Yuma AI SEO Teardown: Strong Blog, 12 Case Studies, and a Site-wide Image Alt Text Problem

Crawl data as of February 24, 2026. Analysis powered by redCacti.

Yuma AI is an e-commerce customer support automation platform built for Shopify, Gorgias, and Zendesk brands. They automate support tickets, cite automation rates between 70 and 89%, and have built a case study library around recognizable brands like Glossier, CABAIA, and EvryJewels.

Their web presence reflects genuine investment in content. Forty-two blog posts. Twelve live case studies with specific outcome metrics. Twenty-two use case pages. Seven competitor comparison pages. The foundation is strong.

The crawl tells a more complicated story. Image alt text coverage runs at 30.9% on the homepage and 9.4% on the about page - not as an isolated oversight, but as a pattern that holds across nearly every section of the site. Two internal links on the FAQ page point to case study pages that return 404. The competitors page carries 2,100 words and sits completely disconnected from the rest of the site with zero incoming internal links. And the case studies index - the hub page that 601 other pages link to - has 41 words of content.

The content engine is running. The technical layer underneath it has compounding problems.

What We Crawled

Yuma AI operates across two domains:

• yuma.ai - 136 pages (main marketing site)
• app.yuma.ai - 1 page (product sign-in, not a content target)
• Total: 137 pages crawled

The main site is meaningfully more complex than a typical SaaS marketing site. It covers six distinct content sections:

That is a meaningful content footprint for an e-commerce SaaS. The blog alone has posts in the 4,000 to 5,800-word range. The case studies publish specific outcome metrics: 87% response time reduction at Glossier, 74% cost reduction at CABAIA, 3x ROI in three months at Clove. The use case section covers 22 specific workflows with individual pages for each.

The issues in this crawl are not about missing content. They are about technical execution.

Section 1: The Image Alt Text Problem - Site-wide and Consistent

The single most consistent finding across the entire crawl: image alt text coverage is poor on almost every page category.

The blog posts are the best-performing section at 58.2% coverage - and that is still below an acceptable threshold.

The about page is the most severe case: 164 of 181 images have no alt text. This is a page that likely contains team photos, office imagery, and visual brand content. Screen readers encounter that page as almost entirely blank on image content. Google processes it similarly.

Why this matters beyond accessibility: Alt text is one of the signals Google uses to understand page content and the context of images. For a company that relies on product screenshots, UI walkthroughs, and customer-facing brand visuals, missing alt text on 70 to 90% of images means signal that never reaches the crawler.

Why it is happening: The pattern is consistent enough across sections to point to a template or workflow issue rather than page-by-page neglect. The most common cause: a CMS image upload workflow that does not enforce alt text entry, or a site design where background images and decorative images were never included in the accessibility layer.

How to fix it: Audit the CMS template for each page type. For programmatic image blocks - hero images, team photos, product screenshots - configure alt text as a required field rather than an optional one. For existing pages, a template-level batch update is possible for repeated image slots. The blog posts are already at 58.2%. Applying that same standard to landing pages, case studies, and the about page would close most of the gap.

Section 2: Broken Internal Links - The FAQ Points to Two 404 Pages

The crawl found 8 broken links total: 2 internal and 6 external.

Both internal broken links originate from the same page: yuma.ai/faq.

Both destination pages return 404. Live versions of both case studies exist at different URLs:

• /case-studies/how-javvy-coffee-achieved-70-automation-and-slashed-response-time-from-24-hours-to-12-minutes-with-yuma-ai (200 OK)
• /case-studies/how-clove-achieved-3x-roi-40-ai-automation-and-25-cost-savings-in-just-3-months-with-yuma-ai (200 OK)

This is a URL migration that was not fully followed through. The case studies moved from short /case-studies/[brand] slugs to descriptive /case-studies/how-[brand]-achieved... slugs, but the FAQ page still points to the old locations. The FAQ page has 3,021 words of content and is a meaningful organic target for support automation queries. Broken outbound links from that page pass no context to the destinations - and the 404 pages they point to are actively degrading the crawl.

