Creative direction, campaign design, editorial and influencer work for a national brand. I set the idea, write the words, brief the people who execute it, buy the media that carries it — and I came up through the camera, so nothing gets promised that can't be shot.
Creative that proves itself organically before a rupee of paid media is spent.
300K+
Organic cross-platform reach built without paid boost
20+
Commercial reels & video assets directed, shot and cut
4.2x+
Average ROAS achieved across performance ad campaigns
Top 3
Organic search rankings built for high-intent keywords
Audited performance across Daewoo Nepal, e-commerce storefronts, and commercial client campaigns (2023–2026).
Six things, one person. The value is that the idea, the words, the casting and the media budget are all decided by someone who has to answer for all four.
The idea before the assets: what we're saying this quarter, why anyone should care, and the rules that keep it recognisable. I write the brief, choose the treatment, and hold the line when the work drifts.
One idea, built out across the places it has to live — feed, story, creator, landing page, search. Sequenced so each piece earns the next, with a number attached to it before anything is made.
Blogs, scripts, ad copy and the site's own pages. Written to be found and to be finished — search intent at the top of the draft, a human sentence at the end of it.
Casting, briefing, negotiating and reviewing. A creator who understands the idea outperforms one with a bigger following, so I brief for intent rather than handing over a shot list.
Meta campaigns built, tested and cut on evidence — return, click-through, profit after ad spend. Creative and buying decided by the same person, which is why the tests are worth reading.
Technical, On-Page, Off-Page and Local SEO, App Store Optimization, and Generative Engine Optimization (GEO). Structured so both traditional search algorithms and modern AI engines find, understand, and quote the work — turning search intent into measurable, compounding organic growth.
Most campaigns fail on paper, before anyone opens a camera bag — the idea was never sharpened, so every later choice is a guess. I write the decision down first: one audience, one thing we want understood, one measure of whether it landed. After that, direction is mostly protection. Creators get room to be themselves inside the idea, editors know what the cut is for, and I can kill a beautiful asset without a debate because we agreed in advance what the work was supposed to do.
Who it's for, what they should understand afterwards, and how we'll know. One page, agreed by the people who can change their minds later.
The right creators, editors and shooters for that idea, briefed on intent rather than shot lists — and told plainly what success looks like.
Video, copy, blog, landing page, ad set — written and sequenced together, so the idea survives the translation between them.
Reach, retention, click-through, return. The next brief is written by the evidence, not by whichever asset the room happened to like.
Head of Marketing at Daewoo Nepal: I own the idea, the calendar, the creators, the copy, the media budget, the Daraz storefront and the relationships that hold it all together. What follows is how that job actually runs, and what it has produced.
Daewoo Nepal, in full.
An appliance brand. On purpose.
Nobody wakes up wanting content about a refrigerator. Appliances are bought twice a decade, researched in a hurry and decided on trust — so a feed of product shots earns exactly the silence it deserves, and paid spend on top of it just buys a wider audience for the same shrug.
Talk about the household, not the machine. A publishing rhythm the brand could actually hold across three platforms; creators cast where a real kitchen was more persuasive than a studio; editorial written for the questions buyers ask before they walk into a showroom.
Organic reach past three hundred thousand. Paid spend widening only what organic had already proved, judged on return rather than impressions. And a brand site I built and rank, doing the convincing while nobody is watching.
{{ d.kicker }}
{{ pt }}
{{ role.org }}
{{ role.when }}
{{ pt }}
If the sentence isn't clear, the campaign isn't either. Blogs that answer a real search, scripts a creator can perform without sounding sold to, ad copy that survives being read in half a second, and site pages written to be quoted back by an answer engine.
{{ w.kicker }}
{{ w.body }}
Technical search strategy, buyer's guides, and editorial pieces written to rank, inform, and persuade.
Intent first: the piece exists because somebody typed something, and the opening paragraph should end their search rather than delay it. Structure second, so an answer engine can lift a clean statement out of the middle of the page and attribute it.
Then the part most brands skip — reading it aloud. Keyword density has never persuaded anyone to buy a washing machine. I've done this freelance for two institutions as well as in-house, and the discipline is the same: earn the click, then deserve it.
Twenty-plus videos I directed, shot and cut myself before I moved into direction — fifteen of them here. I keep them for one reason: when I brief a creator or an editor now, I know exactly what I'm asking for and what it costs to deliver. Every card plays its cover and opens the video on Instagram; the address sits under each one if your browser blocks the jump.
