Lauren O'Connor

They say part of their product marketing work is adapting that explanation for whoever is in the room: sales teams, marketing staff, or executives who have heard the terms but cannot yet connect them.

• Rather than present a personal framework, the post recommends Sanjeev Mohan's article as “the best explanation I've found,” describing it as a treatment of the terms as layers, not competitors.
• It also quotes a warning that a context layer on top of ungoverned metadata and inconsistent semantics is “just a faster way to be wrong.”

The post opens with the repeated question—“what's the difference between a semantic layer, a knowledge graph, and an ontology?”—then points readers to Sanjeev Mohan's stacked-layer explanation and highlights the quote about a context layer accelerating wrong answers when the foundations are not governed.

Read original

At the IDC CIO Summit, Lauren O'Connor asked how many definitions of revenue existed across the organizations represented in the room.

• Half the audience laughed—a reaction they read as painful recognition rather than humor.
• The roundtable's first question was how to sell data governance to a C-suite that has already bought an AI platform.
• Its stated throughline: “your AI is only as trusted as your data.”

The IDC CIO Summit recap describes executives at different points in their AI journeys facing the same root problem: fragmented, ungoverned data making AI unreliable at exactly the wrong moment.

Read original

Reporting from a Gartner event, they note there was no BI bake-off and treat that absence as a signal: the BI tool debate has not disappeared, but AI agents and infrastructure were the loudest topics.

• They quote a Gartner prediction that by 2030 universal semantic layers will be treated as critical infrastructure alongside data platforms and cybersecurity.
• They then describe the everyday friction of using AI agents for data questions: an exhausting back-and-forth just to agree on definitions before asking the real question.
• They call it a missing-foundation problem, not an AI problem.

The Gartner post moves from the quoted 2030 prediction to a concrete symptom—AI agents that do not know how the business defines anything—and concludes that the problem is a missing semantic foundation, pointing to a longer write-up.

Read original

They cite a background as an instructor and an analyst, and now in product marketing, and say an English and Spanish major is why words and definitions matter to them.

• The posts show that translation in practice: a terminology explainer for peers, a conference session and roundtable for executives, and commentary on analyst and vendor news for a general professional audience.
• Over this period, a short announcement of a new position as Vice President, Product Marketing at Strategy is followed by public representation of Strategy at industry events and commentary on Gartner and semantic-layer themes.
• Recent material focuses on AI readiness and semantic layers rather than the role announcement.

The terminology post says the challenge is explaining semantic layer, knowledge graph, ontology, and context layer differently for sales teams, marketing folks, and executives. The IDC and Gartner posts show the same translation method applied in live rooms and industry commentary.

An announcement of the new role appears early in the posts here. Later posts represent Strategy at the IDC CIO Summit and comment on Gartner and semantic-layer news, shifting the visible output toward AI readiness and data governance.

Read original · Original 2 · Original 3 · Original 4

Lauren O'Connor starts from the question they keep getting asked: what separates a semantic layer, a knowledge graph, and an ontology, and what a vendor means when it says "context layer."

• They describe the backgrounds they draw on to answer it, including work as an instructor, an analyst, and now product marketing, and say an English and Spanish major is why definitions matter to them.
• The post then does the practical part of the job it describes, adapting the same explanation for sales teams, marketing staff, and executives who have heard the terms but cannot connect them.
• Rather than presenting their own framework, they point to Sanjeev Mohan's article as "the best explanation I've found," describe it as treating the terms as layers of a stack instead of competitors, and quote the line that a context layer on top of ungoverned metadata and inconsistent semantics is "just a faster way to be wrong."

For brands planning a campaign

For a data infrastructure brand shaping product marketing messaging, this shows how a technical taxonomy becomes language for sales, marketing, and executives in one piece. It also shows a willingness to credit another person's framework rather than repackage it—a useful signal for a brand choosing who to put in front of its audiences.

It is the most compact statement of the thesis.

Reporting from a Gartner event, they note that there was no BI bake-off this year and read the absence as a signal, not a change in the underlying debate: the tool argument has not gone away, but it is no longer the loudest thing in the room, which was instead full of AI agents and infrastructure.

• The post quotes a prediction verbatim: by 2030, universal semantic layers will be treated as critical infrastructure alongside data platforms and cybersecurity.
• They connect that forecast to a recognizable symptom: without a semantic layer, an AI agent has no idea how the business defines anything, leaving users in "this exhausting back and forth just to get the definitions right before you can even ask the real question."
• They frame this as a missing-foundation problem rather than an AI problem and point to a longer write-up.

