Turning Dead Traffic Into 5x ROAS with Performance Max
A pan-Indian competitive exam prep platform was burning ₹1.23 lakh on Google Display ads and earning just ₹7,096 in return. First Launch rebuilt the entire campaign strategy — and delivered a 5x ROAS turnaround.
The Client & The Ask
A leading ed-tech platform for pan-Indian competitive exam preparation — and a remarketing campaign that wasn't working.
A leading ed-tech platform helps students across India prepare for competitive exams through live classes, mock tests, interview prep, and personal mentorship programs — accessible via both web and app-based learning environments.
Despite building a sizable base of signed-up users, the platform was struggling to convert them into paying customers through their existing Google remarketing setup. The brand approached First Launch with a clear mandate: nurture top-of-the-funnel sign-ups to purchase using a targeted Google remarketing campaign.
A Campaign Spending Without Earning
The brand's existing Display remarketing campaign had a fundamental mismatch between audience intent and campaign structure — high visibility, near-zero return.
| Audience Targeted | 5–8 Lakh users |
| Total Impressions | 5.2 Million+ |
| Total Clicks | 40,651 |
| Total Ad Spend | ₹1,23,000 |
| Revenue Generated | ₹7,096 |
| Return on Investment | 0.05x ROI |
By migrating to a Performance Max campaign with refined audience signals, First Launch enabled dynamic targeting across all Google inventories — reaching users with demonstrably higher purchase intent and converting them at scale. The result was a consistent 5x outperformance of the original Display setup.
Rebuilding the Strategy from First Principles
Instead of incremental tweaks on a failing setup, First Launch made three decisive structural shifts to the campaign.
- Recognised that broad 6-month sign-up lists lacked purchase intent
- Shifted to using audience lists as intent signals, not rigid targeting parameters
- Allowed the algorithm to expand reach toward users with higher conversion likelihood
- Moved away from standard Display remarketing that had plateaued despite continuous optimisations
- Launched a generic Performance Max setup with multiple products bundled as asset groups
- Used a paid PPC analysis tool to rapidly identify the best-performing networks
- Disabled URL expansion to send users to the most relevant landing pages
- Turned off auto-generated assets to prevent mismatched ad combinations
- Built comprehensive brand exclusion lists to protect organic traffic from paid overlap
Four Phases of Disciplined Optimisation
Performance Max campaigns need a learning window. The real gains came from data-driven refinement at every stage of the campaign lifecycle.
The Data Behind the Turnaround
Performance Max consistently outperformed the baseline Display campaign. Continuous creative refinement and audience signal updates maintained strong upward momentum throughout.
Questions about this engagement
Common questions about the Performance Max strategy and results delivered for this ed-tech brand.
-
The Display Remarketing campaign had run through multiple optimisation cycles without meaningful improvement — generating over 5.2 million impressions but only ₹7,096 in revenue. The fundamental issue was a mismatch between audience intent and campaign structure. Performance Max addresses this by dynamically targeting across all of Google's inventories using audience lists as signals rather than rigid constraints, allowing the algorithm to find users with a higher likelihood to convert.
-
Performance Max campaigns go through a learning phase in the first 2–3 weeks as the algorithm gathers conversion data. During this period, bidding was set to "Maximise Conversions" to build signal volume. Once the campaign stabilised past the 3-week mark, the team switched to "Maximise Conversion Value", which is when the ROAS improvement became measurable and consistent.
-
In this engagement, the audience list of 5–8 lakh users was used as a signal rather than a hard targeting criterion. Performance Max uses this list as a starting reference and expands outward to find similar high-intent users across Google's network. A sizeable seed audience improves signal quality, but the value comes from how the list is used — as intent intelligence, not as a ceiling on reach.
-
The bidding switch from "Maximise Conversions" to "Maximise Conversion Value" after the 3-week learning window was the pivotal moment. This shifted the algorithm's goal from generating as many conversions as possible to maximising the revenue value of each conversion — directly scaling ROAS rather than just transaction volume.
-
Brand exclusion lists were built at the account level to explicitly prevent the Performance Max campaign from bidding on branded search terms. This ensured that users searching for the brand by name were captured organically or via dedicated branded campaigns, while Performance Max focused exclusively on reaching new or non-branded intent audiences.
-
Yes — the core principles apply broadly to any business with a large pool of signed-up but unconverted users. The key is repositioning the audience list as a signal rather than a strict targeting segment, pairing it with smart asset management, and allowing the campaign to optimise toward conversion value rather than raw volume. First Launch has applied similar frameworks across ed-tech, fintech, and SaaS contexts.
Explore More Case Studies
Take a look at First Launch case studies for tips & tricks on how to improve your own brand performance
One content system change took Line of Thought from 105 to 235 followers — with 39% of the audience in...
A greenfield Google Ads build for a US health-tech education platform. First 30 days. 148.5 course sign-ups. 24.79% conversion rate....
See how First Launch helped City Power & Gas rank #1 on Google for 5 competitive energy keywords across New...