Adapted from Duane Forrester's essay on Duane Forrester Decodes. Aurora extends his framing for marketing teams running AI-search visibility programs.
Most SEO and GEO teams do not fail at AI search because they lack vision. Roughly 70% of enterprise SEO teams now understand intellectually that AI-driven changes are necessary. Only about 30% have restructured roles to match. The 40-point gap between knowing and building the structure that makes it real is where the actual failure happens, and it is a change-management problem, not a strategy problem.
Forrester names three recurring stall patterns. Each is fixable, and each requires a structural commitment most teams skip.
The three stall patterns
Analysis paralysis. Teams accumulate research, build internal cases, monitor platform behavior, and never commit to a starting point because the platforms keep moving. Waiting for stability is not diligence. It is avoidance dressed up. The platforms will not stop moving. The teams that get value out of AI visibility work commit early to a starting point and revise on a cadence.
Pilot purgatory. A 2026 survey of 200 US marketing leaders found 82% of teams using AI for campaigns remain in pilot or experimental mode. 61% deploy AI only at the individual level rather than integrating it into collaborative workflows. These pilots rarely fail cleanly — they simply never graduate to production. The blocker is structural: pilots without budget, headcount, or KPI redesign cannot become operating disciplines.
Reorg fatigue. Teams that have lived through repeated transformation announcements develop scar tissue. The next announcement gets met with skepticism unless leadership demonstrates structural commitment — budget moved, headcount allocated, KPIs rewritten. Slide decks announcing transformation no longer move teams that have seen the previous five.
The four resistance patterns
Forrester maps four distinct resistance types, each requiring a different response.
Seniority-based. Senior practitioners' skepticism is often legitimate pattern recognition and appropriate vendor-hype defense. Dismissing experience burns the team's most valuable asset. The right move is to frame the transition as additive — the fundamentals of relevance and trust do not disappear in AI search, they compound. Senior practitioners who bridge this gap become accelerants.
Skills-based. This is knowledge or ability gap, not motivation. The ADKAR change-management model — Awareness, Desire, Knowledge, Ability, Reinforcement — diagnoses it cleanly. Treating skills anxiety as resistance wastes resources and confirms team members' fears that leadership misreads their actual challenges.
Political. When AI visibility expands SEO's scope into retrieval architecture and cross-functional coordination, budget ownership becomes contested between marketing, IT, and content. This resistance surfaces indirectly — slow approvals, ambiguous priorities — rather than as direct opposition. Resolution requires explicit ownership decisions, not hoping collaboration produces clarity.
Legitimate skepticism. When team members ask for revenue connections, they deserve honest responses. Manufacturing certainty damages credibility faster than acknowledging measurement gaps while demonstrating directional progress. Document methodology consistently from the start so you are building a proprietary baseline as standards emerge.
Running both operations at once
Most organizations cannot switch in a single restructuring cycle. The realistic near-term pattern is parallel operations — traditional SEO continues while AI capabilities develop alongside it. The most common version is also the one most likely to fail: existing SEO gets handed AEO/GEO responsibilities alongside current work, budgets do not expand to match, and the team figures it out. That setup guarantees pilot purgatory.
Two operating principles matter during parallel periods. First, not every traditional SEO activity needs equal intensity. Technical hygiene, crawl accessibility, and structured data are infrastructure both disciplines depend on; do not deprioritize them. High-volume tactical content production is the capacity that can be reallocated to AI-era work without meaningful current risk. Second, the AI visibility workstream needs dedicated ownership. Work that lives in everyone's job description at the margin of their other responsibilities does not graduate from pilot.
Sequencing role transitions
Attempting simultaneous restructuring manufactures reorg fatigue. Phased sequencing reduces disruption and builds momentum.
- Content strategists transition first — shortest conceptual bridge, high upskilling potential, low new-hire dependency. Early wins build credibility for later phases.
- Technical SEOs transition next — vector index hygiene, structured data beyond standard schema, AI bot crawl management. Not every practitioner will pursue this; the upskill-vs-hire decision is consequential here.
- New roles get named — AI visibility analyst monitoring retrieval inclusion and brand representation, and someone owning machine-facing content architecture. These often start as partial responsibilities before justifying dedicated headcount.
- Reporting and metrics restructure last. Teams accountable for AI visibility outcomes but evaluated entirely on traditional organic traffic produce compliance theater. Design this phase during phase one and communicate it clearly so teams understand the trajectory.
The 90-day transition scorecard
Forrester proposes leading and lagging indicators that measure the transition separately from the visibility outcomes it is meant to improve. The leading indicators:
- At least one role with formal AI visibility responsibilities
- A named owner for the dual operating model
- At least two active retrieval experiments generating learning data
- Completed skills-gap assessments for every team member against phase-three role definitions
The lagging indicators connect to outcomes: brand citation share in AI-generated responses, retrieval inclusion rates across engines, accuracy of brand representation when content surfaces. Aurora is built to feed the lagging side — per-prompt citation rates, per-engine visibility scores, source intelligence rollups, competitor exposure deltas — but the leading indicators are organizational and have to be measured separately.
The actual differentiator
The organizations that navigate this transition successfully will not be the ones with the clearest vision of what AI search requires. They will be the ones that converted that vision into structure: named owners, phased timelines, honest skills assessments, and measurement that tracks the work before it tracks the outcomes. The teams winning eighteen months from now are the ones building that structure right now.
