AI advertising · In-house teams

Advertising and AI on a Small Team: Running ₹10-50 Lakh a Month Without an Agency

Platform AI now does the bidding and targeting an agency used to bill for. Here is how a small team covers the rest, and when it should not try.

Ajay Kumar • •8 min read
Illustration of a results chart for a small advertising team

Quick answer: can a small team run ₹10-50 lakh a month of ads without an agency?

Yes, if three jobs have clear owners. Platform AI now handles bids, audiences and placements, so a team of 2-3 people can run ₹10-50 lakh a month by owning the numbers, the creative supply and a weekly decision routine. It works badly when nobody owns profit targets, or when creative output cannot keep up.

Why a small team can now run this much spend

A few years ago, an account spending ₹30 lakh a month needed someone adjusting bids and audiences every day. Today the platform does most of that work itself.

Meta's Advantage+ sales campaigns use AI for audience, placement and budget. Google's Performance Max and AI Max do the same across Search, YouTube and Display. The hours an agency once spent inside the account have shrunk.

What remains is real work, but it is different work: deciding targets, supplying creative, checking data and saying no. A small team can do that, if each part has a name next to it.

This article is about the operating model. For the five inputs that decide results, start with our founder's guide to advertising and AI.

Who does what: a 3-role operating model

You need three roles, not three full-time hires. In many brands spending ₹10-50 lakh a month, two people cover all three.

RoleOwnsTime a weekAI helps with
Numbers owner (founder or marketing lead)Profit target, budget, the weekly decisionAbout 2 hoursBidding, budget split, daily summaries
Creative owner (in-house or freelance)3-4 new angles a month, approvals, brand voice2-3 daysFirst drafts, resizing, variations
Tracking and money check (part-time)Pixel and server events, tax, invoicesAbout 1 hourAlerts when store orders and platform purchases drift apart

The hours are our suggestion for a typical account, not a benchmark. The point is the split. If one person holds all three roles, the creative work always loses, because it has no deadline.

Write down a backup for each role as well. A small team's biggest risk is not skill. It is one person on leave during a sale week.

What to hand to AI, what to supervise and what to keep

Sort every recurring task into one of three boxes. The mistake small teams make is automating the third box because it feels like the hardest.

For the middle box, both platforms give you tools. Meta's automated rules check your campaigns, ad sets and ads, then either make a change or notify you. You choose the trigger, the action and what the rule applies to.

Google can apply its recommendations automatically. Its help page says auto-applied recommendations will not increase your budget, and that you can review the queue and dismiss anything that does not fit your goals.

Start in notify mode

Set each new rule to send a notification first. Read what it would have done for 2-3 weeks. Switch it to act alone only if you agreed with it nearly every time.

We compare software and service in more detail in AI Meta ads tool vs a Meta ads agency.

The weekly rhythm: about 5 hours in three sittings

A small team cannot watch the account all day, and should not. Meta says an ad set usually leaves the learning phase after about 50 results in the week after its last significant edit. Every extra edit restarts that count.

  1. Monday, 30 minutes: numbers. Profit after ads against target. Store orders against platform purchases. Note anything odd. Change nothing yet.
  2. Wednesday, 2-3 hours: creative. Review last week's new ads. Brief or approve the next batch. Check that every claim has proof.
  3. Friday, 60 minutes: decisions. One batch of changes: pause, scale, launch. Log each one with its date and reason.
  4. Monthly, half a day: audit. Break-even ROAS (return on ad spend), tracking health, tax on invoices and access rights.

Between sittings, only three things justify opening the account: broken tracking, a rejected ad or payment failure, and runaway spend.

Our daily, weekly and monthly checklist lists the exact checks for each sitting.

Is in-house cheaper than an agency? Run the sum

In-house is cheaper only above a certain spend, because its cost is mostly fixed while an agency fee usually grows with the budget. Work out your own crossing point before you decide.

Compare management cost as a share of spend

Add up what it costs to run the ads yourself, then divide by monthly spend. Do the same for the agency quote.

Worked example: a brand spending ₹30 lakh a month

Step 1: Add up the in-house cost

Example: one performance marketer at ₹1,20,000 a month, freelance creative at ₹80,000 and software at ₹40,000. That is ₹2,40,000 a month. These are round example figures, not market rates.

Step 2: Compare with the agency quote

The agency quotes 10% of ad spend. At ₹30,00,000 a month, that is ₹3,00,000.

