How The Manga Store Went from ₹4L/Month and Eight Failed Agencies to ₹8 Lakh in a Single Week

The clearance sale had been live for a few hours. Ad spend: ₹8,000. Revenue already in: ₹1.25 lakh. ROAS: 15x.

The Arlox team immediately pushed the daily budget to ₹80,000 and aimed for ₹4 lakh by end of day.

Thirty-three days earlier, The Manga Store had onboarded with a clear brief: they were frustrated. Seven to Eight previous agencies had burned through budgets, attracted the wrong audiences, confused the brand's data, and left the founder Divyansh Pandey deeper in operational debt than when he started. The brand was doing ₹4L a month with a ROAS of 2.5x. But the sales weren't profitable. They weren't even reaching the right people.

Five weeks after going live with Arlox, the clearance sale ran for six days. Total revenue: over ₹8 lakh. Final-hour ad ROAS: 10x. Overall clearance sale ROAS: 3.4x. April target set immediately after: ₹15 lakh a month.

This is the story of how an anime streetwear brand niche, brand-protective, and burned by generic marketing found its audience, its hero products, and its scale.

BRAND SNAPSHOT

• Industry: Anime & Manga Streetwear (D2C, India)

• Products: Premium anime-themed apparel — graphic tees, hoodies, joggers, sweatshirts, jackets, jerseys (Vinland Saga, Devilman Crybaby, Demon Slayer, Sasuke, Akira, Kaiju, and more)

• Geography: India (pan-India D2C, Shopify)

• Starting Revenue: ₹4L/month, ROAS 2.5x (unprofitable due to wrong audiences)

• Milestone Achieved: ₹8L+ in a single week; 15x ROAS on sale launch morning; 10x ROAS in final-hour campaign; April target ₹15L/month

• Services: Meta Ads, Creative Strategy, Website CRO, Conversion API Integration, Microsoft Clarity, Partial COD Implementation, Fake Order Filtering, Audience Rebuilding

THE PROBLEM

The Manga Store had something rare in the Indian D2C space: a genuinely differentiated product in an underserved niche. Anime-themed premium streetwear — not mass-market prints, but carefully designed, limited-run apparel for an audience that takes its fandom seriously. Demand existed. Community existed. The products were real.

But the brand's marketing history was a case study in misalignment. Three to four agencies over multiple engagements had run campaigns that drove traffic — just not the right traffic. The analytics data had been polluted by non-target audiences responding to broad, generic promotions. The campaigns increased spend and debt simultaneously, without building a reliable customer base.

By the time Divyansh came to Arlox, the frustration had a specific shape. It wasn't just "ads aren't working." It was: "Every agency that's run ads for us has brought in people who aren't our audience, wrecked our pixel data, and left us worse off than before we started."

The brand was revenue-generating but not profitable. The data was noisy. The audience was undefined in the system. And the founder had watched budget after budget disappear into campaigns that didn't understand what The Manga Store was actually selling — or who it was selling to.

WHY IT WAS HAPPENING

Three compounding failures sat underneath the surface:

1. Audience misalignment had corrupted the entire data layer. When previous agencies ran broad campaigns to generate impressions or cheap traffic, they attracted audiences with no genuine interest in anime streetwear. Every click from the wrong person taught the algorithm the wrong thing. By the time Arlox inherited the account, the pixel data was muddy — filled with signals from people who'd never buy. Rebuilding the audience from scratch wasn't optional. It was the only path forward.

2. No conversion infrastructure existed. The brand had high add-to-cart numbers but low purchase conversion — a pattern visible from the first week of data. Customers were arriving, showing intent, and then leaving before completing the purchase. No abandoned cart recovery was in place. The checkout experience had not been optimized. The Conversions API (CAPI) had not been connected, meaning the pixel was operating with incomplete data even on legitimate traffic. The gap between interest and purchase was wide open.

3. Fake COD orders were bleeding cash and inflating the numbers. The Manga Store was processing a significant volume of cash-on-delivery orders that turned out to be fraudulent — fake orders placed by non-serious buyers. In the early weeks, the founder flagged a batch of suspicious COD orders worth ₹58,000. These fake orders created a false revenue picture, tied up inventory, and added fulfillment costs that would never be recovered. Without a filter in place, scaling ad spend would have scaled the problem.

THE SOLUTION

Onboarding happened on February 20, 2025. Ads went live on February 25 five days after the first conversation.

Mythos — Creative Advantage:

The brand's product catalog was immediately promising from a creative standpoint. Anime streetwear writes its own briefs — every product has an existing cultural reference that resonates deeply with the target audience. The challenge wasn't finding a story. It was finding which story converted.

The creative team began testing across multiple product lines simultaneously: hoodies, joggers, sweatshirts, jerseys, graphic tees. Formats tested included lifestyle video, static images, carousels, and anime-inspired POV hooks. The first major insight came early: the static image outperformed everything, hitting a ROAS of 6x on one campaign in early March. This wasn't a fluke — it pointed to an audience that responds to clarity and product aesthetics over production-heavy content.

