Two cases. Two stories.

The client

A founder-led D2C luggage brand, in a category that does not reward newcomers easily.

A North American direct-to-consumer luggage brand. Founder-led, small operating team, premium positioning in a category dominated by long-established incumbents. Travel and luggage is genuinely competitive: you are not only up against other small D2C challengers but against Samsonite, Away, Rimowa, and the steady tide of unbranded inventory that floods Amazon at lower price points. Margins can be reasonable when you can rank, but ranking requires either money or patience.

When the founder reached out in late 2024 the brand had been live for under two years. Revenue was growing, but almost entirely on paid Meta and direct traffic. The organic channel was effectively dormant. Search Console showed impressions, but on the wrong queries: informational, brand-adjacent, rarely commercial.

The founder had also been through two SEO consultants. Both promised page one in ninety days. Both delivered keyword reports, content briefs and very little measurable change. By November 2024 the founder had specifically lost interest in another deck. They wanted someone who would do the work and tell them when, realistically, it would land.

The situation

A Shopify store with no SEO infrastructure, no content, and no measurable organic visibility.

The pre-engagement audit, run in the first week of November 2024, found a picture that was both bad and recoverable. The storefront looked fine to a customer: clean photography, decent UX, working checkout. Below the surface the foundations were not broken so much as absent. No structured data anywhere. No XML sitemap submitted. A robots.txt using a Shopify default that was unintentionally blocking several useful URL patterns. Render-blocking JavaScript pushing Largest Contentful Paint above four seconds on mobile.

On-page work was almost entirely missing. Product pages had three or four line descriptions, generic titles built off Shopify defaults, no internal linking strategy, no schema, and no image alt text beyond file names. Category pages were not really category pages, just collection pages with no introductory copy and no reason for Google to treat them as landing experiences.

The largest gap was content. The brand had no blog at all. In a category where buyers research extensively before purchase, sizing guides, material comparisons, durability, airline regulations, packing method, the absence of editorial content meant there was nothing to crawl, no internal linking surface to build authority on, and no editorial signal for the AI engines that were beginning to matter.

The backlink profile was thin but clean. A handful of directory mentions and two or three small lifestyle references that had appeared organically. Nothing toxic, nothing penalised, just a domain no real publication had reason to link to yet. Starting from zero is slow but predictable. Repairing a damaged profile is much harder.

The diagnosis

Not a one-thing problem. A foundation-has-not-been-built problem.

Some SEO engagements are diagnostic puzzles, where the site ranks well where it should not and badly where it should, and the work is finding what is distorting the signal. This was not that. The brand was not being outcompeted by smarter SEO. It was missing the foundational work any ecommerce brand at its stage should already have had.

We laid it out in a one-page document at the end of week one. Technical foundations were poor but fixable, and we expected most of those to ship inside the first month. On-page was almost entirely missing and was the highest immediate-leverage area, because the work was straightforward and would compound quickly. Content was the largest gap and the slowest to fix, with no realistic presence before month three or four. Authority was low but not negative, and would be the slowest of the four streams.

We held the start of formal outreach until month four. That decision is one we would revisit, and we discuss it in the reflection at the end.

The approach

Four parallel work streams. Real timelines. Honest expectations.

We told the founder up front that meaningful ranking movement on commercial keywords would not appear before month four, and that anyone promising faster was either lying or about to put the domain at risk. What they would see by the end of month one was technical fixes shipping, first on-page rewrites live, schema deployed and a content strategy locked. By month three, movement on long-tail commercial queries. By month six, movement on competitive primary keywords. Revenue would follow rankings, never lead them.

The founder agreed to a twelve month framing. Not a contract, since we do not lock clients in, but an aligned expectation that success would be measured at twelve months rather than ninety days.

Cadence was bi-weekly strategy calls for two months, then monthly, with a live dashboard throughout and no written monthly reports. The founder was specifically relieved by that. Two previous agencies had sent forty page documents that took longer to read than the month's actual SEO work had taken to perform.

The execution

Month by month, what shipped and when.

1. Month 1 · December 2024

   The technical audit surfaced roughly forty discrete issues. We prioritised aggressively: anything affecting indexability or Core Web Vitals went into week-one work. Schema for organization, product and breadcrumb types shipped by mid-December 2024. Render-blocking JavaScript was resolved through asset reorganisation and deferred loading, and LCP dropped from 4.2 seconds to 1.8 across templated pages. Robots.txt was rewritten and the sitemap rebuilt and submitted. By the last week of December 2024 the top ten commercial product pages had new titles, meta descriptions, heading hierarchy fixes and a first round of internal linking.

2. Months 2 to 3 · January to February 2025

   Content strategy locked in mid-January after a two week buyer-journey mapping exercise, splitting the keyword universe into three intent stages. Research-stage queries were targeted with long-form pillar pages, comparison-stage queries with structured comparison content, and purchase-stage queries through product page optimization and category rebuilds. The first three pillar pieces went live in February: 2,500 to 3,500 word documents with internal linking, FAQ schema, comparison tables and Q&A structures built for both classic search and AI extraction. Cadence was set at two pillar pieces plus four supporting articles a month, all written internally.