Management cost as a share of spend
In-house: ₹2,40,000 ÷ ₹30,00,000 = 8%. Agency: 10%.
Step 3: Find the crossing point
Break-even spend
Fixed in-house cost ÷ agency fee rate = ₹2,40,000 ÷ 0.10 = ₹24,00,000 a month

Below ₹24 lakh a month, the agency is cheaper in this example. Above it, in-house is. At ₹50 lakh, the gap is ₹2,60,000 a month.

Cost is only half the answer. An agency also brings cover for leave, a wider view of what works across accounts, and a creative bench. Put a value on those before you cancel the contract.

For a fuller breakdown of tools, fees and people, see what AI advertising really costs.

When a small team should keep an agency

Going without an agency is a bad idea in four situations. In each one, the gap is not software.

There is a middle path. Keep strategy and numbers in-house, and buy creative production or a monthly audit from outside. Many teams find this cheaper than a full retainer and safer than doing everything alone.

Our rule of thumb

Bring ads in-house only when you can name the numbers owner, the creative owner and a backup for each. If any of the four names is missing, fix that first.

Tools can cover part of the daily watching. ScaleWin, for example, reads the ad accounts and the store every day and sends a WhatsApp summary, so the Monday sitting starts with the facts already collected.

What most guides get wrong about small teams and AI

Most advice on this topic is written to sell either a tool or a retainer. These are the claims that do not survive contact with a real account.

Related reading for teams at the lower end of this range: AI digital marketing for lean teams.

Key takeaways

  • Three roles are enough: one person owns the numbers, one owns creative, and one checks tracking and money part-time.
  • Hand bids, placements and audience expansion to platform AI. Keep offers, claims, targets and creative approval with people.
  • Start every automated rule in notify mode. Let it act alone only after it has been right for a few weeks.
  • Plan about 5 hours a week in three sittings. Daily edits keep resetting the learning phase.
  • In our example, in-house becomes cheaper than a 10% agency fee above ₹24 lakh a month of spend. Cost is not the only test.

Advertising and ai: frequently asked questions

How many people do I need to run ads in-house with AI?

Three roles, which two people can often cover: someone who owns the numbers and the weekly decision, someone who owns creative, and a part-time check on tracking and billing. Platform AI handles bids, audiences and placements, so nobody needs to adjust those by hand each day.

At what ad spend does in-house become cheaper than an agency?

Divide your fixed monthly in-house cost by the agency's fee rate. In our example, ₹2,40,000 of in-house cost against a 10% fee gives a crossing point of ₹24 lakh a month. Your own figures will differ, so run the sum with real salaries and quotes.

Which advertising tasks should I not automate?

Keep the offer, prices, claims, profit targets and the final approval of creative with people. These depend on facts the platform cannot see, such as your margins and legal duties. Automate bids and placements, and supervise pausing, scaling and platform recommendations.

Are Meta automated rules safe to use?

They are safe if you start carefully. Meta's automated rules can either notify you or make a change when a trigger is met. Run each new rule in notify mode for a few weeks, check that you would have agreed with it, and only then let it act by itself.

Should I turn on Google's auto-apply recommendations?

Use it selectively. Google says auto-applied recommendations will not increase your budget, and you can review the queue and dismiss items. Start with maintenance-type recommendations, read what was applied each week, and leave anything that changes targeting or bidding goals for manual review.

How much time does it take to manage ₹10-50 lakh a month of ads?

About 5 hours a week for the numbers and decisions, in three sittings, plus creative production time. Add half a day each month for an audit of targets, tracking, tax and access. More time in the account rarely helps, because frequent edits restart the learning phase.

Sources

Ajay Kumar
Founder, ScaleWin · 12+ years in performance marketing

Ajay is the founder of ScaleWin. He works hands-on with Meta Ads, Google Ads and SEO for Indian D2C and service brands, and built ScaleWin so that the daily work of a performance marketer - reading numbers, pausing losers, scaling winners - happens automatically.

Drafted with AI assistance. How these articles are written

Let ScaleWin read your ad data every day

ScaleWin checks your Meta and Google Ads daily, judges every ad on profit and tells you exactly what to change.

Start free, no card

Get marketing tips that grow your business

One practical email a week on ads, SEO and growth. No spam, no fluff. Unsubscribe anytime.