April 2020 ARVIC

The April 2020 ARVIC virtual conference convened insights and product professionals at the sharpest inflection point of the early pandemic. Talks ranged from toilet paper hoarding studies and real-time media tracking to behavioral economics, neuroscience wearables, and research policy at scale. Beneath the variety, three concerns ran through nearly every session: how to keep research credible under pressure, how to translate findings into decisions, and how organizations should adapt their tools and processes when the ground shifts fast.

COVID-19 as a Pressure Test for Research Practice

Several talks used the pandemic as a live case study for what good research practice looks like under stress. A qualitative study on toilet paper hoarding demonstrated that fast, conversation-based discovery methods can surface emotional drivers, comfort-seeking, fear of social exclusion, and desire for control, that structured surveys miss entirely. A real-time media insights session showed that consumer behavior shifted in days, not weeks, and that teams with existing processes and data relationships were able to reprioritize quickly while others hesitated. A shopper insights talk made the straightforward recommendation to pause standard behavioral research during periods of abnormal consumer behavior, because pantry-loading dynamics make the data non-comparable to baselines.

• Fast, light-touch discovery methods are well suited to pre-discovery phases when you do not yet know what questions to ask.
• Reprioritization requires relationships and processes already in place, not improvised ones built during a crisis.
• Know when not to field. Behavioral data collected during abnormal periods can mislead more than it informs.

Rigor, Bias, and the Limits of What Data Can Tell You

A recurring theme across methodologically focused talks was the gap between what research outputs appear to say and what they can reliably support. A session on A/B testing argued that experimental statistics describe the past only and that treating two-week results as predictive of long-term outcomes leads to bad product decisions. A behavioral economics session demonstrated live, using audience experiments, how anchoring, decoy effects, and loss aversion systematically distort survey responses. A neuroscience talk showed that self-reported liking predicts actual behavior only 15 to 30 percent of the time, while neurological immersion measures consistently outperform it.

• Audit question order and numeric context in surveys for unintended anchoring before fielding.
• Reframe A/B experimentation as a quality gate and risk-reduction mechanism rather than a revenue forecasting tool.
• Most survey designs evoke deliberate, effortful thinking; most purchase decisions do not. Closing that gap is a design problem, not just an analysis problem.

Translating Findings into Decisions

Multiple talks addressed the persistent problem of research that is methodologically sound but organizationally inert. An analytics leader from a major consumer goods company presented a practical blueprint: know your audience before you build your content, name the opportunity cost explicitly, tell a specific consumer story rather than leading with aggregates, and make the path to yes visibly low-risk. A personas case study showed how consolidating prior research into a shared foundation before fielding, and involving creative and commercial teams early, produced segments that were immediately actionable across social, retail, and web strategy.

• The most durable influence is guiding decision-makers to a conclusion they feel they reached themselves.
• Aggregate data alone rarely moves a room. A named individual with real challenges gives leaders something concrete to empathize with.
• Major decisions often require the same message delivered multiple times and in different contexts before it lands. Timing is a real variable.

Ethics, Privacy, and Policy as Operating Conditions

Three talks treated privacy and policy not as compliance checklists but as active design considerations. A data privacy session argued that the pandemic had accelerated personal data collection to an all-time high and that the research industry must recommit to privacy-by-design principles before regulators eliminate traditional professional exemptions. A talk on research policy at a large technology company showed that framing policy as an enabler rather than a blocker, and embedding it into every phase of the research lifecycle, produces more consistent compliance than treating it as a separate step. A UX research session covering regulated industries made the case that identifying legal and organizational constraints before starting a project, not after, shapes better and faster research plans.

• Seventy percent of people are more likely to share data when collectors clearly explain why it is needed and how it will be used.
• Brands that buy research have a role to play: preferring suppliers that subscribe to a recognized code of ethics reinforces good practice across the industry.
• In regulated environments, documentation is both a legal shield and a working discipline. Audits are real, and the cost of non-compliance is concrete.

Remote Work, Remote Research, and What Sticks

Talks on remote work and remote qualitative methods shared a common argument: the shift was already underway, COVID-19 simply accelerated it, and the habits formed now are likely to persist. Practical guidance covered dedicated physical workspace, synchronous versus asynchronous tool selection, and the deliberate manufacture of human connection in distributed teams. On the research side, a senior design researcher at an automotive company shared rapport-building protocols for remote interviews and noted that online settings make it easier for participants to discuss sensitive topics and give candid negative feedback, a feature, not a limitation. Video highlights and testimonials in deliverables were identified as an emerging expectation that will outlast the crisis.

• Set up a dedicated physical workspace before solving for tools or meeting schedules.
• Distinguish synchronous from asynchronous work and match the right tool to each type. Using one tool for everything recreates the interruption patterns of an open office.
• Remote interview formats lower the barrier for participants to share sensitive or negative feedback. Design for that, rather than trying to replicate in-person conditions exactly.

Organizational Structure and the Outsourcing Question

Two talks addressed how insights teams should structure their work and partnerships. A framework from an insurance company mapped four project factors, business decision scale, methodology complexity, available tools, and recruit difficulty, onto a simple matrix to guide the DIY versus outsourcing choice. The consistent recommendation was to let the business decision drive the methodology and the sourcing choice, not budget pressure or available tools. A separate talk on UX research in regulated industries reinforced this: vendors are subject to the same regulatory requirements as in-house teams and must be vetted accordingly.

• Simple decision, simple methodology: strong DIY candidate. Complex decision, complex methodology: bring in a full-service partner.
• Hybrid approaches are common and valid. Moderating in-house while outsourcing recruit, or using a vendor tool for programming while handling analysis internally, are both reasonable configurations.
• When evaluating strategic research partners, prioritize transparency, full-service capability across qual and quant, and a willingness to set realistic expectations.