Managing ₹10Cr+ Monthly Ad Budgets: How Performance Agencies Buy Media at Scale

Managing ₹10L/month in ad spend and managing ₹10Cr/month are not the same activity. The systems, team structure, risk management, approval processes, and daily cadence are fundamentally different. A ₹10L/month operation can run with two people making real-time decisions. A ₹10Cr/month operation requires a 12-15 person team with defined governance, budget allocation protocols, pacing management, creative production velocity, account structure, and reporting infrastructure. We’ve guided brands from ₹5Cr to ₹12Cr/month spend, and the transition is always painful—not because the math changes, but because the operational model has to completely rewire itself.

The Scale Media Buying Framework

Modern agencies managing ₹10Cr+/month operate using a structured framework with six components:

1. Governance Structure (Who approves what at what spend level)

At ₹10L/month, one person might approve all budget changes. At ₹10Cr/month, you need a governance layer that defines decision rights by spending authority. Here’s a typical structure:

• Level 1 (Under ₹1L daily change): Campaign manager approves. Day-to-day optimization, creative swaps, minor bid adjustments. Approved by individual team member, no escalation.

• Level 2 (₹1L – ₹5L daily change): Team lead approves. Larger budget reallocations, new audience launches, bid strategy changes. Requires written brief (why, expected impact, risk).

• Level 3 (₹5L – ₹15L daily change): Director approves. New channel launches, major budget reallocation across channels, strategic pivots. Requires 24-hour notice, data-driven justification, ROAS/CAC impact modeling.

• Level 4 (Over ₹15L daily change): Executive + Finance approval. Major reallocation (e.g., shifting ₹3Cr from Meta to Search), pausing channels, portfolio restructuring. Requires board-level visibility, quarterly planning, external data (MMM, incrementality tests).

We run this at Clicksbazaar. Campaign managers optimize within their daily budgets. Team leads make tactical channel shifts. Directors handle strategic reallocations. CEO/COO review any decisions above ₹15L. This prevents chaos (everyone has authority) while maintaining control (nobody can unilaterally blow the budget).

2. Budget Allocation Protocols (Channel caps, reallocation triggers)

At scale, you need systematic rules for how the budget flows across channels. Otherwise, you’ll have teams competing for spend, and you’ll lose coherence.

These caps prevent overconcentration. Without max caps, teams naturally optimize toward their highest-ROAS channels, and you end up with 60% of budget in one channel—leaving others underfunded. Without min spend thresholds, low-performing channels get starved and never get enough volume to scale or recover. Without reallocation triggers, you don’t know when to move the budget (you’re relying on intuition).

This structure creates autonomy: teams know their budget range and when escalation is required. They also know: if my channel underperforms by X%, I can expect budget cuts.

3. Pacing Management (Daily vs weekly optimization)

At ₹10L/month, you check spend and pacing daily. At ₹10Cr/month, you need automated pacing management because manual oversight becomes impossible.

Daily spend at ₹10Cr/month = ₹33L/day (roughly). If one channel underpaces by 10%, that’s ₹3.3L in lost spend—which compounds over a week to ₹23L in lost volume. Automated pacing rules prevent this:

• Spend velocity targets: Each channel has a daily spend target (e.g., Google Search targets ₹65L/day ± 5%). If actual spend deviates >5%, automated rules trigger (increase bids if underpacing, decrease if overpacing).

• Weekly rebalancing: Every Monday, reconcile the prior week’s actual spend vs. target. If a channel underpaced >7%, escalate to team lead for investigation and correction.

• Holdout window management: On months with holidays or events that disrupt normal patterns, create “no optimization” holdout windows. During Diwali or year-end, don’t run automated pacing rules—switch to manual oversight.

Most ₹10Cr/month operations use platform-native pacing tools (Google Ads Campaign Budget Optimization, Meta campaign budgets with pacing) combined with custom dashboards that track actual vs. target daily spend. When pacing tools start failing (which happens when budgets are large enough to hit platform limitations), you move to programmatic buying via DSPs.

4. Creative Production Velocity (Matching spend growth)

At ₹1Cr/month across channels, you might need 20-30 new creative assets per month to prevent fatigue. At ₹10Cr/month, you need 150-250 per month.

Creative velocity requirements:

Most brands at ₹10Cr+ have a dedicated creative studio (in-house or agency partner). They run continuous creative testing cycles. Production happens in 2-3 day turnarounds, not 2-week turnarounds. The creative team produces variations, not entirely new concepts—user-generated content adaptations, testimonial videos, product demonstrations with different messaging angles, lifestyle imagery in different scenes.

