AI search and discovery that connects people to the products, answers, and knowledge they need.
San Francisco, CA
Joined January 2009
- Tweets7.9K
- Following2.1K
- Followers6.9K
- Likes1.4K
AI agents don't fail on models. They fail on context.
Too little, and they guess. Too much, and they lose the thread. The wrong context, and they overshare.
Eight weeks on getting it right starts now: na2.hubs.ly/H08yFPQ0 #AIagents
ALT Navy graphic with the headline “AI agents don't fail on models. They fail on context.” and the line “What an agent is shown decides what it says.” A cluster of blue, teal, and pink dots fills the lower right corner. Lucidworks logo. A band along the bottom reads: What Lucidworks does. Search and AI context that make enterprise agents trustworthy. lucidworks.com.
In a regulated industry, the citation is the product.
Fluency is table stakes. Traceability is what makes it deployable.
The fix for a confident wrong answer isn't a bigger model. It's grounding, answers from approved content, with an audit trail that survives review.
500,000 SKUs.
700 attributes per product.
Data from hundreds of suppliers in dozens of formats.
One search box.
Generic search was built for a few thousand clean consumer products. Not superseded part numbers and specs buried in a 2011 PDF.
Your buyer has a machine down.
If your best sales rep left tomorrow, you'd replace them within a month.
How long has your search been underperforming?
67% of B2B buyers now prefer a rep-free experience (Gartner). On those deals, your search bar is the rep.
"We'll see returns in 2028" is not a fundable sentence.
391% ROI over three years. Payback beginning in under six months.
When every budget is being scrutinized, the second number is the one that gets the project approved.
Forrester TEI, commissioned by Lucidworks.
"The results we've had with Lucidworks are nothing short of astounding."
— Marc Desormeau, Global Search Lead, Lenovo
55% better click-to-first-result relevancy.
95% increase in search-driven revenue.
Relevance came first. Revenue followed. That order is the thesis.
A "no results" page isn't a dead end.
It's a checkout page for your competitor.
On a large catalog, zero results is rarely a catalog problem. The product is there — the buyer just described it differently than your index expects.
That's a language problem. Solvable.
The vendor with the most features wins the demo.
The vendor with the best results wins the renewal.
Don't ask what a search platform can do. Ask what it produced — on your data, against a number you picked before the pilot.
What's the last feature you bought and never used?
Our latest benchmark found that 66% of organizations seeing measurable AI value cite productivity as the biggest benefit. Is your AI strategy producing measurable business outcomes or just more AI initiatives?
Download the 2026 AI Benchmark Report: okt.to/DxaBmP
The biggest AI risk may not be falling behind, but thinking you're ahead. Most organizations believe they are outperforming peers in AI maturity and
execution, but practical deployment data points to recurring challenges.
Explore the findings: okt.to/pq28by
Lucidworks is honored to be recognized on the 𝟮𝟬𝟮𝟲 𝗞𝗠𝗪𝗼𝗿𝗹𝗱 𝗔𝗜 𝟭𝟬𝟬, an annual list celebrating organizations advancing the future of artificial intelligence for knowledge management: okt.to/KBztj9
Who’s getting more value from AI: B2B or B2C? While B2C orgs are more mature in digital experience, B2B orgs are investing in AI faster, and reporting stronger benefits.
Download our 2026 State of GenAI in Business Report to see the data behind the trend: okt.to/A0i7f5
Your product catalog was built for humans. That’s the problem.
If an AI agent audited your catalog today, how would it score? okt.to/Ii2HO0
As enterprises adopt more AI models, one challenge is becoming impossible to ignore: keeping AI costs predictable. Our CEO, Mike Sinoway, shares his perspective in Forbes on why billing visibility is emerging as a key requirement for scaling responsibly: okt.to/bCEM5I
"I need a laptop for my daughter."
That's not a keyword. It's how customers search today.
Hear how Lenovo is combining AI with keyword precision to deliver more relevant search experiences, and preparing for the future of agentic commerce: okt.to/Cdavy5
If your search platform can't reliably find:
✔️ Exact part numbers
✔️ Technical documentation
✔️ Compatibility data
...your AI can't either.
That's why hybrid search is the foundation for manufacturing AI. Learn more: okt.to/bTmoAq
B2B buyers need both keyword (precise match) and semantic (intent based) search.
Because real-world queries look like:
• “3/8 stainless zinc-plated bolt”
…and also:
• “corrosion-resistant fastener for marine use”
We broke down how it works: okt.to/n8Dcap
Our latest research offers an AI a reality check, combining responses from thousands of executives and employees with real-world AI implementation data.
See where organizations are succeeding and where they're overestimating progress. Download the report: okt.to/sLCNEW
What does it take to have your products show up when an AI agent is doing the shopping? Structured product data. Semantic clarity. Technical specs that machines can parse. The companies that figure this out will own the next era of product discovery: okt.to/Nxkrqa
The biggest misconception in B2B commerce? That buyers always know exactly what they're looking for. They don't.
They compare, refine, backtrack, and ask questions.
Our latest guide explores what that means for modern B2B commerce teams. Read it here:
okt.to/vZKg3l
