What I Would Do Differently If I Started My SEO Portfolio from Scratch
Building this portfolio has been one of the most valuable learning experiences of my BBA journey. I went from zero knowledge of GitHub and Jekyll to a fully functional, SEO‑optimised site with case studies, research articles, and a custom domain (still on the list). But looking back, there are several things I would do differently.
This post is not a “look how perfect I am” piece. It’s an honest reflection to help other students avoid the same pitfalls.
1. Publish Earlier, Even If Imperfect
What I did: Spent weeks tweaking colours, layouts, and fonts before publishing the first post.
What I’d do instead: Publish a minimum viable post (e.g., “Hello World – I’m learning SEO”) on day one. A live site with one paragraph is infinitely better than a perfect site that lives only on your hard drive.
Why: Google needs time to crawl and index. The clock starts when you publish, not when you’re happy with the design.
2. Focus on Content, Not Theme Tinkering
What I did: Endlessly customised the Minimal Mistakes theme – sticky headers, back‑to‑top buttons, category filters, search overlays (some of which I eventually removed).
What I’d do instead: Leave the theme 90% default. Write the case studies first. After you have 3–4 solid posts, then add polish. Recruiters care about your insights, not your border radius.
Why: Time spent on CSS is time not spent on keyword research or client outreach.
3. Use a Simpler Blogging Workflow from Day One
What I did: Wrote posts directly in GitHub’s web editor, often fighting with YAML errors and indentation. I also experimented with various search implementations that never worked perfectly.
What I’d do instead: Start with a local Jekyll installation and a text editor (VS Code). Draft posts in Markdown, preview locally, and only push when ready. For search, I’d skip custom code and rely on Google’s site search or a simple /search.json fallback.
Why: A local workflow catches errors before they break your live site, and it’s much faster for writing.
4. Optimise Titles and Meta Descriptions from the Start
What I did: Wrote functional but boring titles like “Ahmedabad Cafe SEO Audit”. Only later learned to add numbers, power words, and a clear benefit.
What I’d do instead: Apply the formula “[Number] + [Adjective] + [Topic] + [Result]” to every post from day one. Example: “Ahmedabad Cafe SEO Audit” → “How I Boosted a Cafe’s Traffic by 150% (A Complete Case Study)”.
Why: Titles and meta descriptions affect click‑through rates immediately. Changing them later means recrawling and losing early momentum.
5. Start Internal Linking Immediately
What I did: Published several posts in isolation before linking them together. The Market Research Hub came late.
What I’d do instead: Build a simple “Hub” page after the first two posts. Add a “Related posts” section at the bottom of each article (Minimal Mistakes supports it with related: true in _config.yml).
Why: Internal links pass authority and help Google understand topic clusters. They’re free SEO juice.
6. Not Over‑Engineer Contact Forms
What I did: Spent hours integrating Formspree, setting up redirects, and creating a thank‑you page. It works, but it was overkill.
What I’d do instead: Just put my email address and a LinkedIn link on the Contact page. For a portfolio site that gets fewer than 100 visitors a month, a simple mailto: is enough.
Why: Simpler is more reliable. You can always add a form later if traffic grows.
7. Set Up Google Analytics and Search Console Immediately
What I did: Added GSC after several weeks, and GA4 even later. Missed early traffic data.
What I’d do instead: On day one, add the GA4 tag and verify site ownership in GSC. Then, spend 15 minutes to understand the key reports (Performance, Coverage, Sitemaps).
Why: You can’t improve what you don’t measure. Early data helps you spot what works.
8. Plan a Content Calendar, Not Just Random Topics
What I did: Wrote posts whenever inspiration struck. Ended up with an unbalanced mix – many SEO case studies, fewer market research pieces.
What I’d do instead: Decide on 5–7 core topic clusters (e.g., local SEO, consumer psychology, BBA career advice). Aim for at least one post per cluster before expanding.
Why: A portfolio with depth in a few areas is more impressive than shallow breadth.
9. Not Fear the YAML
What I did: Avoided front matter customisation because I was afraid of breaking the build. Consequently, I missed opportunities like adding schema: Article, custom excerpts, and last_modified_at.
What I’d do instead: Read the Minimal Mistakes documentation and experiment on a local branch. YAML errors are fixable; missing structured data is a lost ranking opportunity.
Why: Schema markup directly improves how your site appears in search results.
10. Ask for Feedback Sooner
What I did: Kept the portfolio private until it was 90% complete.
What I’d do instead: Share the link after the first 2–3 posts – with peers, on LinkedIn, with my internship mentor. Their feedback would have saved me weeks of wasted effort.
Why: Early feedback validates direction or corrects course when it’s cheap to change.
What I Got Right (So I’m Not Only Criticising Myself)
- Choosing a static site generator – Jekyll + GitHub Pages is free, fast, and teaches real‑world tooling (Git, Markdown, CI/CD).
- Writing original research – The EV adoption and consumer psychology posts are unique to my experience and cannot be copied.
- Keeping it live – Despite the mistakes, the portfolio is live and improving.
Advice for Other BBA Students
You don’t need a perfect portfolio. You need a real portfolio. Start with a single post – 500 words on a topic you understand. Publish it. Then repeat. Iteration beats perfection.
And remember: the goal of a portfolio isn’t to impress recruiters with technical wizardry. It’s to prove that you can take an idea, execute it, and deliver results. Everything else is noise.
Have you started your portfolio? Questions or want feedback? Connect with me on LinkedIn.
This post is based on my personal experience building karandave211.github.io. Your mileage may vary – but the principles are universal.