For brands planning a campaign

This post offers a reusable demand-generation hook for a semantic layer or data platform vendor: a third-party analyst forecast paired with a familiar pain, building a positioning narrative without the brand having to make the case from scratch. It also shows how they turn an industry event into a message, rather than simply report attendance.

It shows what they do beyond publishing and whom they do it for.

Recapping a session and roundtable at the IDC CIO Summit in New York, they describe asking the room how many definitions of revenue exist across their organizations.

• Half the audience laughed, which they read as recognition rather than humor.
• Their summary: attendees were at different points in their AI journeys but shared one root problem—fragmented and ungoverned data making AI unreliable at the wrong moment.
• The roundtable ran out of chairs.
• The first question raised was how to sell data governance to a C-suite that has already bought an AI platform, and discussion of the future org chart drew strong reactions.
• They close with the line that an organization's AI is only as trusted as its data.

For brands planning a campaign

Useful evidence for briefs involving executive events, roundtables, or sales enablement: it shows firsthand facilitation with senior buyers. The question of how to sell governance to a C-suite that has already bought an AI platform could anchor content or a webinar for a data governance vendor.

Rather than treating these as competing products, they present them as a stack in which each layer has a specific job.

• They argue that the hard part is explaining the stack differently to sales teams, marketing colleagues, and executives who have heard the terms but cannot yet connect them.
• That framing gives a brand a ready-made structure for defining its category language and training internal teams on which term to use with each audience.

Publishes a short, audience-calibrated explainer that names the confusion, then resolves it with a layered model rather than a product pitch.

For brands planning a campaign

Strong fit for a messaging and positioning project that needs to distinguish adjacent category terms and give sales and executive audiences different versions of the same story.

Examples: Source 1

At the IDC CIO Summit, they asked a room of executives how many definitions of revenue existed inside their organisations.

• The response was recognition rather than surprise.
• Their conclusion is that fragmented, ungoverned data makes AI unreliable at exactly the wrong moment, and that a context layer on top of ungoverned metadata is just a faster way to be wrong.

Starts with an observation from the room, asks the audience a diagnostic question, then reports the reaction to make the problem feel familiar before making the argument.

For evidence of performance, ask for customer results or deployment examples; Lauren O'Connor's posts report neither.

For brands planning a campaign

Useful for a governance or data-quality offering that must justify itself to buyers who have already purchased an AI platform and believe the governance problem is solved.

Examples: Source 1 · Source 2

Their account of using AI agents to answer data questions describes an exhausting back-and-forth to agree on definitions before asking the real question.

• They call this a missing-foundation problem, not an AI problem.
• It is an argument a vendor could use to move a semantic layer conversation out of feature comparison and into architecture planning.

Anchors the argument in an observed industry shift, quotes an external prediction, then grounds it in the everyday friction of agreeing on definitions before analysis.

For brands planning a campaign

Suited to an awareness or category-education campaign aimed at buyers evaluating AI agents who have not yet framed semantic layers as a prerequisite.

Examples: Source 1

They cite their background as an instructor and analyst, and earlier as an English and Spanish major, as the reason definitions matter to them.

• They describe product marketing as figuring out how to pitch the same concept to whoever is in the room.
• Their posts put that into practice: a terminology explainer for peers, a conference session and roundtable for executives, and commentary on industry announcements for a general professional audience.

For brands planning a campaign

Relevant when the goal is understanding rather than reach—for example, enabling a sales team, briefing executives, or producing category education a prospect can follow without a deep technical background.

Examples: Source 1 · Source 2 · Source 3

They also show a shift over the period covered: an announcement of a new position as Vice President, Product Marketing at Strategy, followed by public representation of Strategy at industry events and in commentary on vendor and analyst news. Recent posts focus on AI readiness and semantic layer themes rather than the role announcement itself.

Uses a consistent signature line—your AI is only as trusted as your data—to connect posts, talks, and longer write-ups.

Ask to review the two linked longer articles before commissioning work; they are not reproduced in full here.

For brands planning a campaign

Expect a mix of on-stage, written, and social formats. Brief them on current AI-and-data themes rather than on background or career narrative.

Examples: Source 1 · Source 2 · Source 3

The posts present Strategy as their employer, not a client or sponsor.

View example

The post describes a speaking appearance.

View example

Gartner is a research source they react to.

View example

They use structured explanatory posts to identify a foundation problem before discussing the technology.

