Levelheaded · The full build
The short version is the case study. This is every step, 00 to 08: every report, what each one is, what the six delivered ones found for Levelheaded and what was built from them, the one still open, then the campaign the roadmap pointed at. Save it as a PDF from this page.
+200%
Campaign traffic
+82%
Form fills
156
Competitors analyzed
Levelheaded · Campaigns
Client · The ask
Levelheaded resolves landlord, roommate and neighbor disputes with Keel, its resolution engine, out of Denver. Keel was launching to off-campus renters, and nobody asking an AI about deposits, repairs or roommate disputes was being sent to Levelheaded. The ask: the roadmap first, then the traffic.
0.0%
AI citation share before
49.7%
The leader's share

Report 01 · Inputs · Discovery
156 competitors across eight channels: Google organic, the SERP features, paid and AI answers, measured over 3,000-plus buyer queries. Levelheaded held 0.0% of the AI citations; the leader held 49.7%. Finding 01 of five: 368 keywords a relevant competitor ranks for where Levelheaded was absent entirely, with the go-after list ordered shallowest competitor, highest volume first.
156
Competitors analyzed
8
Channels measured
0.0%
Citation share
49.7%
AI answer leader
What we built from it
The go-after list set the order for everything after it: which renter questions to answer first, and which audience the launch had to reach.
Report 01
Competitive analysis
Inputs · Discovery
What it answers
Who search engines and AI answers trust in your category, and where you stand against them.
What lands in the portal
A positioning matrix, the category boundary, and every move ranked by impact and effort.


Report 02 · Stage 1 · Audits
Delivered in the portal. Page by page, which renter questions belevelheaded.com answered, which pages were thin, and the fixes in priority order. It fed the content gap.
What we built from it
The rewrite list for the site, and the rule the campus page follows: the renter's question in the headline, the answer under it.
Report 02
Content audit
Stage 1 · Audits
What it answers
Which pages are thin, which are generic, and which are missing the keywords and AI readiness buyers look for.
What lands in the portal
A page-by-page review with fixes in priority order.
Report 03 · Stage 1 · Audits
Delivered before the campaign. Where visitors dropped off between landing and asking for help, and the top three fixes for the next 30 days. The campus page follows it: one page, one action, get it resolved.
What we built from it
The campus page as it shipped: one offer, one button, the price and the time to resolution next to it.
Report 03
CRO audit
Stage 1 · Audits
What it answers
Where visitors drop off and why.
What lands in the portal
The top three conversion fixes for the next 30 days.
Report 04 · Stage 1 · Audits
The Playbook showed where renters and students ask about deposits, repairs and roommate disputes, and which sources AI answers cite. That told us where to show up and what to say.
What we built from it
Where to show up and what to say: the questions the campaign had to answer, and the sources worth being cited by.
Report 04
AI visibility
Stage 1 · Audits
What it answers
How often ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews name your brand, who gets named instead of you, and which sources they pull from.
What lands in the portal
Your score, your mention rate, a per-engine map of who gets named instead of you, and the content and website fixes that change it.

Report 05 · Stage 1 · Audits
The one report in Levelheaded's Playbook not yet delivered. It runs against belevelheaded.com and the campus page: crawl, index, speed and schema, page by page, so Google, Bing, Claude and ChatGPT read every one of them.
Report 05
Technical SEO audit
Stage 1 · Audits
What it answers
What stops Google, Bing, Claude and ChatGPT from reading and indexing your site properly.
What lands in the portal
Crawl, index, page speed and schema issues, page by page, with the fix for each.
Report 06 · Stage 1 · Audits
Built on the content audit and the AI visibility report: which renter questions competitors answer that Levelheaded didn't, scored by difficulty, each with a brief and a publishing order. Delivered, and it fed the channel plan.
What we built from it
The publishing order for the renter questions, and the three the campus page answers on the page itself: landlord, roommate, neighbour.
Report 06
Content gap
Stage 1 · Audits
What it answers
Which buyer questions competitors answer and you don't.
What lands in the portal
Difficulty-scored content opportunities, each with a brief and a publishing order.
Report 07 · Stage 2 · Channels
Where off-campus renters actually spend time and what it costs to reach them there. The plan came back in three parts: Reddit and Meta, the campus audience, one landing page.
What we built from it
The campaign itself. Reddit and Meta into /campus for the Keel launch, run every week with the results in the portal.
Report 07
Channel optimization
Stage 2 · Channels
What it answers
Where your buyers actually spend time, and what it costs to reach them there.
What lands in the portal
A placement and spend plan for Meta, Reddit, Google and LinkedIn.
Build · The campaign
For the Keel launch, we pointed Reddit and Meta ads at the campus landing page for off-campus renters. One page, one action: get it resolved. Campaign traffic up 200%, form fills up 82%. Next in the Playbook: the technical SEO audit, then the next campaign off the same plan.
+200%
Campaign traffic
+82%
Form fills



Your turn
With our Business Process Engine's deterministic compute that delivers results without the AI drift.
Fill out the form.