Loading performance story
Loading page...
This professional cosmetic brand was a sophisticated $780K/month business — with an AI-driven agency, advanced tooling, and years of institutional knowledge. They weren’t broken. They were stuck. This is the story of what a human expert sees that an algorithm cannot.
The client is a USA-based beauty brand specializing in professional-grade lash extension products — adhesives, shampoos, primers, and care tools trusted by lash artists and beauty professionals across the globe. The brand competes in one of Amazon’s most saturated and aggressively advertised beauty subcategories, where copycat products and aggressive bidding make sustained visibility both expensive and fragile.
By the time Clickstera entered the picture in March 2024, the brand was already mature by any reasonable measure. ~$780K–$800K in average monthly revenue. Active across Amazon US, Canada, UK, and Walmart. A well-established catalog. A multi-year relationship with one of the world’s leading Amazon advertising platforms. By all appearances, a brand that had figured it out.
The CEO wasn’t unhappy. They weren’t shopping for an agency. They were running a strong business and had full confidence in the tools and teams managing it. That’s what made what we found in the audit so significant — and so surprising to them.
The brand had been managed by a sophisticated AI-driven optimization platform for over three years. This wasn’t a weak or outdated setup — it was one of the industry’s leading Amazon advertising tools, built to automate bidding, keyword harvesting, and budget allocation at scale. For many brands, it delivers excellent results.
For this account, performance had plateaued. Average monthly revenue had been stuck in the $780K–$800K range for the better part of three years, growing at roughly 2% year-over-year — a rate that, adjusted for Amazon’s natural category growth, essentially meant standing still. The ceiling wasn’t visible from inside the account. The reports looked reasonable. The ROAS was acceptable. But the brand’s true revenue potential was sitting untapped, invisible to the algorithm managing it.
“A great tool optimizes within the structure it’s given. It cannot redesign the structure itself. That requires a human who understands not just what the data says — but what the data isn’t saying.”
Our founder, Aadarsh Shaw, had spent 2021–2023 at Teikametrics — contributing to the very category of AI-powered Amazon advertising tools that this brand was using. He understands precisely what those systems are optimized for, and crucially, where they stop. When we ran our pre-engagement audit on the account, the gap he spotted wasn’t a flaw in the tool. It was a structural problem that no automated system was positioned to solve.
10 keywords were responsible for approximately 75% of all advertising-driven sales across the brand’s hero ASINs — and every single one of those keywords was running in multiple campaigns simultaneously, competing against itself for budget, diluting bid signals, and preventing any single campaign from fully capitalizing on demand.
This wasn’t a bidding problem. The bids were fine. It wasn’t a budget problem — there was plenty of spend. It was a structural architecture problem: the account was built in a way that guaranteed the 10 most important keywords could never be managed with the precision they deserved, because they were buried inside broader campaigns alongside dozens of lower-value terms, sharing budgets and diluting signals.
Fixing it required tearing down that architecture and rebuilding it from scratch — at a brand doing nearly $800K/month, with $140K in monthly ad spend in flight. That’s not a task for an algorithm. That requires judgment, experience, and the willingness to take calculated risk on behalf of the client.
The CEO received a Clickstera outreach email in early 2024. They responded — not because they were dissatisfied, but out of curiosity. The email promised an audit perspective, and as a data-driven leader, they were open to outside input.
What followed was a month-long process. Initial meeting. Deep account audit. Strategy presentation. And then — because the CEO is exactly the kind of operator who does their homework — a formal background check and an NDA before any access was granted. They weren’t handing a $140K/month ad account to anyone without verification.
The audit finding that changed the conversation was the 10-keyword insight. The CEO had never seen the account through that lens before. The previous management system reported performance by campaign and by product — but it hadn’t surfaced the concentration risk hiding underneath: the vast majority of revenue dependent on a handful of terms, each of which was structurally limited from reaching its full potential.
“Even though they were doing $750–$800K, there was so much more to capture.” That realization — not frustration, not failure, but untapped opportunity — was what moved them to take the chance.
“They didn’t sign because we told them something was broken. They signed because we showed them a ceiling they didn’t know existed — and a concrete plan for pushing through it.”
The account rebuild followed a clear sequence: isolate the revenue drivers first, restructure the architecture around them, add the DSP layer with surgical precision, then optimize everything manually — daily — at a level of granularity no automation handles well.
