What Berlin taught us about polling and power

Polling holds a particular kind of authority in the think tank world. It signals seriousness, suggests empirical grounding and offers the promise of clarity in an environment often defined by noise.

But as we explored at our recent WonkComms Breakfast Club in Berlin, polling is not inherently influential. Nor is it inherently responsible. It is a powerful instrument — and like most powerful instruments, its impact depends on the intent and discipline with which it is used.

Bringing together communications leaders from More in Common and the European Council on Foreign Relations (ECFR), the discussion moved beyond methodology into a more strategic question: what does it actually take for polling to shape public debate constructively?

The answers were less about survey design and more about narrative control, launch discipline and reputational risk.

Polling as a strategic intervention

The clearest message from Berlin was that polling should not begin with curiosity. It should begin with intent.

For More in Common, polling feeds a mission cycle: understand society as it is, reveal insights that challenge assumptions, and strengthen democratic culture as a result. 

ECFR described its polling as a way to surface political majorities that are often invisible in media narratives. In both cases, polling is not simply content. It is positioning.

That distinction matters for organisations with sustained publishing cadence and multiple stakeholder audiences. When polling sits outside the broader narrative architecture of an institution, it becomes episodic — a spike of attention disconnected from longer-term influence. When it is embedded in strategy, it reinforces and sharpens an organisation’s voice.

The starting point, then, is not “What would make an interesting chart?” but “What debate are we trying to shift — and what do we need to understand in order to shift it?”

The lens shapes the debate

Another important thread running through the session was the idea that polling’s influence lies less in volume and more in interpretation.

Large cross-national samples create credibility. But credibility alone does not change conversations. What shifts narratives is the lens through which findings are framed.

More in Common’s segmentation work — including the ‘Invisible Third’ — demonstrates how polling can move beyond simplistic binary framings of public opinion. By identifying attitudinal groups that cut across political lines, segmentation can reconfigure how issues are understood and discussed.

This is where polling becomes powerful. Not when it confirms polarisation, but when it complicates it.

For communications teams, that requires restraint and clarity. The most dramatic statistic is not always the most strategic one. The job is not to amplify noise, but to surface meaning.

Launch is choreography, not distribution

If polling is positioning, then launch is choreography.

ECFR reflected on years of pan-European polling and the coordination required to ensure findings travel across markets. That impact rests on a number of deliberate choices:

• Timing releases to political moments
• Aligning with relevant news hooks
• Securing trusted media partners
• Preparing materials well in advance
• Planning how insights can be revisited in the weeks that follow

None of this is incidental – it is structural to impact.

Design, too, plays a strategic role. A line that captured the room was simple but telling: if a chart needs explanation, it is too complex. In a compressed media environment, clarity determines uptake. Visuals must carry a single insight without sacrificing nuance, reducing friction, not adding to it.

For organisations managing complex publishing ecosystems and varied audience journeys , this often means translating the same dataset differently for journalists, policymakers, funders and specialist communities. That translation is not a technical afterthought; it is core communications work.

The structural risk of misuse

Perhaps the most sobering part of the discussion was the recognition that polling will not always travel as intended.

Speakers acknowledged examples where findings were reframed in national media contexts in ways that altered their emphasis . There was open discussion of the risk that data can be instrumentalised by actors seeking to reinforce polarising or extremist narratives .

This is not a rare occurrence. It is a feature of today’s information environment.

Publishing polling, therefore, is not simply about amplification. It is about governance. It requires pressure-testing headlines, anticipating points of distortion and being clear about limits and uncertainty. It demands thinking adversarially before others do.

For research organisations whose influence depends on long-term credibility across complex stakeholder ecosystems , this responsibility cannot be delegated.

Influence is a system

What emerged from Berlin was not a checklist for better surveys, but a reminder that influence is systemic.

The organisations that consistently make polling matter:

• Anchor it in mission and long-term positioning
• Identify the lens that reframes debate
• Invest in disciplined, well-timed launches
• Treat visual clarity as strategic, not decorative
• Anticipate risk as part of publication planning

Polling, on its own, does not create impact. It strengthens — or weakens — the communications architecture already in place.

Handled thoughtfully, it can reveal nuance in polarised debates and surface unexpected majorities. Handled carelessly, it can contribute to oversimplification or be absorbed into narratives that distort its intent.

The difference lies in strategic coherence.

Why this conversation matters

WonkComms was created to host exactly these kinds of discussions — grounded, candid exchanges among people responsible for translating research into public influence.

Berlin was a reminder that publishing data is not a neutral act but an intervention. It shapes how issues are understood, who is heard and which narratives gain legitimacy.

For those leading communications in think tanks and research institutes — particularly those navigating multi-market audiences and sustained publishing demands — polling is neither a silver bullet nor a side project. It is a high-stakes instrument.

The question is not whether to use it, but whether it is embedded within a communications system capable of carrying its weight.

If that is a conversation you want to be part of, join us at a future WonkComms session in London, Berlin or Washington DC.

Take a look at ECFR’s data portal here.