Two seats. Architecture seat — I define the model; your team publishes. Operating seat — I also run day-to-day CMS publishing. One operating seat at a time. Contentstack and Sanity. Measured with Profound and CrownPeak.
Limited partnerships. New York. Multi-month, not a package.
9.6 / 10
CrownPeak score (platform avg 6.9)
~1,000 → baseline
WCAG issues, 12-month meter
~4,000 → ~500
Total CrownPeak issues, 24-month meter
0
Priority issues across 378 pages
On a 378-page headless Shopify Plus catalog, CrownPeak accessibility score reached 9.6 / 10 against a platform average of 6.9. Priority issues: 0. Average issues per page: 1.14.
Scores moved by cleaning CMS output — not an overlay. Component HTML stays with engineering.
When I started as content lead, the catalog was two mattresses. It is now a full sleep brand: 15+ mattress models plus furniture, bedding, and pillows, published in Contentstack while that assortment grew.
Live citation pattern in the CMS: intent collection paired with a supporting guide. Example: Best Firm Mattresses collection plus the 2026 firm-mattress guide — query-shaped hub, short product set, comparison table, FAQs written to shopping questions.
The same heading, label, and entity fixes that cleared CrownPeak priority issues are what answer engines extract from collections, PDPs, and guides.
Stack fluency
No junior team. One senior partner.
No overlay vendor. No junior team. One senior partner inside the CMS.
Prompt sets, citation gaps, schema, Profound monitoring, collection and guide pairs.
Content models, FAQs, modules, collection and PDP patterns.
CrownPeak on CMS content; fixes in entries and modules; no overlay.
Shoppers ask ChatGPT, Claude, Gemini, and Perplexity for product recommendations. Those systems answer with brand names. If your content, entities, and structure are unclear, you are invisible — even if the store converts well in traditional search.
The same semantic structure that helps a screen reader helps an answer engine. Fixing both with one partner eliminates the dual-vendor waste that most brands accept by default.
20-minute review of current AI visibility and major accessibility issues. Used to decide fit.
Two weeks. Profound prompt set, CrownPeak content audit, gap analysis, and a prioritized action plan — CMS-ready changes plus a separate list of component-level issues for engineering.
Six-month minimum. Monthly loop: prompt set, citation gap, CMS changes, CrownPeak re-scan, Profound re-measure.
All engagements start with a free Snapshot. No commitment required.
It starts with a free 20-minute AI Visibility Snapshot. If there's a fit, we move to a two-week Baseline Diagnostic, then a six-month minimum partnership with a monthly measurement loop.
Occasionally, if the stack is headless and the CMS is enterprise-grade. The tools and method transfer — Profound and CrownPeak are platform-agnostic.
Traditional SEO optimizes for crawlers. This work optimizes for answer engines — the systems that synthesize recommendations rather than return a list of links. The overlap is structured content and clear entity signals; the goal is different.