The second insight came from the data itself. Vinland Saga Shirt and Devilman Crybaby consistently ranked as the top-performing products — both delivering 2x to 3.3x ROAS on winning ad sets. Once identified, these two became the creative pillars. The team doubled down on both: new reel variations, offer statics, product-specific landing pages instead of collection pages. Every piece of market intelligence from the founder was turned into a creative brief immediately.

When the clearance sale launch was planned for March 24, the team had already built a creative infrastructure — tested hooks, known winners, product-specific formats — that could handle aggressive scale. The sale wasn't improvised. It was the output of five weeks of systematic testing.

Sentinel — Scientific Media Buying:

Daily performance reports began on day one. Not weekly summaries — daily breakdowns with specific notes on which ad sets were winning, which were being cut, what the ROAS trajectory looked like, and what the immediate action plan was for the next 24 hours. The founder always knew exactly where the account stood.

Budget decisions followed the data, not a predetermined calendar. When early campaigns showed high add-to-carts but weak conversion, budgets were held steady while CRO fixes were implemented on the website. When Vinland Saga and Devilman Crybaby began consistently delivering 2x+ ROAS, those ad sets were scaled while underperformers were cut.

The clearance sale media buying demonstrated what systematic scaling actually looks like. Sale opened with ₹8,000 in spend and generated ₹1.25 lakh by mid-morning at 15x ROAS. The team moved budget to ₹80,000 per day immediately, with the goal of ₹4 lakh by end of day. Over the six-day sale window, ₹5.8 lakh was generated in the first three days alone. The final-hour urgency campaign — a last-24-hours push with all remaining inventory on a single landing page — ran at 10x ROAS. The overall clearance sale ROAS finished at 3.4x, exceeding the internal target.

When the Microsoft Clarity integration revealed that customers were dropping off at specific points in the checkout flow, the team flagged it immediately and prescribed fixes. Data didn't just live in reports — it drove the next action every single day.

Vault — Brand Value Engine:

The conversion problem from week one had a clear set of solutions, and the team prescribed them as an operational checklist.

First: connect the Shopify Conversions API (CAPI) to Meta. Without server-side event tracking, the pixel was operating on browser data alone — missing a significant share of actual purchase events. The team walked the founder through the full CAPI integration step by step, restoring data accuracy at the pixel level.

Second: fake COD orders were costing the brand real money. The solution was a partial payment system for COD orders — requiring a small upfront payment to filter out non-serious buyers. The implementation was confirmed on March 10. Fake order volume dropped immediately, protecting both cash flow and inventory.

Third: Microsoft Clarity was integrated to give visibility into actual user behavior on the website. This surfaced the specific drop-off patterns — where customers were abandoning — that the ROAS data alone couldn't explain. Insights from Clarity informed creative decisions (product-specific landing pages over collection pages) and checkout optimizations.

Fourth: the clearance sale itself was architected as a conversion event, not just a promotion. A dedicated landing page was built at themangastore.in/pages/clearance-sale with a countdown timer, all remaining products in one view, and urgency copy. The founder and the team coordinated the exact launch timing down to midnight.

THE RESULTS

• ₹8L+ in a single week — March 24–29, 2025, from a standing start with zero prior D2C scale

• 15x ROAS on clearance sale launch morning — ₹1.25L in revenue from ₹8K ad spend in the first few hours

• ₹5.8L in 3 days — clearance sale revenue generated March 24–27

• 10x ROAS on the final-hour campaign — last-24-hours urgency push with remaining inventory

• 3.4x overall clearance sale ROAS — exceeding the team's internal target

• 6x ROAS on static image creative — identified as the first breakthrough ad in early March

• Devilman Crybaby hitting 3.3x ROAS and Vinland Saga at 2.3x+ consistently — hero products confirmed and scaled

• Fake COD order bleeding stopped — partial payment system implemented, protecting real revenue and inventory

• ₹48.8K single-day milestone (March 9) before the sale even launched — the largest non-sale day in brand history

• April 2025 target: ₹15 lakh/month — set immediately after the clearance sale results were confirmed

The founder's message after week three: "Since you guys took charge of our marketing, we've witnessed your effective efforts daily, consistently pushing our sales higher and higher. Having worked with 3-4 agencies in the past, we've never seen this level of constant improvisation and adaptability, which truly sets you guys apart. Your commitment and strategic approach make us believe in you as the right partners to help us achieve our goals."

LESSONS FOR SIMILAR BRANDS

Corrupted pixel data is not a minor inconvenience — it's a strategic liability. When previous agencies run broad campaigns to cheap audiences, every wrong click teaches the algorithm the wrong pattern. The damage isn't just in the wasted spend — it's in the polluted data that future campaigns will learn from. Cleaning and rebuilding the audience layer is not optional. It is the foundation everything else is built on.