We worked with a ₹8Cr/month ecommerce brand. Their in-house creative team produced 15 assets per week. That was insufficient. Within six weeks at high spend, creative fatigue set in and CTR declined 24%. We helped them scale to a 4-person team + outsourced UGC production. New target: 35-40 assets per week. After 12 weeks, creative fatigue was manageable again.

5. Platform Account Structure (Organizing massive spend)

At ₹10L/month, one or two Google Ads accounts might suffice. At ₹10Cr/month, account structure becomes critical.

Typical structure at scale:

• Google Ads: 2-3 accounts (splitting by business unit, geography, or product line). Each account: ₹3-4Cr/month. Each account has 4-5 campaigns by type (Search, Shopping, YouTube, Performance Max). Avoids hitting account-level limits and allows team specialization.

• Meta: 2-4 business accounts (each tied to a different ad account for performance isolation). Spread spend across multiple ad accounts to avoid throttling. Meta sometimes limits spend velocity on individual ad accounts, so diversification helps.

• Google Campaign Manager / DSP: At ₹10Cr+ spend, many agencies use programmatic buying infrastructure. Fluent, The Trade Desk, or DV360 manage real-time bidding and audience buying at scale more efficiently than individual platform interfaces.

• Reporting/Analytics: Centralized data warehouse pulling data from all accounts (Google Ads, Meta, CRM, ecommerce). Single source of truth for spend, conversions, and ROAS across all channels.

Splitting spend across multiple accounts also reduces risk. If one Google Ads account gets suspended (policy violation), other accounts keep running. One Meta ad account gets throttled; others maintain velocity.

6. Reporting Infrastructure (Daily dashboards, weekly syncs)

At ₹10Cr/month, you need automated reporting infrastructure. Manual reporting is impossible.

Standard reporting stack:

• Daily dashboard: Automated pull of spend, conversions, and ROAS by channel and campaign type. 24-hour lag is acceptable. Used for trend spotting and pacing monitoring.

• Weekly performance report: Blended metrics (overall ROAS, blended CAC, media mix). Used for executive oversight. Includes prior-week comparisons and annotations (creative changes, bid strategy shifts, external events).

• Ad hoc analysis requests: BI tool (Looker, Tableau, Supermetrics) available for team leads and directors to slice data by audience, creative, campaign, geography. Self-service analytics reduce bottlenecks.

• Monthly reconciliation: Cross-check platform spend totals against finance records. Critical at scale because small discrepancies compound. A 0.8% platform reporting discrepancy = ₹80L/month difference. Reconciliation meetings happen monthly to ensure accuracy.

Most ₹10Cr/month operations use a combination: Google Data Studio (for daily dashboards), a BI tool (for deeper analysis), and manual weekly syncs with leadership (10-15 minutes, 5 key metrics).

The Maturity Model: From ₹1L to ₹10Cr/Month

How operations scale and evolve:

Real Example: The ₹3Cr to ₹8Cr Transition

A performance-driven D2C brand was spending ₹3Cr/month profitably (blended 2.8x ROAS). They secured growth capital and wanted to hit ₹8Cr/month within six months. This required overhauling their operating model.

Month 0 (Planning): They had a media buyer and a campaign manager. We designed a team structure: add a second campaign manager (Google Search focus), add a creative producer, and hire a BI analyst. Implemented governance rules (campaign manager approves <₹5L changes; media buyer escalates >₹5L; CEO approves >₹20L).

Month 1-2: Governance was in place but not enforced. Decisions still happened ad-hoc. We ran three “governance violation” incidents (team members making >₹10L reallocations without approval) before leadership took it seriously. By month 2, governance rules were sticky.

Month 2-3: Creative production was the bottleneck. The new producer was making 8 assets per week. Target was 24/week (for ₹5Cr/month spend). CTR started declining because creative was burning out. We onboarded an external UGC production partner (another ₹2.5L/month cost) to fill the gap. Production hit 35/week by month 3.

Month 3-4: Pacing became unstable. ₹5Cr/month means ₹16.7L/day target. Some days they’d do ₹13L (underpacing), other days ₹19L (overpacing). Causing weekly revenue variance of ₹30-40L. Implemented automated pacing rules: any channel underpacing >8% triggers automated bid increase. Within 10 days, pacing variance dropped to ±4%.