The 6 external broken links:

The Recharge documentation link appears twice - once from the integration page and once from the root-level platform page - both pointing to the same 404 destination. The Notion link in the Siena alternatives post is a reference to an internal team document that was never meant to be public-facing.

How to fix it: Update the FAQ page to point to the live case study URLs. Remove or replace the external broken links. The Notion link should be removed entirely. Redirect /case-studies/clove and /case-studies/javvy-coffee to their live counterparts so any external links pointing to the old URLs pass equity forward.

Section 3: The Orphaned Competitors Page - 2,100 Words, No Incoming Links

yuma.ai/competitors-page has 2,100 words of content. It has an OG image. It has a full meta description. It compares Yuma against other AI support platforms with the kind of detail that a buyer evaluating tools would find useful.

And it has zero incoming internal links from anywhere else on the site.

A page like this sits at the exact intersection of competitive positioning and organic search. Buyers comparing Yuma against alternatives are high-intent visitors. Queries like “Yuma AI alternatives” or “AI customer support comparison” convert. The content exists to serve those visitors. The problem is that no other page on the site connects to it, so Google sees it as a standalone document with no editorial endorsement from the rest of the domain.

This is distinct from Yuma’s structured comparison section under /compare-yuma/, which has 7 dedicated pages targeting head-to-head competitor queries (vs Ada, vs Decagon, vs DigitalGenius, vs Siena, vs Sierra, vs Tidio, vs Zendesk AI). Those pages receive internal links. The competitors-page is a separate piece of content that was published but never wired into the site architecture.

How to fix it: Add a link to yuma.ai/competitors-page from the compare hub at yuma.ai/compare-yuma-ai, the pricing page, or the navigation. One relevant internal link from a well-connected page pulls it out of orphan status. Given that the comparison section is already a defined content cluster, this page belongs inside it.

Section 4: The Case Studies Index - 41 Words, 601 Incoming Links

yuma.ai/case-studies is one of the most internally linked pages on the entire site. 601 pages link to it. It is the hub for all customer outcome content.

It has 41 words.

The page likely renders as a grid of case study cards. The design is doing the work of a content page. But 41 words gives search engines almost nothing to index - no editorial framing of what these case studies demonstrate, no context about the types of businesses or outcomes represented, no signal about what a visitor should expect to find.

This matters because hub pages in well-linked content clusters can rank for category-level queries: “e-commerce AI customer support case studies,” “Yuma AI results,” “AI support automation ROI.” With 601 incoming links and almost no content to work with, the page is not positioned to capture those queries.

How to fix it: Add 150 to 250 words of introductory text to the case studies section. A short paragraph on what the case studies show, what types of businesses appear in them, and what outcomes recur across them. This does not require redesigning the page - it is a text block that sits above or below the card grid.

Section 5: What the Content Footprint Gets Right

Yuma’s content operation is one of the stronger ones in the e-commerce SaaS category.

Blog depth: The blog has 42 posts, and many of them are substantive. “AI Hallucinations in Customer Service” runs 5,218 words. “The Best 6 Siena AI Alternatives” runs 5,851 words. “Top 5 AI Tools for Gorgias” runs 5,073 words. These are not thin SEO posts. They are detailed guides with the kind of depth that can rank and hold position.

Case study specificity: The live case studies cite specific metrics: 87% response time reduction at Glossier, 74% cost reduction at CABAIA, 3x ROI in 3 months at Clove, 89% ticket automation at EvryJewels. Specificity in case studies matters for both organic credibility and conversion. Yuma’s case study library has it.

Use case coverage: 22 dedicated pages cover specific workflows - WISMO automation, returns processing, billing issue resolution, cart abandonment recovery, social comment moderation. Each page targets a specific buyer job-to-be-done rather than a generic feature description.