Niche brands need niche-accurate creative, not scaled-down mass-market ads. The Manga Store's audience is specific: they know their anime references, they have strong aesthetic standards, and they do not respond to generic e-commerce formats. The static image outperformed everything else not because static is inherently better — but because it led with product aesthetics in a format that respected the audience's intelligence. Understanding who you're selling to is the first creative brief.

A clearance sale is a revenue lever, not a distress signal. The clearance sale wasn't run because the brand was struggling — it was run because the team had built enough creative infrastructure to scale aggressively, had identified winning products, and saw the opportunity to move inventory at strong ROAS while generating a significant cash injection. The 15x ROAS on the launch morning and 3.4x overall were the output of five weeks of disciplined preparation, not luck.

Fake COD orders are a scaling tax you cannot afford to ignore. At ₹4L/month, fake orders are an annoyance. At ₹15L/month, they become a cash flow crisis. Implementing partial payment for COD is not about losing customers — it's about filtering buyers from browsers and protecting the inventory reserved for real orders. The partial payment filter should be one of the first infrastructure decisions any Indian D2C brand makes before scaling.

Daily data is the difference between fast learning and slow guessing. The gap between a good agency and a great one is how quickly they look at the data and how specifically they act on it. Every daily report for The Manga Store came with a named winner, a named underperformer, and a specific action planned for the next 24 hours. That cadence compressed months of learning into weeks.

CHALLENGES WE FACED

High add-to-cart rates, low purchase conversion in the early phase. In the first two weeks, the data showed consistent patterns of strong add-to-cart activity with conversion rates significantly below expected levels. This pointed to a funnel problem — customers interested enough to add a product weren't being given a strong enough reason to complete the purchase. The team prescribed a multi-pronged fix: Conversions API integration for better pixel data, Microsoft Clarity for session recording and drop-off analysis, product-specific landing pages instead of collection pages, and a partial COD payment requirement. The conversion issue was not solved in one move — it was diagnosed, treated, and monitored until the metrics moved.

Fake COD orders masking real performance. In early March, the founder flagged a cluster of suspicious COD orders representing ₹58,000 in reported revenue that would never convert to actual cash. This was the visible tip of a broader problem: without a pre-purchase filter, any scaling of ad spend would also scale the fake order rate. The team recommended and implemented a partial payment system for COD within a week of the issue being surfaced. Post-implementation, order quality improved measurably.

Ads running through the wrong Instagram page. Early in the campaign period, the founder identified that some ads were serving through a secondary account rather than the brand's primary page. This meant the brand-building impression value of those ads was accruing to the wrong account, and the brand identity was inconsistently presented in the feed. The team paused the secondary account's ads, implemented a review process for creative approval before launch, and confirmed with the founder that all subsequent ads would go through the correct page.

Inventory selling out at peak sale demand. The clearance sale generated 2,300 add-to-cart actions in a concentrated window — but with most products out of stock by that point, the conversion rate on those carts was artificially suppressed. Microsoft Clarity confirmed that customers were seeing out-of-stock statuses and dropping off mid-funnel. The team responded by creating a final-campaign push that consolidated all remaining available products onto a single landing page, with urgency copy and a countdown timer — and that final push ran at 10x ROAS. Inventory planning for the next launch cycle was discussed immediately in the post-sale call.

BELIEFS CHANGED

"Previous agencies proved paid ads don't work for niche brands." The Manga Store came in with Seven to Eight failed agency experiences and a belief that digital advertising fundamentally misunderstood niche, brand-protective categories. The clearance sale — 15x ROAS on launch morning, 3.4x overall, ₹8L in a week — didn't just refute that belief. It showed that niche audiences, when targeted correctly, convert with exceptional efficiency. The brand's specificity was an asset, not a handicap. The previous agencies' failure was not a verdict on paid advertising. It was a verdict on generic advertising.

"A ROAS of 2.5x is the ceiling for this category." Before Arlox, the brand was running at 2.5x ROAS while still losing money — because the audiences receiving those ads weren't the right ones, and the conversion infrastructure wasn't converting them efficiently. When the data was clean, the creative was right, and the funnel was fixed, 6x ROAS on a static image was achievable in week two. 15x ROAS at sale launch was achievable at week five. The ceiling wasn't the category. It was the execution.

"We have to discount to sell." One of the founder's early concerns was how to move inventory without compromising the brand's premium positioning. The answer wasn't permanent discounting — it was a structured, timed clearance event that created genuine urgency without training the audience to wait for sales. The clearance sale at 3.4x ROAS demonstrated that urgency-driven promotions, done correctly, are compatible with brand health. The regular-day record — ₹48.8K on March 9, before the sale — showed that the brand could generate strong revenue at full price too.