giving their time and money across the country), the ww2 parallels (the resistance and fighting the bad guys), and then a bunch of “inspiring” stories of protagonists overcoming adversity - stories with nice emotional payoffs, and good energy.

I’m looking to do the kind of stuff that goes down well with our target demo on social. And the key insight is that people on social media aren’t looking for information. They’re looking for entertainment - something to help them through the stresses and boredoms of their everyday life by giving them a few moments of new emotional energy. 

And then, after we’ve developed media assets out of these hypotheses, we’ll keep an eye on the engagement data and see what’s working and what’s not. And very simply we’ll do more of what works and less of what doesn’t. 

Everything is an experiment. We’ll get further insights from how people respond.

Connecting audience insights with strategy

One more example, this time with a quantifiable outcome. 

We ran a simple experiment for Action Aid Denmark, creating new messages based on audience insights to see how much more effective we could be at getting people to sign climate petitions. 

So we did some research, looking at their audience, competitors, the conversation and the culture around climate.

And the big insight we got was that although up to 88% of Danes are worried about the climate, very few engage with climate content. 

So why is that? Obviously it's because the current stories aren't working for them.

And that’s probably because most climate content just repeats the same climate emergency story that they’ve heard a million times before. 

We figured that if people are worried about something (a pain) but they’re not getting a solution, there’s a disconnect. So we came up with a few different hypotheses angled towards solutions. 

We turned those hypotheses into quick creative ideas, put them out there and got up to 27% cheaper ad costs per petition signature. 

Check out the more detailed case study here.

Democracy needs to catch up

This applied behavioural science (aka digital advertising) is what I learned on private sector projects with brands like Apple and Novo. It’s what Zohran’s team is definitely doing. 

And exactly the same techniques are being applied with devastating effect by very smart disinformation strategists in Moscow and St Petersburg. And by Tommy Robinson’s team and Charlie Kirk’s team and Trump’ team. 

The tools are apolitical. They’re just tools. The ethics comes in the ideas they're applied to.

And in 2025, the only places that aren’t using these tools, unfortunately, are the NGOs and the centrists and the democrats, the UN and the EU. And that’s a big part of why democracy - the thing they’re desperately trying to protect - is going downhill.

Other tools

You can get a long way with free tools for audience insights. 

• Google Trends shows the relative popularity of search terms over time, indicating public interest in specific topics.

  
  

• Answer the Public collates and visualises search engine queries, revealing the specific questions people are asking.

  
  

• More in Common has really nice open source polling 

  
  

• LLMs - trying asking “what are the pains of my INSERT TARGET AUDIENCE?”

And if you’ve got deep pockets you can pay for big data tools that’ll mostly tell you what you already knew or guessed.

Talkwalker, Brandwatch, SEMRush etc are all examples. But start with the free ones, learn how to use them and then expand when you reach the limits of what you can do before you think about paying.

Reach out if you want to talk about how you can use audience insights to increase the resonance of your org.

And if you're interested in more new media solutions to the political chaos that arrived with social media, follow us on LinkedIn.