3. Months 4 to 6 · March to May 2025

   First observable ranking movement appeared in mid-March. Long-tail commercial queries started surfacing the brand in positions 30 to 50, where Google routes high-relevance but low-authority pages while it gathers signal. By mid-April several had moved to page three, and by the end of May three were on page two with two trending toward page one. Backlink outreach began in earnest in March across three strategies: digital PR around the comparison guides, expert quote pitching through journalist briefs, and strategic guest contributions on adjacent travel publications where the editorial fit was honest.

4. Months 7 to 9 · June to August 2025

   The compounding started. Pages on page three in May were on page one by July for medium-competitive queries. Organic traffic doubled between April and July, a meaningful absolute increase rather than a large percentage off a tiny base. The first observable spike in organic-attributed revenue came in July, and the founder renewed without discussion in early August. The library passed thirty long-form pieces, and internal linking was restructured across two sprints in June and August. Three significant backlinks landed: a feature in a mid-tier travel publication, a comparison reference in a higher-authority lifestyle title, and an expert quote in a national consumer publication.

5. Months 10 to 14 · September 2025 to January 2026

   Scale and defend. Content shifted toward seasonal queries: back-to-school travel in September, holiday gifting in October and November, January travel-resolution content in December. Several foundational pillar pieces published in February were now ranking page one for their primary keywords, and each was updated in November 2025 with refreshed information, new internal links and additional FAQ content. By the January 2026 measurement window the brand was on page one for the keywords that mattered most to commercial revenue.

The results, in context

The headline numbers, and what they actually mean.

The temptation in agency case studies is to lead with the biggest figures and stop talking. That is how case studies stop being credible, so here is the detail behind each one.

The +217% revenue figure is organic-channel-attributed revenue growth, comparing the same months in the year before engagement to the same months during it. It excludes paid acquisition, direct traffic and email, which were running independently throughout. Total brand revenue grew at a different, smaller rate.

The +700% organic traffic figure compares the engagement's start month, November 2024, to its end measurement month, January 2026. That is a fourteen month comparison, not year over year, and it benefits from a low starting baseline. We say so explicitly because it matters: a brand starting from a higher organic baseline would not see 700% in fourteen months even if the same work were performed at the same quality.

The +100% CTR figure is average click-through rate across commercial queries in Search Console. It came from two compounding sources: better positions, and better titles and meta descriptions, which produced measurable gains even where ranking positions held steady.

More important than any single metric, the brand ended with a structural foundation that compounds for years if maintained. Page one rankings, once earned, are easier to defend than to win. The founder did not need to renew to sustain the results. The follow-on scope was AEO and GEO, not SEO maintenance.

Thank you for saving me from the worst marketing teams that never knew anything. I am glad I met you.

What we'd do differently

Two things, in retrospect.

The client

A premium bedding brand with high order value, a long consideration cycle, and a multi-year relationship with our team.

A North American direct-to-consumer brand selling luxury linen bedding. Premium positioning, high price points and high gross margin per unit, but a long consideration cycle. Customers genuinely shop around, comparing materials, weave types, country of origin and sustainability claims. Repeat purchase rates are healthy because the product is durable, but new-customer acquisition is the engine of growth.

The relationship with ScaleArk started in 2023, initially scoped around SEO and brand work, with paid advertising managed by a separate longstanding agency. The two ran in parallel, generally without conflict, for nearly two years. In November 2025 the founder asked us to evaluate whether taking Google Ads over would be a useful change. We took over operationally on 1 December 2025.

One thing worth naming up front, because it shapes the case: we already had context. We knew the catalogue, the customer profile, the seasonal patterns and the brand voice, and we had been inside Search Console and Analytics for years. The four month timeline was possible partly because we did not need to spend the first month learning the business. A new client would see structural CPA improvements on a similar trajectory, but realistically over five to six months.

The situation

An account producing revenue, but quietly drifting in the wrong direction.

Cost per acquisition in April 2025 stood at $132.88. For a brand with high order value and strong margin per unit that was not fatal, and the brand was profitable on a contribution-margin basis. But CPA had been creeping upward for twelve months, from roughly $90 in early 2024, past $100, past $115, and now north of $130 with no sign of reversing.

The previous agency was running five active campaigns at once: branded search, two non-branded prospecting campaigns with overlapping audiences, a Performance Max campaign and a remarketing campaign. Optimization scores hovered between 60 and 75 percent, meaning the platform was actively flagging recommendations nobody was actioning. Conversion tracking had observable drift, with server-side events missing for iOS-affected traffic and the GA4 setup not matching the Google Ads setup.

The founder's frustration was not that campaigns were failing, because they were producing revenue. It was that CPA had drifted upward for a year, the incumbent agency could not articulate why, and the monthly reports kept getting longer while the meaningful data inside them kept getting sparser. That pattern is familiar. Monthly optimization is, in practice, a euphemism for somebody glancing at the account for ten minutes.

The diagnosis

Structural waste, not tactical inefficiency. The account did not need more spend, it needed less clutter.