Twenty-plus made. Fifteen kept. The bad ones taught more.
Seven rooms around the valley, each selling a different thing — the char, the crowd, the view, the price. The work was deciding which, then shooting only that.
Someone else's guidelines, someone else's product, a deadline attached to the brief. Useful training in fitting a voice that isn't yours — the thing I now ask creators to do.
No lighting, no second take, no control over anything but where I stood. Documentary work teaches you to decide fast — the habit direction runs on.
Two griddles and a pair of hands that had done this longer than I'd been alive. Shot close, cut slow, nothing styled.
Personal work lives in the highlights at @_.anuraaag.
I tell stories for brands and direct the work that carries them. Engineer by training, shooter by apprenticeship, and now the person who decides what a brand is saying and why anyone should care.
The engineering degree wasn't a detour. It taught me to take a system apart before arguing about it — the same instinct that makes a campaign legible: find the part doing the work, cut the rest. Robotics gave me deadlines with a physical verdict at the end, and hackathons taught me to ship something honest in the time available instead of something perfect in theory.
Then a couple of years with a camera in kitchens across the valley, which is the best training in judgement I've had and the reason my briefs are realistic. What kept me was never the gear — it was noticing that the videos which travelled had a story in them and the polished ones often didn't. So I moved towards the story: what a brand is actually saying, who it sounds like, and why a stranger should care by the second sentence. Direction is where that landed — the idea, the words, the casting, the media, and the argument for why any of it was worth the money. Kathmandu-based, and most useful to a team that wants one person holding the whole thought.
Electronics & Communication Engineering at Advanced College of Engineering and Management — IOE, Tribhuvan University. Science before that, at Global School of Science.
I built a robot good enough to carry Nepal's name to Techfest at IIT Bombay, and won enough hackathons that the weekends blur together.
And I once wrote someone's college application essay. They're studying aerospace engineering at Berkeley now — the clearest proof I have that writing moves people.
Looking for a full-time seat in marketing or creative direction. Email travels fastest.
Direct answers for search engines, client teams and founders evaluating the work.
A Creative Director defines the central thesis and guardrails before any budget is committed or camera is packed. I decide what the brand will say and refuse to say, write the brief, cast creators for authentic fit rather than vanity follower counts, script the delivery, and review edits against intent — ensuring work proves itself organically before paid media scales it.
Traditional SEO focuses on indexing, backlinks, and keyword placement to rank clickable blue links on search engines like Google and Bing. Generative Engine Optimization (GEO) structures semantic entities, verified facts, and direct answers so AI answer engines (ChatGPT Search, Perplexity, Google AI Overviews, Claude) extract, synthesize, and cite the brand in conversational responses.
Creative is the single highest-leverage variable in performance advertising. When creative and media buying are directed by the same person, losing tests reveal which messaging premise failed rather than just which image didn't convert. Higher creative relevance earns cheaper CPMs, drives stronger click-through, and delivers compounding ROAS (averaging 4.2x+ across campaigns).
Intent first, structure second, human craft always. I research the exact commercial comparisons buyers search before purchasing, fix technical site architecture and schema for bots, and write buyer's guides and editorial that end the search on the page. Search intent earns the click; honesty and clear writing deserve it.
Ask for case studies with numbers you can actually check, not just claims. Ask whether the work includes technical fixes like schema markup and site structure, not just blog posts. And ask whether the same person is deciding the message and the technical execution, or whether that gets split across people who never talk to each other — that split is where most SEO work quietly falls apart.
An agency splits the work across departments — a strategist writes the brief, a different team executes creative, someone else runs media, and the person accountable for the result usually isn't the person who made any single decision. Hiring one person means the idea, the words, the casting, the media budget and the technical SEO underneath all sit with someone who has to answer for how they fit together — slower to scale past a certain size, faster and more coherent below it.
It depends more on scope than on title. A one-off technical audit is a fixed, bounded cost. An ongoing content and search strategy is usually a monthly retainer, priced against what building and ranking real content actually takes — not a flat rate that's the same whether the site has 5 pages or 500. Ask for a scoped quote against your actual site, not a generic package price.
Longer than anyone selling it wants to tell you, and faster than doing nothing. Daewoo Nepal's storefront went from zero organic history to indexed, growing traffic and a page-two average position within two months of launch — with the technical foundation done correctly from day one, not months of cleanup first. A brand-new domain competing for a genuinely competitive term should expect 6 to 12 months for real movement; a specific, narrow, well-targeted page can move faster.
