Pulse Labs

Can AI-generated survey responses tell us what real people actually think? Pew Research Center recently looked at how well AI-generated survey responses match the answers of real people. The findings are eye-opening: - AI answers were off by an average of 12 percentage points. - On nearly half of the questions, at least one response option was never selected by any AI respondent. - AI struggled to reflect the true diversity and nuance of real human opinion. It highlights an important question for research teams: Just because AI can generate plausible, well-written responses doesn’t mean those responses reflect how people actually think, feel, or behave. As synthetic respondents become more common in research conversations, it's worth looking closely at what they can tell us, and where real participant input still matters. Curious how others in research are thinking about this: What would you need to know before using synthetic respondents in a study? Read the report from Pew Research Center: https://lnkd.in/d7SY4j5g #UXResearch #MarketResearch #AI #PulseLab

Some of the most important research decisions happen before the study even begins. Take participant screening. The criteria you use to decide who qualifies can quietly shape the outcome of your research. Cast the net too wide, and the signal gets lost in the noise. Filter too tightly, and you risk screening out the very people who could challenge your assumptions. Recruit only your most engaged users, for example, and you may never hear why others drift away. It's always worth asking: are we screening for the people who can best answer our research question, or just the people most likely to confirm what we already believe? Getting that distinction right makes all the difference. What's the biggest trap you try to avoid when writing a participant screener? #UXResearch #MarketResearch #ParticipantRecruitment #PulseLabs

On Tuesday, we asked what makes content stand out in a world of endless scrolling. Practical usefulness came out on top. We also got an answer that wasn’t on the poll: “vibes.” And honestly, there’s probably something to that too. A useful test for content might be: Would someone learn something they can actually use? Would it help them solve a problem, make a decision, or do something better? Would they save it because they might need it again? If the answer is no, maybe it needs another look before it goes out. LinkedIn’s recent conversation around AI slop makes this even more relevant. When it’s easier to produce more content, the value of each piece becomes a little easier to question. So what makes content worth stopping for? Start with something useful. Add a little bit of vibes. #ContentStrategy #AI #UXResearch #ResearchOps #PulseLabs

AI is making it easier to create more content. But that doesn’t necessarily mean we need more of it. LinkedIn’s latest look at AI-generated content raises an interesting question about what makes something useful, original, and worth people’s attention. Read the full update from LinkedIn: https://lnkd.in/eHCKwSth

4 ways agencies can make an RFP response stand out When responding to a research RFP, more length rarely adds more value. What clients often want is clarity, flexibility, and a quick read on fit. Here are four adjustments that can make a proposal stronger: 1. Show your team, not just your firm. Clients are hiring people as much as a company. Introduce who will actually run the work, and why they’re right for the project. 2. Highlight domain depth, not just general capability. Being great at moderation isn’t the same as knowing how to talk to clinicians, business buyers, or kids. Lead with experience in the specific audience or industry. 3. Spell out your assumptions up front. Be clear about what’s in scope, who owns participant recruitment, and how timeline shifts are handled. Showing you’ve thought through the friction points builds trust early. 4. Frame deliverables around the decision, not just the method. Don’t just list methodology steps. Show how the final synthesis will help the client make the decision in front of them. The goal isn’t to say more. It’s to make it easy for the client to say yes. For agency leaders and research buyers: what’s one thing in an RFP response that immediately tells you a team “gets” it? #UXResearch #MarketResearch #ResearchAgencies #RFP

A little #PulseLabs team time in Rio 🇧🇷 Some of our Brazil-based team recently had the chance to spend a few days together. Karina Cavalcante, Maria Clara Rezende, Samira Pulatova and Luiza Goto got some time to connect in person, catch up, and enjoy the city together. It’s always nice seeing the people you work with every day in a new setting, especially when your team is spread across the globe. #GlobalTeam #TeamCulture #Collaboration #LifeAtPulseLabs #ResearchCommunity

As Gen Z takes on a larger role in the workplace, understanding what earns their trust and attention is becoming increasingly important. LinkedIn recently looked at how Gen Z is approaching professional relationships and how marketers are thinking about engaging this growing group of decision-makers. The findings offer an interesting look at how expectations around connection, credibility, and communication are evolving. Read the full research: https://lnkd.in/eRSnp4Gt #GenZ #GenZProfessionals #MarketResearch #ConsumerInsights #Research #Marketing

Research teams don’t necessarily need more tools. Sometimes they just need fewer places to look. 🔎 As research stacks expand, insights, vendors, panels, and workflows quickly end up scattered across fragmented systems. That friction slows teams down right when speed matters most. That’s the exact problem we’re tackling with Research Exchange: connecting your partners, tools, and workflow into one ecosystem. Want to see what we’re building to end this fragmentation? Explore Research Exchange: https://lnkd.in/gUrJ5T67 #UXResearch #UserResearch #ResearchOps #ProductDesign #PulseLabs

More and more, industry events are circling the exact same core dilemma: How do we leverage AI speed without sacrificing evidence quality? This week, uxcon vienna is tackling this shift head-on, featuring sessions on the future of UX research in the AI era, evidence quality, and where human research still matters. We’re keen to learn how teams are navigating this in practice, not just in theory. What’s one research task you’ve successfully delegated to AI, and one you’ll always keep strictly human? #UXResearch #UXConVienna #AIinResearch #ProductDesign #PulseLabs