Marketers urged to replace rigid buyer journeys with context-aware decisions
Marketing automation can route a lead, trigger an alert or deliver a requested asset in seconds. But speed alone does not make an interaction useful. At the November 2026 MarTech Conference, a panel on customer engagement argued that teams should move beyond fixed sequences and make decisions based on what buyers need, who is involved and whether the available data is reliable.
The discussion, “From journeys to decisions: The next era of customer engagement,” was moderated by Brianna Miller, director of demand generation at Cohere Health. Panelists were Tim Boylan, owner of Grow-Wire Consulting LLC; Carlos Hidalgo, CEO of Carloshidalgo.co; Mehul Nagrani, CEO of Integrate + CaliberMind; and Jodie Wilson, a freelance lifecycle and marketing technology architect at Jodie Wilson Consulting.
Account for nonlinear buying
Panelists said conventional journey maps can impose a tidy sequence on buying processes that are anything but orderly. Enterprise purchases involve committees, changing priorities and multiple stakeholders, with evaluation often unfolding through overlapping conversations rather than a predictable progression.
Nagrani compared enterprise buying to soccer, where many fluid touches may precede a scoring opportunity. He contrasted that with go-to-market processes that resemble football, moving accounts through predefined stages and handoffs. The analogy points to a practical challenge for marketing and sales teams: internal pipeline rules may not reflect how a buying group is actually weighing a decision.
Hidalgo also questioned the reliance on campaign calendars as the organizing principle for engagement. Customers do not necessarily make decisions on a campaign schedule, he said. Teams need to sustain relevant communication across the evaluation period, adapting context for different stakeholders rather than assuming a single campaign will fit everyone.
Use automation where immediacy helps
The panel did not argue for making every interaction instantaneous. Instead, speakers distinguished between straightforward requests, where automation can reduce friction, and situations where judgment or relationship-building calls for a human response.
Nagrani said a prospect requesting standard product specifications or documentation should not have to pass through an unnecessary sales-development qualification step before receiving it. Delivering the requested material promptly respects the buyer’s time. More nuanced conversations, by contrast, may benefit from a considered human response rather than an automated action.
Boylan applied a similar principle to direct-to-consumer service. When a customer is struggling with a chatbot, automation should help identify that the interaction is not working and connect the person with a human specialist. The point is not to use technology to conceal service problems, but to recognize when a different form of help is needed.
Wilson described predictive models as useful for surfacing topics, channels and timing that may be relevant. She also stressed that marketing teams should retain control of business logic and brand standards. In that model, machine learning informs decisions without taking over the rules that shape the customer experience.
Interpret signals in context
Activity data can be misleading when treated as a universal measure of purchase readiness. A spike in page views or content downloads does not, by itself, establish that an account is ready for a sales push. Hidalgo said teams need to consider which person is engaging and what that person may be responsible for. A CFO’s interest could relate to budget protection or compliance, while an operations leader’s activity may point to workflow challenges.
Wilson cautioned that apparently conflicting research behavior may have a reasonable explanation: one contact could be investigating solutions for separate business units. Nagrani added that generative tools and automation can increase the volume of low-intent digital activity. The task for marketers is to distinguish meaningful movement in an account from surface-level activity, rather than treating every signal as equivalent.
Repair the data behind decisions
Panelists made data quality a prerequisite for effective decisioning. Wilson urged teams to fix their data before changing their technology stack. Nagrani described experiments in which AI produced polished performance dashboards that did not hold up when the underlying data was audited. Automating decisions on top of inaccurate inputs, the discussion suggested, can scale errors rather than improve outcomes.
Boylan offered an example involving a rental car company: SMS opt-ins had been routed to incorrect fields, preventing willing subscribers from receiving important updates. Adding AI to that setup would not have corrected the underlying problem. He recommended starting with one or two significant customer-lifecycle inflection points and cleaning the data tied to them, rather than attempting an all-at-once overhaul.
Nagrani suggested examining a single stalled opportunity: document the available data, the actions taken and the assumptions made, then assess the people and tools needed to understand why progress stopped. The exercise can expose operational friction. Hidalgo recommended combining technical audits with input from sales, customer success and buyers. Together, the panel’s advice was to make relevance a product of sound information, buyer context and deliberate judgment—not automation speed alone.
