About the Brand


AstroUrjaa had a product people wanted and ads that couldn't sell it — cost per first purchase stuck at ₹138. We rebuilt their Meta account around buyer-seeded lookalikes, structured creative testing, and Friday-weighted budgets. Six months later: ₹64.6L in revenue on ₹15.8L spend, with ROAS climbing from 2.6x to 5.5x.


Business Problem


AstroUrjaa is a D2C astrology brand selling to urban 18–35 buyers with high intent and real spending power. The product-market fit was solid — the campaigns weren't. After months of running Meta ads in-house, cost per first purchase was stuck at ₹138 with low volumes. Campaigns were broad, budget was spread flat across the week, and there was no way to tell which ad, audience, or spend level was actually driving purchases.


Diagnosis


Three structural gaps.


  • Targeting relied on interest-based audiences — the in-house default — instead of Meta's own purchase data.


  • Creatives were being swapped without a testing framework, so nobody could tell whether a number moved because of the creative, the audience, or noise.


  • Budget was treated as a fixed setting: spread evenly across days and audiences, ignoring that Fridays consistently outperformed for first purchases in this category. Nothing was wrong with the product; everything was wrong with the structure.


Execution


  • We rebuilt audiences from buyers, not browsers — lookalikes seeded from customers who'd spent ₹2,000+ or ordered 5+ times, later layering in a top-spender cohort.


  • Remarketing ran in parallel to catch high-intent shoppers who don't convert on first touch.


  • Creative testing became structured: four angles tested simultaneously — age-specific hooks, interest-based angles, problem-first framing, budget thresholds — so variables could be isolated.


  • Budget became a weekly lever: weighted toward Fridays, split by audience freshness, reallocated every week.


Business Impact






Testimonial




Your ads are spending. Are they selling? Book a free 30-minute Meta ads audit — we'll show you exactly where the money is leaking.