YMYL SEO Case Study: fundabroad.com from 41 to 63 in 8 Weeks

fundabroad.com operates in education finance, one of the more competitive corners of Indian search. Queries such as "study abroad loan", "education loan for Canada" and "MPOWER vs Prodigy" are contested by well-funded lenders, aggregators and banks. The startup had a clear product story but a weak visibility foundation, and, crucially, it operates in a category Google treats with extra caution.

What makes this a YMYL SEO case study

"Your Money or Your Life" (YMYL) is the term Google's Search Quality Rater Guidelines use for topics that could significantly affect a person's health, financial stability or safety. Education loans are squarely in that category: a student choosing the wrong lender can carry the consequences for a decade.

For YMYL topics, Google's guidelines ask raters to hold pages to a higher standard of Experience, Expertise, Authoritativeness and Trust (E-E-A-T), with trust described as the most important of the four. In practice, that means a fast, well-written loan page from an anonymous-looking site will struggle against a slower page from a source that clearly shows who is behind it, who reviewed the content and how to complain if something goes wrong.

AI answer engines behave in a similar spirit. When ChatGPT, Gemini or Google's AI Overviews answer a money question, they lean towards sources with visible accountability. That is why our sequencing here was trust first, performance second, content depth third.

Where we started

The first RankStreet audit returned:

For a finance-adjacent site, the E-E-A-T line was the real blocker. The site was effectively unrankable for anything beyond navigational queries, because nothing on it told a search engine or a cautious student who stood behind the advice.

Weeks 1 and 2: building the trust foundation

We started with the signals a sceptical reader, a quality rater and an AI system would all look for.

• Added Organization schema with sameAs links to the company's LinkedIn, X, Instagram, Trustpilot and Google Business Profile.
• Added Person schema with credentials for the founding team on the About page.
• Added visible "Reviewed by" bylines with expert credentials to every loan-product page.
• Published a comprehensive privacy policy, terms of service and a grievance officer contact page, linked from the footer.
• Integrated Trustpilot and displayed the rating badge above the fold.

Here is the shape of the reviewed-by pattern, combining visible text with matching markup:

<p>Written by [Author name], [role]. Reviewed by [Reviewer name], [credential]. Updated [date].</p>

{
  "@context": "https://schema.org",
  "@type": "WebPage",
  "name": "Education Loan for Canada",
  "reviewedBy": {
    "@type": "Person",
    "name": "[Reviewer name]",
    "jobTitle": "[Credential]"
  },
  "lastReviewed": "YYYY-MM-DD"
}

The placeholder names are deliberate: the point is the structure, and the rule is that every name in the markup must be a real, visible person on the page.

Impact: E-E-A-T score moved from 42 to 71, with the trustworthiness section going from 12/25 to 22/25.

Weeks 3 and 4: the performance overhaul

With trust in place, we turned to the failing mobile experience. Students research loans on their phones, often on patchy connections, so mobile was the priority.

• Migrated hero images to WebP with a 180 KB budget and explicit dimensions.
• Enabled server-side caching and Cloudflare's CDN.
• Removed four unused third-party scripts.
• Deferred non-critical JavaScript, including the chat widget and analytics.
• Inlined critical CSS for above-the-fold content.

Impact: mobile PageSpeed moved from 34 to 74, mobile LCP from 5.1s to 2.2s and CLS from 0.31 to 0.06. The same playbook, in more detail, is in our Core Web Vitals case study for viraaprofessional.com.

Weeks 5 and 6: content depth and FAQ schema

The site had product pages but nothing that answered the questions borrowers actually ask before they apply. We added three things:

• An FAQ block on every loan-product page with six to ten real questions sourced from the sales team, marked up with FAQPage JSON-LD.
• A "Compare loans" page with structured comparison tables.
• Country hubs for the top five destinations (US, UK, Canada, Australia and Germany), each answering visa, cost-of-living and loan-specific questions.

The country hubs are where content depth paid off. A thin page that says "we offer loans for Canada" gives an AI system nothing to quote. A hub that explains, in plain sentences, what documents a lender typically asks for, how proof-of-funds requirements work and how collateral-free loans differ from secured ones gives it several liftable passages.

An example of the before-and-after on one question:

Before: "Get the best education loan for Canada with fast approval. Apply now!"

After: "An education loan for Canada can be secured (backed by property or a fixed deposit) or unsecured (based on the student's admission and the co-applicant's income). Unsecured loans usually need no collateral but may carry higher interest rates. The right choice depends on the loan amount, the course and whether a co-applicant has collateral to offer."