The first week was diagnostic. We pulled six months of campaign-level performance data, ran auction insights across every active campaign, and audited the conversion tracking end to end. The findings were structural.

The five campaigns were cannibalising each other. The two prospecting campaigns had over 60% audience overlap, competing in the same auctions for the same impressions with the same brand paying both sides. Performance Max was buying clicks on branded queries the dedicated branded campaign should have owned at lower cost. The remarketing campaign was being credited with conversions branded search had originated, because last-click assigns the conversion to whichever campaign closed the click rather than the one that generated the demand.

Optimization scores below 80% meant a backlog of unactioned recommendations: bid strategy mismatches, with campaigns on Maximize Clicks when the conversion data clearly supported Target CPA, audience signal underuse, ad copy improvements and negative keyword expansion. None of these are exotic. They are the basic maintenance any account this size should get weekly.

Conversion tracking was leaking events. Server-side tracking was not deployed for iOS-affected traffic, which since the iOS 14.5 privacy changes is not a small problem. The reporting the founder had been seeing for a year was mechanically distorted. The diagnosis was easy to communicate and hard to deliver in agency terms: we do not need to spend more money, we need to fix what is already running.

The approach

Cut the campaigns down, fix tracking first, no new channels for at least a quarter.

We told the founder in the first call after the diagnostic week that we were not going to add new campaigns. We were going to cut five down to two or three, get every active campaign to a 100% optimization score before touching creative or scaling spend, and fix tracking inside fourteen days. CPA would move within four weeks and revenue would follow within ninety days if the fundamentals stabilised.

We were equally specific about what we would not do. No new channels, with TikTok, Reddit and Pinterest off the table for at least a quarter regardless of how they were performing for other brands. No test budgets: every dollar went into existing campaigns until those were structurally clean. No new creative until the existing creative had been tested fairly against a recovered tracking baseline. The founder agreed quickly, because the plan matched their instinct: less, not more.

The execution

Four months, each phase doing what the previous phase had earned the right to do.

1. Days 1 to 14 · Early December 2025

   Conversion tracking audit completed. Server-side tracking deployed through Tag Manager for iOS-affected events. GA4 and Google Ads conversion mappings reconciled, so the two systems finally reported the same events with the same definitions, which they had not done for at least a year. Two of the five campaigns paused immediately: the redundant prospecting campaign and the remarketing campaign that was double-counting branded conversions. Performance Max was restructured to exclude branded search queries.

2. Days 15 to 30 · Late December 2025

   The three remaining campaigns were each pushed to a 100% optimization score, which meant actioning a backlog that had accumulated for months. Bid strategy switches from Maximize Clicks to Target CPA, and Manual CPC to Maximize Conversions. Ad copy variants added, audience signals cleaned, negative keywords expanded using actual search-term report data rather than theoretical lists. Early-month CPA readings were still carrying $130-plus from the old structure, but mid-December readings were trending toward $100, with no additional spend. The improvement was entirely structural.

3. Days 30 to 60 · January 2026

   By mid-January the new structure was producing CPA in the $80 to $90 range, a meaningful improvement while the platform was still in its post-restructure learning phase. We deliberately held creative steady to let the bid strategies stabilise against a consistent baseline. Late January brought the first creative refresh: three new variants per campaign, differentiating top-of-funnel story-led creative from bottom-of-funnel offer-led creative.

4. Days 60 to 120 · February to April 2026

   February stabilised at $75 to $80. March moved to $70. April hit $65.56, the headline number. The architecture stayed at three active campaigns throughout, all at 100% optimization scores, all with stable bi-weekly creative refresh cadences. April 2026 revenue was three times April 2025, and Q1 2026 revenue was four times Q1 2025. The improvement was not from spending more, since total spend was roughly flat to slightly down. It came from spending the same money against a structure that had stopped bleeding to overlap and waste.

The results, in context

The same caveats that applied to the SEO case apply here.

The CPA reduction is across the entire account, not any single campaign. Branded search CPA was already low, because branded search is structurally efficient when implemented well, so there was little room to improve it. The dramatic improvement was in prospecting, which dropped from over $200 to roughly $80, and that single shift pulled the blended account-level figure down sharply.

If you are reading this to judge whether your own prospecting CPA could drop similarly, the honest answer depends on whether your account carries the same structural waste. Accounts managed well from the start do not have this kind of recoverable inefficiency sitting inside them.

The 3x April and 4x Q1 revenue figures are year over year, comparing a structurally improved account against a baseline year. A meaningful share is attributable to better account structure, and another share to brand-side improvements the founder shipped in parallel: better landing pages, a refreshed catalogue, seasonal positioning handled internally. We do not claim the entire revenue lift. We claim the CPA reduction unambiguously, because that came directly from structural changes we made inside the account.

Most importantly, the brand had a multi-year relationship with us before we took over paid. There was no relationship-building overhead in the first month. This is not a magic four month playbook. It is a sound diagnostic playbook layered on top of zero relationship friction.

Thank you Majeed, for everything you do and everything that you have done.

What we'd do differently

One thing, in retrospect.