Rask is a video content translation and dubbing SaaS platform, allowing you to localize video content to any language, apply lip-sync, make it look and sound cool.
We partnered with the Rask.AI, marketing team in August 2023 to scale their Google Search Ads effectively and quickly—without hurting overall marketing performance.
Google Search Ads is the money prinitng machine when it's cooked right, with patience and precision. We took our slow, careful approach and made it work.
DISCLAIMER:
We operate under strict NDAs, and this case is no exception. We agreed with the marketing team that we can mention the case and some logic behind, but not disclose any details.
Then their internal marketing team was rearranged, our cooperation was put on hold. And I’m not even sure if the current team is okay about sharing anything at all.
Thus, no business-sensitive info here. I'll focus on our general approach instead.
Challenges
An ultra-competitive AI market at its peak.
Google Ads requires time and budget to collect meaningful data, test, and "train" campaigns.
Low trust in external marketing teams and agencies due to past negative experiences.
The Approach
Full analytics tech stack review and rebuild.
Test campaigns to ensure valid analytics and data flow.
Conversion event and revenue attribution optimization.
Simultaneous campaign strategy testing at minimal budget.
Campaign segmentation by:
Conversion and revenue attribution timeframe.
Geo & language targeting (refining from global English to local markets).
Keyword groups: Use Cases, Competitors, Pains, Values, Hype.
Keyword match types: Broad, Phrase, Exact.
Parallel scaling of proven strategies while testing new ones.
Ongoing competitive, geo and language-specific research and analytics.
The Essence
We don’t rush to scale budgets, even when millions are available (we’ve had fintech cases like this). This caused some tension early on, as the team wanted fast scaling.
We don’t blindly copy competitors strategies, even when we know they work, because:
Traffic behavior is never identical, even for similar products.
In-depth testing is critical to understanding how budget, targeting, optimization, and UX correlate.
Only our own campaign data is valid for making decisions.
We analyze the entire customer lifecycle—from ad impression to landing page, signup, and even second or third subscription renewal (for SaaS) to uncover dependencies and patterns between campaigns, pricing, and UX.
Why so nerdy and slow?
People ask this a lot. Many agencies, especially when working with well-funded startups or established brands, go all-in.
Launching big campaigns with $XK/week budgets right away.
I respect that approach and learn from those who do it, but it comes with trade-offs.
The fast approach has its advantages:
Faster performance data collection and campaign learning.
Large paid Google Ads volume can boost organic traffic.
A surge in new leads can create viral loops through referrals and social media.
More social buzz = potentially higher ad click-through rates for at Google search result pages.
But it also has immediate and long-term risks:
High Volume = Less Precision
Google Ads traffic is finite. You can’t buy more than what’s available.
Large-scale campaigns can't be as precisely targeted or optimized.
Traffic is Unstable & Unpredictable
Ad performance is influenced by millions of external factors:
Elections, seasonality, holidays, sports events, school cycles, competitor activity, even unrelated industry shifts.
Broad, high-budget campaigns feel these shifts much harder. Adjusting strategy without hurting overall performance becomes a nightmare.
For example, if AI copilots suddenly trend, ad costs spike. This could hurt ad performance for unrelated dev tools like analytics software.
If you’re running 1-2 broad campaigns with a massive budget, your entire marketing economy suffers—and without backup options, you won’t even know what caused the change.
Slow, iterative scaling = better data
Marketing attribution relies on cohort-based analysis (grouping users by signup date: daily, weekly, monthly).
By scaling gradually and testing strategies in both parallel and sequential modes, you get more diverse cohorts—which means better insights.
A strong data foundation = flexible scaling
Once you've tested enough strategies at a granular level, you gain flexibility in scaling effective ones, while optimizing other.
You make data-driven decisions instead of relying on luck, experience, or gut feeling.
Most agencies earn a commission on ad spend volume, so their focus is on fast, large-scale campaign execution. And this is high-level science too.
We take a different approach. We optimize the entire product economy, working across Pricing, UX, and in-depth marketing and product analytics to ensure clarity and strategic reasoning behind every move.
Have questions or need expert-level support for revenue growth?
Talk to me: MJei.me