Comparison positioning: 7 structured comparison pages under /compare-yuma/ target high-intent competitor comparison queries. For a company competing against Siena, Ada, DigitalGenius, and others, this is the right content investment. Buyers comparing platforms directly are close to a decision.

Internal linking: Most pages have strong incoming link counts. The blogs index, FAQ, and case studies hub all accumulate substantial link equity from the rest of the site. The site architecture is generally well-connected.

Section 6: Schema Markup - Underdeployed Across Most Page Types

The crawl found schema markup on a small number of pages. The about page carries AboutPage schema. Most other pages have no schema at all.

For a site with Yuma’s page types, the gaps are significant:

• SoftwareApplication schema on the homepage and product pages (Support AI, Chat AI, Sales AI, Social AI) tells Google this is a software product and enables app-category SERP features
• FAQPage schema on yuma.ai/faq (3,021 words) could trigger accordion-style rich snippets for support automation queries - a high-value format on competitive SERPs
• Article schema on blog posts signals structured content to Google and supports correct snippet generation across 42 posts
• BreadcrumbList schema across the use case and case study sections signals content hierarchy and improves how deep pages appear in search results

Schema alone does not move rankings. But for a site with 42 blog posts, 22 use case pages, and a 3,000-word FAQ, deploying the right schema types at the template level is a one-time configuration that creates structured signal across dozens of pages in a single change.

How Yuma AI Can Think About Optimization

Quick Wins

• Update the FAQ page to point to the live Javvy Coffee and Clove case study URLs (2 broken internal links, same-day fix)
• Add one internal link to yuma.ai/competitors-page from the compare hub or pricing page
• Add 150 to 250 words of introductory text to the case studies index page
• Remove the Notion internal document link from the Siena alternatives blog post
• Update the Recharge documentation link on both /integration/recharge and /recharge-payments

Next Phase

• Audit the image upload workflow in the CMS to enforce alt text as a required field for new uploads
• Batch update alt text on key landing pages: homepage, about, pricing, and comparison pages
• Add FAQPage schema to yuma.ai/faq
• Add SoftwareApplication schema to the homepage and product pages
• Set up 301 redirects from /case-studies/clove and /case-studies/javvy-coffee to their live equivalents

Long-Term Investments

• Implement Article schema across all 42 blog post templates (one template change covers all posts)
• Systematically update alt text across case study pages, use case pages, and integration pages
• Review canonicalization on root-level platform pages that mirror /integration/ URLs (Shopify, Gorgias, ShipBob, Zendesk, Kustomer, Loop Returns, Recharge)
• Add BreadcrumbList schema to the use case and case study section templates

Key Takeaways

• Image alt text is the dominant technical gap - coverage ranges from 9.4% on the about page to 58.2% on blog posts, affecting accessibility and image search signal across nearly every page type
• Two broken internal links on the FAQ page point to case study URLs that were migrated but the FAQ was never updated to reflect the new paths
• The competitors page has 2,100 words and zero incoming internal links - it exists on the site but is invisible to internal link equity flow
• The case studies index has 41 words despite 601 incoming links - the strongest link equity hub on the site has almost no content to rank on
• Schema is almost entirely absent - FAQPage on the FAQ page, SoftwareApplication on product pages, and Article on blog posts would each cover multiple pages in a single template change
• The content foundation is strong - 42 blog posts with real depth, 12 live case studies with specific outcome metrics, 22 use case pages, and 7 comparison pages targeting competitive queries
• 8 broken links total (2 internal, 6 external) with the internal ones concentrated on a single high-value page

Yuma AI has built the content side of SEO more thoroughly than most companies at this stage. The issues in this crawl are almost entirely in technical execution rather than content strategy: alt text workflow, internal link gaps, broken links from a URL migration, and undeployed schema. These are fixes that do not require new content - they require a systematic pass through what is already there.

Want to run a crawl like this on your own site? Try redCacti