Research Capacity Is Rising, but Decision Windows Are Shrinking

Maze’s latest State of Market Research Report describes a familiar tension in unusually practical terms. Research teams are being asked to cover more ground, respond more quickly and connect a wider mix of evidence, while budget and headcount fail to keep pace.

Its survey of more than 200 people involved in market research found that respondents had taken on an average of 4.5 additional responsibilities over the previous two years. Market research, customer insight, analytics, competitive intelligence, strategy, AI governance and tool management are increasingly arriving in the same job description.

That is not simply a workload problem. It changes the design of the insight function.

The risk is that teams become highly efficient at processing requests, but less able to shape the decisions that generated those requests. Research may be faster, more visible and more broadly used, yet still arrive after the commercial direction has hardened.

The issue is not simply workload

Maze reports that 70% of participants had seen research requests increase and 78% had experienced faster delivery expectations over two years. Meanwhile, 57% said their team lacked the resources to meet current demand. The research is not a representative census of the UK profession — its sample is weighted towards North America — but the pattern will resonate with many in-house insight leaders.

A request for “customer research” now often contains several different questions. Is there demand in the market? Which segment matters? What do existing customers struggle with? Is the proposition credible? Which competitor has set the reference point? What evidence will persuade sales, product and marketing to make the same call?

Those are related questions, but they do not require the same evidence or the same timing. Combining them under one broad brief can create work that is ambitious but indecisive: a large study, several stakeholder groups, a long list of findings and no clear point at which a choice can be made.

The expanding role of insight should therefore not lead organisations to expect every researcher to become a universal substitute for product, strategy, analytics and commercial teams. Its value lies in joining up evidence around a decision, including where the evidence conflicts and where uncertainty remains.

That is a more demanding role than fielding a study. It requires teams to be clear about the decision owner, the choices genuinely available and the consequence of being wrong.

Faster work can miss the moment that matters

The most revealing finding in the Maze study concerns timing. While 89% of respondents said research was viewed as a strategic partner, only 33% said it typically entered when a problem or opportunity was first being defined. A further 27% said it came in while options were being explored.

That leaves a narrow period in which insight can alter the brief, not merely validate a preferred answer.

This matters especially in B2B organisations, where apparent decisions often begin long before a formal project appears. A sales team identifies a recurring objection. Product sees an adjacent use case. Marketing wants a new positioning platform. Leadership notices a competitor’s claim. By the time someone asks for research, the organisation may already have selected the problem it wants evidence to solve.

The best response is not to insist that every emerging issue requires a full programme. It is to create a lighter route into the earliest stage: a short evidence review, a decision-framing session, a rapid set of customer conversations, or a check against existing segmentation, win/loss and behavioural data.

The question at that stage is not “what should we ask customers?” It is “what would we do differently if this assumption were wrong?” If the answer is nothing, the request may not yet justify new research. If the answer is a different target market, proposition, investment level or route to market, the timing is probably right.

Treat the request queue as a portfolio

More capacity is arriving through AI. Maze found that 84% of participants believed it increased research capacity, with teams using it across planning, synthesis, reporting and knowledge management. But only 17% said AI was integrated across the research process with clear ownership, and 96% agreed that outputs required review.

That gap should not be treated as a narrow technology-policy concern. It is a question of where newly created capacity goes.

It can be spent producing more summaries, more rapid concept reactions and more stakeholder-ready slides. Or it can be directed towards the work that tends to be crowded out: clarifying the real decision, locating prior evidence, identifying the customer groups missing from a brief and testing the assumptions that have become accepted too quickly.

A useful operating model separates requests into three types.

First, there are decisions that need fresh evidence: a new market entry, material repositioning, pricing move or major proposition choice. These deserve a defined learning plan and appropriate methodological depth.

Second, there are decisions that need existing evidence made usable. Here the task may be synthesis, retrieval and interpretation rather than another survey.

Third, there are requests for reassurance after a direction has effectively been chosen. These can still have value, but teams should label them honestly as validation or optimisation work rather than early-stage discovery.

That distinction protects research from being judged against an impossible standard. It also makes the queue easier to manage, because the urgency of a request is no longer confused with its likely influence.

Build an evidence base that can be reused

The pressure on insight teams will not ease simply because more tools become available. Voxpopme and the Market Research Institute International have just placed research automation, data quality, the gap between stated and observed behaviour, and demonstrable impact at the centre of their new State of Insights study. The agenda reflects a profession trying to make faster work more dependable, not merely more abundant.

The practical answer is to make each project contribute to a shared commercial understanding. That means recording not just findings, but the decision addressed, audience covered, methods used, confidence limits, relevant segments and what later happened.

This is different from creating a bigger archive. An archive stores outputs. A working evidence base helps a team recognise what it already knows, what has changed and what must be tested before the next decision closes.

Research capacity is rising. Its commercial value will depend on whether insight teams use that capacity to get upstream of decisions — while organisations still have a meaningful choice to make.