The second version is longer, but it answers the question. That is what AEO looks like at sentence level. For the JSON-LD template, see our FAQ schema markup guide.

Impact: the AEO friendliness score went from 38 to 76, and the site began appearing in Google AI Overviews for "study abroad loan for [country]" style queries.

Weeks 7 and 8: launching a blog with freshness built in

• Launched a blog with eight substantive posts covering loan comparisons, forex, visa financial requirements and repayment strategies.
• Gave every post an author byline with credentials, a "Reviewed by" line and a visible "Updated on" date.
• Cross-linked each post to the relevant loan-product pages and country hubs.

Impact: the audit's blog freshness signal flipped from "No blog" to "Active", and internal linking depth improved significantly.

Results after eight weeks

The overall score crossed 60, the point at which we stop describing a site as structurally weak, and the site passes Google PageSpeed on both mobile and desktop.

What is still on the list

A 63 is progress, not a finish line. The next priorities are:

1. Scaling AEO tracking. Add 20 more prompts to the RankStreet AEO/GEO tracker across ChatGPT, Claude, Perplexity and Gemini, and iterate the content that lifts mention rate. Our guide to getting cited by ChatGPT covers the framework.
2. Earning coverage. Backlink data from Common Crawl shows sparse coverage, so the plan is earned mentions in edtech and fintech publications.
3. More country hubs. Ten more destinations, at the same depth as the first five.
4. Real-user monitoring. Cloudflare Web Analytics is live; Chrome UX Report data is next, to catch field regressions that lab tests miss.

Lessons for other YMYL sites

• Trust first. On money and health topics, speed and content cannot compensate for anonymity. Show who wrote it, who checked it and how to reach you.
• Markup must mirror reality. Reviewed-by schema with no visible reviewer, or credentials nobody can verify, is worse than nothing.
• Answer the actual question. Borrowers ask about collateral, co-applicants and visas. Pages that answer those questions get quoted; pages that say "apply now" do not.
• Ship weekly. Eight weeks of small, measurable changes moved this site further than a redesign would have, and every step was attributable.

Looking towards 2027, we expect scrutiny of YMYL content to tighten rather than loosen, as AI assistants take on more of the first-pass research for financial decisions. Sites that can demonstrate accountability will be the ones those assistants are comfortable recommending.

YMYL mistakes we see on almost every first audit

fundabroad.com was not unusual. On finance, health and legal sites, the same problems appear again and again:

• Anonymous advice. Articles about loans, insurance or treatment with no author, no reviewer and no date.
• Credentials that cannot be checked. "Our experts" with no names, or bios with no link to a professional profile.
• Buried accountability. Grievance contacts, licences and policies hidden several clicks deep, or missing entirely.
• Calculators without explanation. An EMI calculator with no text explaining assumptions gives readers and machines nothing to trust or quote.

Each of these is cheap to fix compared with the visibility it costs.

Run the same audit on your site

Run a free RankStreet audit on your most important page to see the same section breakdown, including the E-E-A-T and AEO friendliness scores used in this case study. If your site is where fundabroad's was at the start and you want the same team to run the playbook, talk to Digistreet Media.

Frequently asked questions

What is a YMYL website?

YMYL stands for "Your Money or Your Life". Google uses it for topics that can significantly affect someone's health, finances or safety, such as loans, insurance, medical advice and legal matters. These pages are held to a higher E-E-A-T standard.

Why did this YMYL SEO case study fix trust before speed?

On YMYL topics, a fast page from a source with no visible accountability still struggles to rank or be cited. Trust signals such as named reviewers, organisation details and grievance contacts unlock the benefit of every later fix.

Does "Reviewed by" schema improve rankings on its own?

No single markup property is a ranking switch. Reviewed-by information helps when it reflects a real, qualified reviewer shown on the page. It makes the page's accountability clear to readers, quality raters and machines.

How quickly can a YMYL site improve its SEO score?

In this case the RankStreet score moved from 41 to 63 in eight weeks with weekly changes. Ranking and AI-citation gains typically lag behind on-site improvements, and in competitive finance niches they keep compounding for months.

Which schema types matter most for finance sites?

Organization with sameAs links, Person for authors and reviewers, FAQPage for genuine questions, and Article or WebPage with dates and reviewer details. Product-type markup such as FinancialProduct can help where it accurately describes the offering.