Bidgood grounds the argument in consumer-goods commercial operations, including trade spend, forecasts, retail media, and product hierarchies. O'Connor draws on data-infrastructure vocabulary and product marketing translation, using conference rooms and analyst predictions as examples.

Read original · Original 2 · Original 3 · Original 4

The posts share O'Connor's foundation-first framing and LinkedIn explainer format.

Trebels writes from a founder and partner perspective, promoting a specific knowledge-extraction product and drawing on event takeaways framed around climate and sustainability. O'Connor writes as a product marketing translator, explaining category terms and industry shifts.

Read original · Original 2 · Original 3 · Original 4

O'Connor's posts emphasize definitions as a precondition for trustworthy AI.

Lancellotti focuses on ERP consolidation, customer success, and CDP-driven account operations, often laying out architecture recommendations for transformation leaders. O'Connor explains overlapping data terms and positions them for sales, marketing, and executive audiences.

Read original · Original 2 · Original 3 · Original 4

His posts carry the same underlying argument O'Connor makes — that AI fails on inconsistent definitions, not on models — into operating language a commercial audience already uses: trade spend optimized against the wrong signals, forecasts that are "confidently wrong," retail media ROI that cannot be trusted, and agents that work technically but miss business intent.

• He treats trusted data as a boardroom concern rather than an IT initiative, the right framing when the buyer's C-suite has already bought an AI platform and what remains to be sold is agreement on what the words mean.
• Choose him when the message has to land with marketing, supply chain, and finance leaders inside a manufacturer or retailer, and a short, pointed post naming a specific business failure fits better than a layered explainer.

The tradeoff is range and depth.

• Bidgood's posts are narrower in both industry and format: they focus on consumer goods, using examples like promotions, margin, and retail execution, and work best when the audience already knows that vocabulary.
• A general data infrastructure vendor selling a semantic layer to technology and data teams would find the terminology work thin: Bidgood asserts that definitions break rather than teaching the difference between a semantic layer, a knowledge graph, and an ontology the way O'Connor does.
• He also writes less for multiple audiences at once; his posts address senior business decision-makers, not sales, marketing, and executives with three versions of one message.

Examples: Source 1 · Source 2

The posts describe a knowledge extraction pipeline that builds semantic knowledge graphs from filings, clinical reports, product manuals, and compliance records, and argue explicitly that the missing piece is a semantic layer between raw information and machine comprehension.

• That is useful positioning for a vendor that has to explain auditability, traceability, and deployment control: Trebels names on-premises, AWS, and Azure options and an LLM-agnostic posture, and points to regulated sectors from financial services to government.
• Choose Trebels when the message must answer "how do we defend this to a regulator and keep control of our infrastructure" rather than "how do we explain this category to a sales team."

The tradeoff is independence and cross-audience reach.

• Many of the posts promote a specific partner product and its architecture, so they read as vendor-adjacent explanations rather than neutral category teaching.
• A company hiring for positioning would need to separate the reusable argument from the product pitch.
• Trebels also writes for a technical, regulated-industry reader.
• If the message must work for executives and sales, ask for versions tailored to each audience.
• Where O'Connor starts from the questions people keep asking about the terms themselves, Trebels starts from a problem the buyer is already paying to solve.

Examples: Source 1 · Source 2

The posts make a structured argument that AI is replacing the workflow layer ERP was expected to own, and that ERP and CRM's durable role is as the system of record and compliance backbone, with a semantic layer in the middle.

• For a vendor whose buyer is a CIO, CTO, or transformation leader, that is a ready-made executive narrative, ending in concrete architecture recommendations: consolidate the system of record, stop building workflows inside ERP, build an orchestration layer on top, and invest in master data and semantic modeling.
• A second post adds a customer-operations angle, using connected finance, service, and CRM data to serve accounts too small for human coverage.
• That is useful when the product's value is framed as operational rather than definitional.

The tradeoff is distance from the terminology problem itself.

• These posts come from inside large-scale transformation programs, so the vocabulary is ERP consolidation, purchase-to-pay, and CDP integration; the semantic layer appears as one component of an architecture rather than the thing being explained.
• A vendor whose main job is to define semantic layer, knowledge graph, ontology, and context layer for a mixed audience would get an executive architecture argument, but not plain-language definitions.
• The posts are also long and dense, suiting transformation leaders better than teams looking for a short, reusable explainer for sales and marketing colleagues.

Examples: Source 1 · Source 2