Search Engine Optimization Specialist

GEO + SEO Strategist
Mumbai (Andheri West) · Full-time, on-site · 3–6 years
The role in one line
You decide what a search engine ranks and what an AI engine says about a client — and then you make both true, on the client's own properties, with evidence, inside whatever regulator governs them.
Why this job exists
Discovery has split in two, and most brands are only equipped for one half of it.
Half your client's customers still search, compare, and click. The other half ask ChatGPT, Perplexity, Gemini, Claude, Copilot, Grok or Google's AI Overview a question and act on the single answer that comes back — no comparison, no second opinion, no ten blue links.
Those two halves are not separate disciplines. The same clean entity data, the same structured schema, the same authoritative page that wins a featured snippet is what an AI engine retrieves and cites. You will own both, because separating them is how brands end up invisible in one while paying for the other.
We do this for listed groups, regulated funds, hospitals, law firms, spirits brands, defence suppliers and family offices. Which is why the discipline below isn't optional.
The outcomes you own
This is not a reporting role. You are measured on movement, and on being able to attribute that movement to something you shipped.
• Answer Ownership Rate — the share of target questions where the client is named, cited, and described accurately. All three, or it doesn't count.
• Category presence — appearing at all on the non-branded questions a buyer actually asks. Most clients start at zero here. Moving that is the headline outcome of most engagements.
• Citation footprint — who is answering for your client today. Their own site, a retailer, a marketplace, a review aggregator, or a competitor. Shifting that mix back toward owned properties.
• Organic performance alongside it — rankings, indexation health, qualified non-brand traffic, and the technical foundations that carry both channels.
• Entity accuracy — every engine returning the correct structure, leadership, credentials and claims. Correcting a wrong answer is often worth more to a client than winning a new one.
Every number we report carries its numerator, its denominator and its sample. Below a minimum sample we publish a range, not a point estimate.
What you'll actually do
• Audit the full estate yourself — domains, subdomains, microsites, sitemaps, robots.txt, schema, status codes, internal linking, crawl budget, Core Web Vitals. Every finding carries its URL, its date, and a tag: Observed, Reported, Hypothesis, or Not Verified. If you couldn't open the page, you write "page not opened." You don't infer it from a search snippet.
• Build the prompt and keyword universe — 150–250 questions a real buyer, allocator, patient or procurement head would ask, banded by whether the client can own it, contest it, or should only watch it. Budget follows the band. Run it across seven engines and conventional search, and grade every answer.
• Set the strategy, not just the task list. Which lane runs first and why, what gets fixed before anything gets published, what the client should deliberately not spend on. You will argue this in front of CXOs and promoters, and you will be asked to defend it.
• Design the content and entity architecture — the pages that answer what a buyer actually asks, structured so both a crawler and a retrieval model can use them. Schema, llms.txt, agents.md, canonical structure, internal linking, and the fact sheet that governs all of it.
• Write the client's fact sheet and defend it. A group with three different store counts across its own sites, an executive carrying a title from a previous employer, a certification withdrawn in 2001 still claimed on a live page — finding these is most of the job, and it's the part clients remember.
• Build the compliance frame before the content. SEBI LODR, ASCI, state excise, RERA, NMC, Bar Council Rule 36, no-solicitation frames for funds raising capital. The frame comes first; content lives inside it.
• Verify everything you ship. Re-crawl it, confirm the schema, confirm the engine saw it. Nothing is marked done on trust.
What you're not doing
Running outreach — we build prioritised placement packs and the client sends them. Executing catalogue-scale data work — we deliver the specification and a worked reference implementation, the client's team scales it. Link schemes, suppression, de-indexing, or anything that manufactures a signal rather than earning one.
What we're looking for
Intellectual honesty under commercial pressure, above everything. The hardest moment in this job is telling a client that a finding you were proud of didn't survive re-testing. Our entire differentiation is that our claims are checkable — and these clients check them.
Plus: genuine research instinct, the judgement to know which finding matters and which is noise, writing a CFO would read without wincing, and real command of technical SEO fundamentals — crawling, indexation, canonicals, structured data, sitemaps, status codes.
3–6 years in SEO, content strategy, digital PR, equity research, consulting or business journalism. We care far more about how you reason than the label on your last role. Regulated-industry experience and enough Python to not be blocked by a 500-row dataset are both strong pluses.
What won't help you here
Traffic graphs with no attribution story. Agency vocabulary without the ability to show the underlying page. Any habit of filling a gap in the evidence with a confident sentence.
How we'll assess you
A two-hour live exercise, not a take-home. We give you a real company's public web estate. We want to see what you find, what you refuse to claim, what you'd do first, and how you write it up.
Finding less but stating it precisely beats finding more and overstating it. We mean that literally in the scoring.
What you get
Direct exposure to founders, CXOs and promoters of some of India's largest listed groups — from the first audit, not after two years. A discipline roughly three years old globally, where the work you do becomes the playbook. And a team that will back a correction over a comfortable claim, every time.
To apply
Send your CV and — more usefully — a short note on one thing you've checked and found wrong. Any subject. Tell us how you knew.
abhay@optimizer360.ai

Work arrangement
No