Start with the funnel, not the service list
Agency websites list services alphabetically, which is useless for buying. Every service exists to do one of five jobs, and once you sort them that way you can see which job you're currently paying nobody to do.
Most scope mistakes are visible in this table. A company with no search demand buying SEO. A D2C brand pouring everything into demand creation while its product pages leak. A B2B firm with good traffic and no tracking, so nothing survives a budget meeting.
| Stage | The job to be done | Services that live here |
|---|---|---|
| Demand capture | Someone is already searching for what you sell. Be there, and be the one chosen. | SEO, Google Ads search, Google Business Profile, marketplace listings, app store optimisation. |
| Demand creation | Nobody is searching yet. Give the problem a name and attach your brand to it. | Organic social, paid social, YouTube, influencer work, digital PR, content marketing. |
| Conversion | Traffic arrives and most of it leaves. Fix the part where the decision happens. | Landing pages, CRO, web design and build, creative, copywriting. |
| Retention | Cheaper than acquisition, and routinely the last thing anyone staffs. | Email and lifecycle, WhatsApp flows, retargeting, community, CRM hygiene. |
| Measurement | Sits under all four. Without it, every budget decision is a guess with a chart attached. | Analytics setup, conversion tracking, attribution, dashboards, call tracking. |
Paid, organic and creative run on different clocks
Agencies bundle these three because they sell better together. They behave nothing alike, and mixing up their timelines is the most expensive mistake a founder makes in year one.
- Paid rents attention. Results arrive in days and vanish the day the card stops. The right first channel when you need to learn what converts, and the wrong only channel forever.
- Organic — SEO, social, content, PR — compounds. Three to six months to real movement, then it keeps working after the invoice stops. Which is exactly why it's underfunded.
- Creative multiplies both. The same ad budget with better creative buys more. The same article with a sharper angle earns more links. Agencies price creative as a cost line; it behaves like a multiplier.
The analytics work nobody puts in the pitch deck
The least glamorous thing an agency does, and the thing that makes every other line defensible. It's also the one most often quietly skipped, because you can't see it on a screen the way you can see a new landing page.
The work is unremarkable: GA4 configured properly, conversion events that match real business events rather than page views, UTM discipline so channels stop stealing credit from each other, Search Console connected, call and form tracking, and a dashboard showing the one number the business cares about.
Without it, you can't tell whether last month's ₹1,50,000 worked, so you renew on vibes and cut on panic. Agencies with weak measurement have an easier life — worth remembering when nobody raises it.
What each service line costs in India
Bands below are wide because quality, city, seniority and scope vary far more than the service label does. Treat them as the shape of the market, not a quote — and remember a low quote usually means fewer hours, not better efficiency.
| Service line | Typical band | Skip it when |
|---|---|---|
| SEO | ₹40,000–₹3,00,000 / mo | Nobody is searching for your category yet. Create demand first. |
| Google Ads / paid search | ₹40,000–₹1,50,000 / mo, or 10–20% of spend | Your monthly spend is under roughly ₹1.5 lakh — the fee eats the return. |
| Paid social | ₹40,000–₹1,50,000 / mo | You have no creative pipeline. Paid social starves without fresh assets. |
| Organic social & content | ₹30,000–₹1,00,000 / mo | Your buyers don't decide on social — most industrial B2B, most local services. |
| Digital PR & link building | ₹50,000–₹2,50,000 / mo | The site has unresolved technical or content problems. Links won't rescue it. |
| Creative & design retainer | ₹40,000–₹1,50,000 / mo | You aren't running enough paid or social volume to consume the output. |
| Web design & build | ₹1,00,000–₹15,00,000 one-off | The current site converts fine and you're rebuilding because you're bored of it. |
| Analytics & tracking setup | ₹50,000–₹3,00,000 one-off | Almost never. This is the cheapest line here and it makes the rest measurable. |
What to skip under ₹5 lakh a year
Five lakh a year is roughly ₹42,000 a month. That funds one channel properly. Spread across four it funds nothing — ten thousand rupees of attention each, enough to produce reports and not enough to produce customers.
Choosing is uncomfortable, so most founders don't, and the budget dissolves. The honest split at that level:
Fund these
- One acquisition channel, matched to how your buyers behave. Search if demand exists, paid social if it doesn't.
- A site you can edit yourself, without a developer and a two-week queue.
- Analytics and conversion tracking, once, properly. It makes everything else arguable.
- Enough creative to feed that channel. Starved channels underperform regardless of who manages them.
Skip these until the budget doubles
- A second and third paid channel. Fragmented budget, fragmented learning, nothing reaching efficiency.
- Influencer programmes and brand video. Expensive to produce, slow to read, hard to attribute at this scale.
- CRO. Below roughly 1,000 relevant sessions a month there isn't enough traffic to test to significance.
- Marketing automation platforms. A licence fee to send emails nobody has time to write.
- Awareness campaigns. At this budget, the one line where you genuinely can't tell if it worked.
How to assemble a scope instead of buying a bundle
Bundles are priced for the agency's convenience, not your funnel. Not a scandal — just worth knowing before you sign one. Building your own scope takes an afternoon.
- Name the constraint. Not enough traffic, traffic that doesn't convert, or leads that don't close. Each points at a different row of the table above, and only one is your problem this quarter.
- Buy the stage that fixes it. One line funded properly, rather than four funded thinly.
- Add measurement first, always. Cheapest item on the list, and it's what lets you fire the rest with evidence.
- Set the review date before you start. Ninety days for organic, thirty for paid — and write down today's number now.
- Leave the next stage unfunded on purpose. If the first works, you'll have the budget and the evidence to expand. If it doesn't, you've lost one line, not four.