How to Choose a Digital Agency in India

Most advice on choosing an agency is written by agencies, and it quietly assumes the reader is comparing firms rather than trying not to lose six months and a large invoice. This one is written from the other side of the table — what we get asked, what we answer badly, and which questions separate a studio that will do the work from one that will sell you a deck.

Start with the uncomfortable framing: you are not buying marketing. You are buying a specific change in your business — enquiries, orders, a category you can enter, a price you can finally charge. Every question below is really testing whether the person in front of you has that outcome in mind or a service menu.

The questions that reveal something

"What will actually be delivered, week by week, in month one?"

A real answer names documents and actions: the audit, the page list, the redirect map, the first set of creatives. A vague answer means nobody has planned it yet and the plan will be billed to you as discovery.

"Who does the work?"

Ask for the name, not the team. In a small studio the person you meet is the person who does it, which is the honest advantage of a studio over a large agency. In a larger firm, ask whether the account person touching your budget has ever shipped anything. Neither model is wrong. Not knowing which one you are buying is.

"What does reporting look like, and can I see a real example?"

Ask for a redacted monthly report from a client similar to you. If the only metrics they show are impressions, followers or "traffic growth", they are reporting activity because outcomes are not there. Good reporting says what changed, what it cost, and what happens next as a result.

"What would make you recommend against spending this money?"

This is the most useful question on the list. Any firm that has never talked a client out of a project is selling, not advising. A credible answer is specific: your margins cannot support the cost per enquiry, your offer is not settled yet, a landing page will teach you more than a website rebuild this quarter.

"What happens if it does not work?"

Not a guarantee — nobody can honestly promise a ranking. Ask what the review point is, what gets changed when a channel underperforms, and what your exit looks like. The answer is about process and notice period, not about courage.

"Can we speak to one client we resemble?"

References you have to ask for are usually fine. References volunteered with a prepared script are not. A studio that cannot connect you to anyone is often a studio with churn, which is a different problem from a small portfolio.

"What do you need from us, and how much of my time is this?"

Every project has a hidden cost in your hours: approvals, content, photographs, access, decisions. Ask for it in writing. If they say "almost none" for a website or a content programme, they have either not done one or they plan to guess at your business and let you correct it later, expensively.

"Who owns the accounts, files and data at the end?"

Domains, hosting, ad accounts, analytics properties, design source files, the content calendar. These should be in your name from day one, with the agency as a user. An agency that holds the keys has engineered your dependency.

"How do you price, and what makes the number move?"

You are not asking for a rate card. You are asking whether the price is tied to scope or to effort. A fixed number against a written scope is manageable. An open retainer with no defined deliverables is where budgets quietly disappear.

Six answers that should worry you

Guaranteed rankings, guaranteed first page, guaranteed leads. Search outcomes depend on competition, relevance and time. A guarantee is either meaningless or achieved by tactics that put your domain at risk.

A price that looks impossible. If a quote is a fraction of a comparable one, something is missing: the work is being done by an intern, the deliverables are thinner than they sound, the tools are unlicensed, or the price rises after month two.

No written scope, but a contract. The contract protects the engagement. The scope protects you. Ask for both, and read the scope first.

Access withheld. You cannot see the ad account, Search Console, or analytics until you ask twice. Ask on day one instead.

Everything is included. "Full digital presence management" with no page counts, no post counts and no reporting cadence is a description, not a scope.

Pressure to sign this week. Discounts that expire before you have read the scope exist to stop you comparing. Legitimate agencies lose deals to better-fitting firms all the time.

What a good proposal looks like

Five things, in this order: what you said the problem is, in their words. What they propose to do about it and what they are explicitly not doing. What you receive, itemised. The timeline with your commitments marked as dependencies. The price, what it covers, and what would change it.

If a proposal cannot be read without the person who wrote it explaining it, it is not a proposal — it is a pitch with a number at the end.

The part nobody tells you

The difference between a good outcome and a bad one is rarely talent. It is whether the business decided what it wanted before hiring someone. Agencies are repeatedly blamed for projects that failed because three partners could not agree on who the customer was. If your internal story is unsettled, spend the first conversation fixing that — with or without us — and every subsequent hour gets cheaper.