No — and anyone who does is either overpromising or setting up a fight over a refund later. What's guaranteed is the actual work: a specific, checkable set of technical fixes and content, built correctly, the same way every time. Ranking follows from that being genuinely done well. Nobody controls Google's algorithm enough to promise a position.
Both, depending on what the site actually needs. A one-time technical and content audit makes sense for a business that wants a clear, prioritized list of what's broken and worth fixing. Ongoing work makes sense once the foundation is solid and the job becomes building and ranking new content over time. Most sites need the audit first, whether or not they continue past it.
Appliances and e-commerce, ride-hailing and marketplace platforms, FMCG, education, and hospitality — Daewoo Nepal, Pathao, Dabur Hajmola, and Advanced College among them, alongside restaurant and café brands. The common thread isn't the industry, it's whether the brand is willing to be specific instead of generic.
Not urgent in the sense that AI search already drives huge volume here — it doesn't, yet. But it's close to free to get right at the same time as normal SEO, mainly by not accidentally blocking the crawlers behind ChatGPT and Gemini through a default robots.txt setting almost nobody checks. Being one of the first properly-configured businesses in a category is a real, if narrow, advantage precisely because almost nobody else has looked.
Yes — local SEO is part of the same technical and content work, not a separate service. A complete, verified Google Business Profile, consistent business information across platforms, and location-specific content all feed the same search visibility as the rest of the site.
Technical SEO is whether search engines and AI crawlers can actually read and trust the site — schema markup, site speed, crawlability, structure. Content SEO is whether what's on the page is worth ranking — real answers to real questions, not filler around a keyword. A site can be technically perfect and still rank nowhere with thin content, or have great writing that search engines can't properly read. Most sites are weak on one side, rarely both equally.
AEO is structuring content so it can be lifted directly into a featured snippet, a voice assistant's spoken answer, or an AI Overview — not just ranked as a blue link. For Nepali businesses specifically, it matters because a search result and an AI-generated answer aren't judged by the same rules: a page needs a clear, direct answer near the top, not just good keyword targeting. Most sites here are still built for the old rules only.
Structure, not luck. Google's AI Overview pulls from pages that answer a specific question clearly and early, back the claim with something checkable, and carry schema that confirms what the content actually is. It favors pages that already rank reasonably well organically and answer the query directly in the first few sentences — chasing the citation without the underlying content and technical foundation being solid doesn't work.
Not a blog specifically — you need content that answers questions your actual customers are asking before they buy. That's often a blog, but it can just as easily be a buying guide, a comparison page, or an FAQ section like this one. The format matters less than whether it's actually answering something, not just filling a content calendar.
Yes, on specific, narrow terms — not on the broadest, most competitive ones immediately. Agencies with a decade of accumulated domain authority will win the biggest head terms for a while regardless of how good new work is. But most of what actually brings in customers is long-tail and local intent, which rewards being specific and correct over being big.
Treating the website as a brochure instead of a source of answers. Most sites here describe the business in general terms and stop — no real product detail, no pricing transparency, no structured data telling search engines or AI systems what's actually being sold. The fix isn't more content, it's more specific content.
Yes — building Daewoo Nepal's SEO foundation from launch was exactly this: structured Product schema, category and comparison content, and technical SEO built around how people actually shop, not just a product catalogue with prices attached.
Less often than most people fear, more often than most sites actually get touched. Technical foundations — schema, structure — are mostly set-once-done-right. Content needs revisiting when the underlying facts change: prices, availability, dates. A stale "last updated" date on time-sensitive content is a real trust signal, not just a formality.
Different jobs, not a straight substitute. Paid ads buy attention now and stop the moment you stop paying. SEO is slower to start but keeps working after the initial investment, and it's the channel most trusted by someone who's already comparing options rather than just noticing an ad. Most businesses need both, sequenced — ads for now, SEO for what compounds.
On a case-by-case basis, yes — usually as a working session against a team's actual site rather than a generic slide deck. Get in touch with what you're trying to solve and I'll tell you honestly whether a workshop format fits or whether the work itself is the better use of the time.
Google Search Console and GA4 for what's actually happening on a live site, not just projections. Beyond that, the tools matter less than reading the real SERP for a keyword before making a recommendation, which is manual work no tool fully replaces. Anyone whose answer is a long list of paid subscriptions is often selling the tools, not the outcome.