The questions worth asking

You do, always. Campaigns run inside your Meta Business Manager and your Google Ads account, with me added as a user. If we stop working together you keep the account, the pixel history, the audiences and the creative — nothing needs migrating. Be careful with anyone who runs your spend through their own account: you lose the conversion history that makes the algorithm work, and rebuilding it costs months.

I do. That's the entire reason to hire a freelancer over an agency — at an agency, the person who wins the pitch is rarely the person in the account, and your campaigns often end up with whoever is most junior and least busy. With me there is no account manager layer, which cuts both ways: you get direct access, but I take on fewer clients because attention doesn't scale.

Below roughly ₹10L a month in ad spend, a good freelancer usually delivers more attention per rupee, because agency retainers price in account management you may not need. Above that — or when you need separate creative, media and analytics teams working in parallel — an agency structure starts earning its overhead. Anyone who tells you freelancers are always better is selling you something; so is anyone who says the opposite.

I ask for three months minimum, and the reason is mechanical rather than commercial: Meta and Google need roughly 50 conversions per ad set per week to exit the learning phase, and a restructured account rarely produces stable, trustworthy numbers before day 30. Judging the work at week three means judging the learning phase. After the initial period, month to month is fine.

Freelance paid media management typically runs ₹15,000–₹40,000 a month in Belgaum and ₹25,000–₹75,000 in Pune, scaling with ad spend, number of platforms, and whether creative and landing pages are included. Agencies generally start around ₹40,000–₹80,000 because you also fund the account management layer. Be suspicious of flat packages that ignore ad spend — managing ₹50,000 a month and ₹5L a month are genuinely different jobs.

Around ₹30,000 a month for lead generation and ₹1,00,000 for e-commerce, and the constraint is data rather than reach. Below that you generate too few conversions per week for the platform to learn, so you pay to sit in the learning phase indefinitely. If your budget is under that, spend it on one platform and one offer rather than spreading it — three underfunded campaigns learn nothing, one funded campaign learns fast.

Start with Google when people are already searching for what you sell, and Meta when they don't know to look for it yet. Google captures existing demand; Meta creates it. A dental clinic should start on Google, because 'dentist near me' is already being typed. A new D2C product nobody knows exists should start on Meta, because nobody is searching for it. The mistake I see most is spending on Google for a product with no search volume, then concluding that ads don't work.

The specific numbers: ₹80L+ in revenue in a single quarter for Vega Auto Accessories at 5.58x blended ROAS; 725 admission leads at ₹238 average cost per lead for SKE Society, with the best campaign at ₹189; and 4.68M views on a 69-second helmet-safety film at 2.9% CTR. Those are blended, account-level numbers rather than a cherry-picked best ad — a single ad set at 12x proves nothing if the account average is 1.8x.

Two to three weeks for a usable cost per lead, six weeks for lead quality to stabilise, and 30–45 days for a trustworthy e-commerce ROAS. The gap exists because tracking usually has to be repaired and campaigns restructured before the platform has clean signal to learn from — and the first fortnight of a rebuilt account is the algorithm learning, not the account performing. Anyone quoting you a specific ROAS in week one is guessing.

Whatever clears your contribution margin — there is no universal number, and treating 4x as a benchmark is how brands lose money profitably-looking. A jewellery brand at 60% margin survives at 2.5x; a phone-case brand at 20% margin is losing money at 4x. Work out the ROAS at which you break even, then set the target above it. If someone quotes you a target ROAS before asking your margin, they're guessing.

You get a Looker Studio dashboard pulling from the ad platforms and GA4 directly, so you're reading the same source I am rather than a slide I typed. I also reconcile platform-reported conversions against actual orders, because Meta and Google both over-report — each claims credit for the same sale, so adding the two dashboards together always overstates reality. The number that matters is what landed in your bank account.

One variable at a time, in a structure where the answer is attributable — usually hook, then format, then offer, in that order. Most accounts test five things at once, get a winner, and cannot say which change caused it, so the learning does not transfer to the next campaign. I test hooks first because creative fatigue almost always starts at the first three seconds, not at the offer, and a new hook on a proven offer is far cheaper to produce than a new offer.

Both, and it is deliberate rather than an upsell. Most underperforming accounts I take over are not losing money on targeting — they are losing it because the landing page contradicts the ad, loads slowly on a mid-range Android, or asks for information the visitor is not ready to give. Running ads against a page you cannot change means diagnosing a problem you are not allowed to fix.

Digital marketing covers everything done online — social, SEO, content, email, ads. Performance marketing is the subset where every rupee ties to a measurable outcome: a purchase, a qualified lead, a booking. A digital marketer might report reach and engagement; a performance marketer reports cost per acquisition and return on ad spend. If the spend cannot be traced to revenue, it is not performance marketing regardless of what the invoice says.

Yes. Paid media runs remotely — the ad account, analytics and creative pipeline are all online, so location only matters if you want in-person strategy sessions. I am based in Pune and much of my campaign work has been for Belgaum businesses, so those two cities show up most in my case studies, not because of any geographic limit.