MLJ Solutions, Powered by Optume.ai Finalist Demonstration. August 21, 2026.
We took the 24 prompts from your own guidance document, word for word, and put them to five AI platforms: ChatGPT, Gemini, Claude, Perplexity, and Grok.
We asked each question twice. Once letting the AI answer from memory. Once with live web search turned on.
That gave us 204 real answers, scored against the four firms you named.
This is what came back.
We built our test panel three weeks before your guidance document ever arrived.
When your document landed, we compared. Your five categories matched ours. Your four competitors were already on our list.
We had guessed your exam paper before you wrote it.
That is not luck. Anticipating how AI systems get asked about you is literally the job. And it is the strongest proof we can offer that we know how to do it.
Littler appears in 19% of the 204 answers.
The other 165 queries? They return your competitors. Or no firm at all.
Ask "which law firms are known for labor and employment law," and you are everywhere. Cited. Linked. Top of the list.
Ask what an employer should actually do about an OSHA inspection, a termination, an ADA accommodation, and you vanish.
The AI knows who you are. It just never uses what you know.
Legal knowledge questions: 44 answers. Littler mentioned in zero.
Jurisdiction questions, California, Texas, the UK, Portugal: 35 answers. Zero again.
Here is the part that should excite you: your competitors are mostly at zero there too.
Nobody owns these questions yet. The first firm that structures its content for them takes the entire category. That firm should be you.
AI answers from memory mention Littler 22% of the time.
Turn on live web search, and it drops to 17%.
Read that again. The machines know your brand better than your website shows up.
That is not a brand problem. That is a website structure problem. And structure is fixable.
When Littler does get mentioned, 85% of the time it comes with a proper link to littler.com.
So the pipes work. The moment your content surfaces, the AIs cite it correctly and describe you positively.
Surfacing is what is broken. And surfacing is exactly what we fix.
Grok likes you most: 26%. Gemini next at 23%. Claude 17%. ChatGPT 15%. Perplexity trails at 12%.
Overall AEO score: 60 out of 100. Improving. But improving from a low floor.
Your schema score is 6 out of 100.
Your attorney bios do carry Person schema. We checked by hand this morning. Erin Webber's page tells the machines her name, her title, her phone number, and her office address.
And that is all it tells them. A business card. No practice areas. No education. No bar admissions. No link to anything she has written. The AI can see that she exists. It cannot see why to cite her.
The front door is thinner still: the homepage carries two generic schema types, has no question-style headings, and fails 32 of 61 automated checks.
And the directory? Fetched without JavaScript, the way many AI crawlers fetch, littler.com/people returns zero attorney links. Your July lateral hire from Jackson Lewis: search his name today, and his old Jackson Lewis bio surfaces. His Littler bio does not.
The machines know the firm exists. They cannot navigate the expertise. That is the gap, and it is fixable.
No question-style headings. 32 of 61 automated checks failing.
AI engines cannot cite what they cannot parse. Right now, littler.com is very hard to parse.
This is our platform, running the audit from this morning.
The dashboard. The memory versus search cards. The website audit. And the head-to-head grid: you against Jackson Lewis, Ogletree, Seyfarth, and Morgan Lewis, prompt by prompt.
Every number you see traces back to a saved AI answer. Nothing is estimated.
The nine keyword categories in your guidance, workforce management, labor relations, wage and hour, litigation, workplace AI, pay transparency, DEI, immigration, remote and global workforce, are the draft map.
In weeks 2 through 6, we validate and prioritize that exact list against real citation-gap data.
Your cover note asked the selected vendor to recommend where to focus. This is how we answer it.
The first 12 weeks build everything: discovery, baseline, competitive benchmark, strategy, roadmap, technical and content plans, your dashboard, and the governance model.
Then the division of labor is simple. We diagnose, recommend, and measure. Your web team implements on Drupal, exactly as your own Q&A described.
Nothing ever publishes without Littler approval. We never touch your CMS. We work from targeted Google Analytics and Algolia reports, nothing more.
Your side of the bargain: reviews within one to two weeks. Your words, not ours.
85,000, one time.
22,000 per month.
32,000 per month — what we recommend.
45,000 per month.
Growth buys you 250 prompts checked every two weeks across six platforms, five competitors tracked, 1,500 pages under management, and the attorney bio program rolling out in waves.
A 4% yearly cap. No hidden fees. No exit fee. And insurance and security were already cleared through your own Q&A process.
The AI engines already believe Littler belongs on every shortlist. They just cannot find your content when employers ask real questions.
Closing that specific, measurable gap is the entire program.
This morning's full audit report can be in your inboxes today.
What the AI Engines Said About Littler This Morning