The first action was to identify the 10 keywords responsible for ~75% of advertising-driven sales — terms like “lash shampoo,” “eyelash extension cleanser,” “lash glue,” and “eyelash extension adhesive” — and extract every single one from the campaigns they were buried in. Each was negated in the existing campaign structure and relaunched in its own dedicated single-keyword campaign with full budget and bid control.
For the first time, each of these keywords had its own campaign, its own budget that couldn’t be consumed by lower-priority terms, and its own optimization track. The signal was clean. The control was total. The results from this change alone began appearing within weeks.
Beyond the top 10 keywords, the remaining campaign structure was rebuilt from scratch around search intent — not product, not match type, but what the shopper was thinking when they typed that query. Three buckets, each with its own bidding logic and budget ceiling:
Brand intent campaigns captured shoppers already searching for the brand by name — high conversion, lower bids, brand defense priority. Generic intent campaigns covered category searches (“professional lash adhesive,” “lash extension shampoo”) — highest volume, most competitive, requiring the most aggressive bid management. Competitor conquesting campaigns targeted shoppers searching for competing brands — lower conversion but high strategic value for new-to-brand customer acquisition.
Each bucket was further segmented by match type — Broad, Phrase, and Exact running in separate campaigns — ensuring clean data, zero overlap, and independent scaling paths for each intent layer.
Amazon DSP was already part of the account but being used broadly. We rebuilt the DSP architecture around a single principle: every dollar of DSP spend should be targeting someone already close to buying, not building brand awareness for unknown audiences.
The rebuild focused DSP exclusively on lower-funnel audiences — product detail page viewers, cart abandoners, and past purchasers within short lookback windows. Each product group received its own DSP campaign structure, segmented by audience type and placement (Mobile App, Web, Desktop) to give clean performance data at every level.
* Screenshot from live DSP campaign structure (client account) showing product-level audience separation across placement types and intent categories. This level of segmentation enables independent performance measurement and budget control at every node.
At ~$140K/month in ad spend, the difference between good and great optimization is not a better algorithm. It’s a human being reviewing the account every day. Our approach at this spend level involves daily search term review — identifying new converting terms to promote, blocking emerging waste before it compounds, and catching intraday budget exhaustion patterns that automated systems typically address weekly at best.
Budget reallocation was done based on demand signals, not schedules. If a product’s conversion rate spiked on a particular day or time window, we moved budget toward it in real time. If a keyword showed early signs of declining efficiency, we adjusted bids before the data fully confirmed it. This is the work that separates human-managed accounts from tool-managed accounts at scale — the thousand small decisions that never make it into a weekly report but compound into meaningful performance differences over months.
The most important number in this case study isn’t the revenue total. It’s the comparison: three years of approximately 2% year-over-year growth, followed by a ~17% jump in the first year of Clickstera management. The brand didn’t change. The products didn’t change. The category didn’t change. The strategy did.
By the time Clickstera onboarded in March 2024, average monthly revenue was running at approximately $780K. Within months of the restructure, that average had climbed to ~$916K — an incremental lift of roughly $136K per month that has held and continued building into 2025 and 2026.
Same brand. Same products. Same ad budget. Different architecture, different strategy, different result.
The CEO’s onboarding conditions were clear: background check, NDA, discounted 3-month trial. They were being careful — appropriately so, for a brand at this scale. Clickstera accepted every condition without hesitation, because the audit findings gave us full confidence in what the results would show.
The trial produced exactly what the audit had predicted. Revenue climbed. The architecture held. The weekly reporting gave the CEO a level of strategic transparency they hadn’t experienced before — not campaign metrics, but account-level thinking, competitive context, and forward-looking decisions explained in plain language.
In November 2025 — twenty months after the initial 3-month trial — The CEO signed an annual contract at full rate. That’s the only testimonial that matters: a sophisticated leader, who started skeptical, chose to make the relationship permanent.
The 2026 target is $1M/month in average monthly revenue. With the foundation now in place, the campaign architecture optimized, and all four marketplaces operating on a unified strategy, that number is a milestone — not a ceiling.
Client photo (optional)
[Written quote from the client CEO to be added — they have verbally expressed strong satisfaction with the engagement, citing strategic transparency and the depth of account-level thinking as key differentiators from previous management.]
If your revenue has plateaued despite strong products and consistent ad spend, the problem is almost certainly structural — not competitive. Let’s find it.
Book a Free Account Audit