Month 4-5: Account structure was consolidating spend into two Google accounts (each hitting ₹3.5Cr/month). Google’s infrastructure started showing stress—bid changes took longer to propagate, CPA spikes appeared without clear cause. Moved to a three-account structure, spread to ₹2.5Cr/account. Stability improved.

Month 5-6: Hit ₹8Cr/month target. Blended ROAS: 2.72x (vs. 2.8x baseline). 3% degradation—better than the 12-18% typical for this scale jump. Why? Disciplined governance, creative velocity, pacing management, and account diversification prevented cascading problems.

Common ₹10Cr-Scale Failure Points

We’ve seen these patterns repeatedly at ₹10Cr+ spend:

Failure 1: Governance without authority. Teams are told “get approval before changes >₹10L” but nobody actually enforces it. Team members make decisions and ask permission after the fact. Result: budget discipline breaks down, conflicting decisions create chaos, metrics start trending down.

Fix: Enforce governance strictly for the first 4 weeks. Make an example of the first violation—public feedback, formal escalation, maybe not a firing offense, but certainly a “this matters” signal. After 4 weeks, it becomes a habit.

Failure 2: Creative production bottleneck. Brands assume they can run ₹10Cr/month with 10-15 creative assets per month. They’re wrong. CTR collapses, CPA spikes, and growth stalls. By the time they realize creative velocity is the problem, it’s month 3-4 of the growth plan.

Fix: Front-load creative investment. Budget for 60-80 assets in month 1, before scaling begins. Run creative testing cycles. Hire producers before you need them (they take 2-4 weeks to onboard and get up to speed).

Failure 3: Account structure misalignment. Brands cram too much spend into single accounts. Google throttles one account; the whole portfolio grinds to a halt. Meta hits learning-phase-reset curves because all spend is in one ad account and budget increases cascade.

Fix: Plan account structure before you hit ₹7Cr/month. Transition to multi-account structures at ₹5Cr/month, before you’re forced.

Failure 4: Reporting lag. Leadership doesn’t see metrics until Thursday/Friday each week. By then, it’s Monday and the week’s performance is baked in. Poor decisions get made Thursday afternoon that affect Friday, and by the time feedback comes, it’s too late.

Fix: Daily dashboards with 24-hour data. Not perfect, but good enough for real-time trend spotting. Weekly deep dives can happen Friday, informing Monday adjustments.

Failure 5: Team communication breakdown. At ₹1Cr/month, everyone’s in the same room and talks daily. At ₹10Cr/month, teams specialize (Search team, Social team, BI team). Without formal communication, they optimize locally and create conflicts (Search increases bids, Social decreases budgets to maintain overall pacing—they work at cross purposes).

Fix: Weekly sync meetings. 30 minutes max. One metric reviewed: did we hit weekly pacing targets by channel? Why or why not? What’s the plan for next week? Keeps teams aligned.

▶ PRO TIP: The Spend Velocity Curve

At scale, watch for a pattern: as you increase monthly spend from ₹5Cr to ₹8Cr, don’t increase all channels equally. Channels have different “elasticity” to spend increases.

Google Search can usually handle 40-50% spend increases without performance degradation (it has lots of search volume). Meta typically degrades at 25%+ increases (audience saturation). YouTube is somewhere in between (30-35%).

When scaling from ₹5Cr to ₹8Cr (+60%), allocate:

• Google Search: +45% (stays healthy)
• Meta: +28% (protects CPA)
• YouTube: +35% (balanced)
• Shopping: +48% (usually has room)
• Programmatic: +52% (scales well)

This isn’t uniform scaling—it’s channel-matched scaling. It requires knowing each channel’s elasticity curve. That’s why running incrementality tests quarterly at scale is critical.

Key Takeaways

Managing ₹10Cr+/month in ad spend requires operational discipline that managing ₹10L/month doesn’t. You need: structured governance (approval rights by authority level), budget allocation protocols (channel caps and reallocation triggers), pacing management (automated systems, not manual daily checks), creative production velocity matching spend growth, account structure diversification, and reporting infrastructure. Most brands underestimate the operational complexity. They hire three extra people and assume that’s enough. They’re wrong. You need team specialization, systems automation, and governance rigor. The good news: once you build this infrastructure at ₹10Cr/month, managing ₹20Cr/month just means scaling each component. The hard part is getting the model right the first time.