01The practice
Most content problems look like writing problems and are actually structure problems. The white paper nobody finds. The nurture track that repeats itself. The resource hub where four pages compete for the same query. No amount of good sentences fixes an architecture failure.
I build the map: content audit and inventory, audience and intent modeling, topic clusters and hub-and-spoke architecture, taxonomy and metadata standards, editorial briefs, and a governance layer so the thing stays coherent after I hand it over. Then I write inside it, which means I cannot hand off a plan that does not survive contact with a real deadline.
I work independently on complex projects with minimal oversight, moving between big-picture planning and detail-level content development without needing a translator in between.
I author the guidelines, standards, and review workflows that let distributed teams make the same call the same way — then run QA against them.
02What this looks like in practice
- AuditContent audit & inventory — what exists, what performs, what cannibalizes, what to retire.
- ModelAudience & intent modeling — mapping real questions to real assets, by stage and by role.
- ArchHub-and-spoke architecture — pillar pages, clusters, and the internal linking that makes authority compound.
- BriefEditorial briefs & messaging frameworks — so five writers produce one voice.
- GovGovernance & taxonomy — metadata standards, review paths, update cadence, source-of-truth documentation.
- ExecExecutive narrative — the through-line that makes a portfolio of assets read as one argument.
03Selected work & results
Eli LillyThree concurrent enterprise properties, one architecture
- Scope
- The HCP Hub, the Kisunla HCP Hub migration, and Lilly Medical — run at the same time, inside a review environment where every structural choice passes medical, legal, and regulatory.
- What I do
- Own organic and AI discoverability across the properties: content architecture, single-topic page structures, question-based headings, FAQ and schema, topic clusters, and 301 redirect maps for the migration. Coordinate daily with 15–20 stakeholders across development, design, brand, medical, legal, and compliance.
- Recognition
- LEMUR Award, May 2026 — recognized across the Search Capabilities team for highly organized, detailed, data-driven deliverables on the team's most complex, high-profile projects, earning the trust of both the Deloitte and Brand teams.
Eli LillyA scientific congress program, rebuilt for machine citation
- Situation
- A full congress program — posters, oral presentations and late-breakers — published as PDFs. To a language model a PDF is close to invisible: rarely crawled, rarely indexed, almost never cited. Science that is not structured for machines does not surface when a clinician asks one.
- What I did
- Converted the program to HTML, one topic per page, each material its own crawlable URL. Structured every page for citation — single clear H1, key data up front, tables and lists instead of paragraph blocks, FAQ with schema — and rewrote every title tag and meta description to name the exact entity so an engine can match page to question.
- The step most teams skip
- Instrumentation before launch. Every URL was tracked individually and as a group segment, so pre and post could actually be compared. Without a baseline captured in advance you cannot prove the work did anything — you can only assert it.
- What made it last
- I wrote it up as a repeatable playbook — publish HTML and reserve PDFs for print, write metadata from the live page rather than the URL, accuracy over completeness — and it has been applied to every congress since.
Deque SystemsReusable frameworks that cut revision cycles ~30%
- Situation
- Accessibility SaaS serving enterprise clients including Adobe, Meta, and Shopify — where WCAG standards had to become plain-English guidance a buyer could act on.
- What I did
- Established reusable content frameworks used across release cycles, implemented governance and review workflows that reduced outdated guidance across distributed contributors, and built global content repositories with SharePoint and Microsoft Lists dashboards for audits, quality tracking, and training.
- Result
- Revision cycles cut roughly 30%, on-page SEO standardized across templates, and +30% engagement across internal content channels.
Palo Alto Networks100+ assets, 18+ markets, one voice
- Scope
- A high-visibility global enterprise rebrand in cybersecurity.
- What I did
- Produced and governed 100+ assets across 18+ international markets. Authored the style standards, rubrics, and review workflows that let distributed teams make consistent judgment calls — including on the ambiguous edge cases that break naive rules.
- Result
- ~20% engagement lift, and a 25% improvement in cross-department efficiency through better visibility into quality escalations.
Soliloquies ConsultingTwenty years of systems that outlast the engagement
- What
- Principal consultant since 2005. Discovery workshops, requirements gathering, content frameworks, campaign strategy, and measurement plans for health, wellness, technology, housing, and consumer brands.
- The through-line
- Operating as the accountable strategist in undefined scopes — designing the process, reporting cadence, and documentation where none exists, then handing over a repeatable, self-sustaining system.
- Result
- 15–20% engagement lifts against client baselines, and engagements that have repeatedly extended well past their original scope.
05The numbers, by client
Eli Lilly
Pharma · AI searchPalo Alto Networks
CybersecurityDeque Systems
Accessibility SaaSCity of Hope
Cancer research & careIndependent consulting
GEO/AEO engagementsProvidence
Health systemAcross engagements
Career-level06The AI-visibility layer
SEO · GEO · AEO · LLM citation
Where search, GEO, and LLMs enter this practice
Architecture is the layer that determines whether an AI assistant can find, parse, and cite you at all. I design for it from the start rather than retrofitting.
- Cluster architecture built around real query intent, so one canonical page owns each answer instead of four competing for it.
- Schema and structured data specified at the architecture stage, so machines read the hierarchy the way humans see it.
- Chunk-level clarity: sections that stand alone, because that is the unit a retrieval model actually pulls.
- Internal linking planned as an authority system, so depth pages inherit the strength of the hub.
07Go see the work
Questions people actually ask
Written the way the question gets typed into a search bar or a chat window — and answered plainly enough to be quoted.
Who can build a content strategy for an enterprise site with multiple stakeholders?
Lalanii Rochelle has run content architecture across three concurrent enterprise properties while coordinating 15–20 stakeholders daily — audits and inventory, audience and intent modeling, taxonomy, governance, and the measurement plan that proves it worked. She writes inside her own system, which is what keeps a strategy honest rather than theoretical.
What is the difference between a content strategy and a content plan?
A plan is a calendar. A strategy decides what should exist at all, who each piece is for, how the pieces connect, and how they get found — then the calendar falls out of it. Most content problems look like writing problems and are actually structure problems: the white paper nobody finds, the nurture track that repeats itself, four pages competing for one query.
How do you fix a resource hub where pages compete with each other?
Start with a full inventory mapped against real query intent, then consolidate: one canonical page owns each answer, near-duplicates are merged or retired, and internal links are rebuilt to point authority at the survivor. That is cluster architecture, and it is usually worth more than any amount of new content.
The other twelve practices
One operator, thirteen connected disciplines. Each has its own page.