SEO in Azerbaijan — a 100-day plan

SEO in Azerbaijan is not the same job as SEO in a saturated English-language market. The keyword volumes are smaller, the SERP is less crowded, and a competent ninety days of work can take a Baku business from invisible to top-three on phrases that actually matter. The mistake is doing the global generic checklist instead of the local one. This is the plan we run.

Days 1–15 — foundation

Nothing visible to the public happens in the first two weeks. This is the part most local SEO contractors skip and the part that determines whether the next 85 days return anything.

Keyword research grounded in Azerbaijani buyer intent. Not "best restaurant Baku" — what real customers actually type. Tools: Google Keyword Planner with Azerbaijan geo, Ahrefs with az.tld scoping, manual SERP inspection in az.google.com using an incognito profile.
Map the keywords to URLs. Each target phrase gets one canonical URL. If you don't have it, plan it. If you have two pages competing for the same phrase, decide which one dies.
Technical audit. Crawl the site with Screaming Frog. Inventory: titles, descriptions, H1s, response codes, redirects, canonicals, hreflang, robots.txt, sitemap, image alts.
Set up the measurement floor. Google Search Console for all three language versions (.com/az/, .com/en/, .com/ru/), GA4 or Plausible, a basic rank tracker (Ahrefs, Semrush, or Serpstat) with az-Baku geo.
Competitor inventory. The five or six businesses showing up for your target phrases. Their content, their structure, their backlinks. You're not copying — you're calibrating.

Days 16–45 — technical and on-page

Now the changes start landing on the live site. Most of these never need to happen again.

Title tags rewritten one URL at a time, primary phrase up front, brand at the end. No keyword stuffing — Google's been done with that for a decade.
Meta descriptions rewritten to actually compel a click. SEO Baku traffic is small enough that click-through rate matters more than position 4 vs 6.
H1 per page, semantic H2/H3 below. One H1. Always.
Image alt text in the page's language, describing what's in the image, not stuffed.
Internal linking pass — each new content page gets at least three contextual internal links from existing pages.
Schema markup: Organization, LocalBusiness with Baku address and geo coordinates, BreadcrumbList, Article schema on journal entries, Product on e-commerce.
Hreflang implemented properly between Azerbaijani, English and Russian versions. This is where most multilingual Azerbaijani sites quietly fail — pages duplicate-compete with their own translations.
Sitemap regenerated, submitted to Search Console for all three language properties.
Page speed pass: Lighthouse target 85+ on mobile, LCP under 2.5s. This isn't cosmetic — Google explicitly ranks on it now.

Days 46–70 — content

This is where the work gets less mechanical and more judgement-bound.

Write or commission 6–10 cornerstone pieces in Azerbaijani. Each one targeting a real, mid-tail keyword cluster — not "design," but "logo design cost in Baku," "how to choose a logo designer," and so on.
Each piece is 800–1,400 words, structured with H2/H3, internally linked, with one or two relevant external citations.
Translate the strongest 3–5 pieces into English and Russian, properly — not via raw machine output. Hreflang ties them together.
Update existing thin pages. Anything under 250 words either gets expanded or merged into a parent page.

Days 71–85 — Google Business and local signals

For a Baku business serving Azerbaijani customers, Google Business Profile is the single highest-leverage SEO surface. Most local businesses fill it out badly and never touch it again.

Claim and fully complete the profile — categories, hours, services, full description in Azerbaijani.
Add at least 20 photographs of the real premises, products, team. No stock images.
Pin the Baku address with the correct coordinates. Get a postal verification done if not already.
Start a steady, organic review pipeline. Asking happy customers in person works. Buying reviews — never.
Local citations: business directories that matter in Azerbaijan (azerbaycan-business.az, sektor.az, Yandex Maps, 2GIS). Each one needs identical NAP — name, address, phone — across all listings. Inconsistency damages local ranking more than absence.

Days 86–100 — links, measurement, iteration

Links are the slowest-moving SEO lever, but they compound.

One genuine, earned backlink from an Azerbaijani publication, association, supplier or partner.
Guest article on a respected industry blog or media outlet, with a contextual link.
Audit existing inbound links — disavow obvious junk, thank good ones with reciprocal mentions where appropriate.
Compare day-100 rankings against day-1. The improvements you'll see at day 100 are the cheap ones; the rest will land over the following 6–9 months. Don't expect immediate position-1.

What this plan doesn't do

It doesn't guarantee top-three for a brutal commercial phrase like "site Baku" or "delivery Baku" — those battlegrounds need 18+ months of work and aren't a 100-day project. It also doesn't fix a fundamentally broken business — SEO sends traffic, but if the offer is weak, the traffic bounces.

What it does do: take a competently-built site from invisible to findable on the queries customers actually type. For most Baku businesses that's the difference between a website as a brochure and a website as a sales channel.

If you want the audit version of this — the day-1 to day-15 part run on your live site — Qaynaq provides it as a flat one-week engagement; write to salam@qaynaq.com with your domain.