Brand Brain

Snapshot#

Penang Media is Eddie Cheng's UK performance marketing agency for DTC founders doing $500k to roughly $10M a year, primarily in the US and Europe, who rely on Meta and have hit a growth ceiling. Eddie is both the agency operator and co-owner of VIBAe, so the work comes from running a brand at scale, not managing other people's money from a distance. Penang replaces wasted spend and improvised media buying with a profit-first creative testing system that turns evidence into traction.

Audience#

Primary: founders and owner-operators of DTC ecommerce brands doing at least $500k in annual revenue, with an informal upper bound around $10M. Meta is their main acquisition channel. Their ad account has plateaued or become volatile, creative cannot keep up, and they are tired of agencies reporting activity instead of profit. They want direct access to someone who understands the economics and thinks like an owner.

No separate secondary customer segment is defined. Brands below $500k are redirected to free resources and the newsletter rather than pushed toward a call.

Pains and Gains#

Pains:

• Paying agencies to spend months "throwing darts" while "everyone under the sun" claims the sale.
• "Burning cash" while the numbers do not add up to a profitable business.
• An ad account that has plateaued, become volatile, or cannot be left alone for a weekend.
• Creative fatigue, slow production, and no repeatable way to decide what to test next.
• Vanity metrics and reports that hide what is happening to net profit.
• Fear that an agency will hand the account to the B-team or treat it like a sandbox.
• Performance creative that risks making the brand look "cheap" or "plastic."

Gains:

• Profitable, predictable growth on Meta.
• A clear, data-backed plan the founder can use with their own team.
• Account stability and the confidence to "start feeding the animal knowing that it is working."
• A creative pipeline where ideas, hooks, and formats are tested systematically and learning compounds.
• Direct access to an operator who shows the account, the decision, and the receipts.
• A partner who owns only the KPIs they can control and says plainly when the economics do not work.

One Word#

• Problem word: Waste
• Top 3 causes:
  1. Blind testing: creative launches without a clear hypothesis, reference, or reason.
  2. Profit-blind measurement: activity and attribution obscure whether the business is making money.
  3. Lost learning: results are not systematically captured and iterated, so every testing cycle starts again.
• Solution word: Traction

Positioning#

The problem lens is Waste. The solution concept is Traction: stop testing blindly, judge decisions against profit, and make each result useful to the next test.

Core messages:

1. Paid media amplifies a business. It cannot repair one.
2. ROAS can be green while the business is bleeding.
3. Creative volume without memory creates more waste, not more traction.

The chosen angle is Operator-in-Residence: "I build ad systems for your brand because I'm building them for mine." The operator shown in the content is the person the client works with, at their revenue stage, with the account open. VIBAe supplies the proof, but the position is broader than one brand. Eddie grew it as a cold-outreach client before buying shares roughly two years later. Results came first, equity second.

The creative system is the product underneath the position. The working diagnosis is that most Meta plateaus come from the creative process rather than the algorithm or targeting. Profit-first judgement decides what stays. Anti-agency language can win attention, but it is common in the market. The real difference is proof density: real screens, real decisions, named evidence, and losses shown as readily as wins.

Offers#

The ladder has one front door:

• Working session: for qualified DTC founders who want an expert diagnosis before considering an agency engagement. £197 for roughly 45 minutes, with a written 30-day plan afterwards. It is founder to founder, contains no agency pitch, and is "refunded before or after the call if it's not a right fit."
• Founder call: a free, application-gated upgrade after the working session. Framed as eight calls a month, with a $500k revenue floor and a 48-hour review buffer. It is not a parallel front-end offer.
• Agency retainer: Penang Media's managed engagement. Standard recorded terms are £4,500 per month, rolling monthly with no locked term, plus a performance component after month three. Confirm current commercial terms before quoting them.
• Newsletter and free resources: the catch-all for people who are not ready or do not meet the revenue floor.

Voice and Tone#

Calm, confident, direct. Founder-peer, never guru. Use plain words, short declarative sentences, concrete examples, physical metaphors, and the reason behind each instruction. Show evidence before making the claim. Admit uncertainty or a mistake plainly when it is real.

Do:

• Say "ad account" in full, "I'll" rather than "I'd," and "the working session funnel."
• Use "Basically" naturally as a connector.
• Keep headlines and on-screen text in sentence case.
• Use generous, low-pressure CTAs such as "give me a shout" and "Links below."
• Sound closest to Eddie's client register: structured, evidence-first, and honest about risk.

Don't:

• Use exclamation marks, em dashes, agency clichés, guru cadence, manufactured urgency, or outcome promises.
• Import the profanity and hype loops from Eddie's peer register into published content.
• Hide behind vanity metrics, vague attribution, or invented numbers.
• Use internal project shorthand in public copy.

Useful language includes: reverse engineer, North Star, guardrails, traction, iterate, hit rate, creative fatigue, low-hanging fruit, rinse and repeat, moving parts, pilot light, feeding the animal. Locked refund line: "refunded before or after the call if it's not a right fit."

Frameworks and Pillars#

The decision behind the number#

Point: "A result is only useful if it changes what you do next."

VIBAe's localized campaigns looked weak in isolation, but established English campaigns serving the same countries improved after the localized ads launched. That did not prove causation. It did justify further investigation and a move from translation toward culture-first concepts.

Framework: Local Result Versus Total Effect. Read the isolated result against the effect across the market, ad account, and business. The four decisions are stop or redesign, keep learning and iterate, check for displacement or hidden costs, and scale with guardrails.

Action: before switching off an apparent loser, compare its local result with the wider effect. Decide whether to stop, investigate, or scale.

The business behind the ad account#

Point: "You cannot make a good ad decision without understanding the business behind it."

The signature story is an IoT prospect whose product economics and available search demand could not support a scalable paid channel. Eddie declined the retainer and recommended proving distribution or recurring revenue first. Keep the company anonymous and confirm any figures before using the story publicly.

Framework: The Amplifier Test. Check four things before investing: existing traction, the signal being amplified, economic headroom, and the measurable trigger for more spend. If an answer depends on hope rather than evidence, validate it before trying to scale it.

Recurring points: the ad account does not decide whether you can scale; the economics do. A profitable-looking ad account can still lose the business money. ROAS is not the business.

Creative that compounds#

Point: "If your creative process has no memory, every month starts from zero."

The signature story is the creative workflow meeting where a task-based system could record output but could not show every combination of concepts, hooks, and formats. That led to the family-tree model: concepts form the trunks, hooks and formats form the branches, and results show what to grow, cut, or explore.

The Branded Response Flywheel has five stages: Deep AI and competitor research, Performance creative strategy, Iterative creative formula, Testing and intelligence extraction, and Profit-driven scaling. Performance intelligence feeds the next round of research and strategy.

The Creative Tree is the flywheel's memory. It maps Concept → Hook → Format and records each combination as validated, unsuccessful, or untested. The next brief pairs a validated element with one promising untested branch.

Other core methods:

• Creative diversity: problem-first, solution, lifestyle, founder video, UGC, static, POV, animation, and other formats prevent sameness and reach different buyers.
• Qualification math: work through AOV, margin, break-even CPA, traffic, and scaling reality before recommending spend. "The fire has to exist before you pump the gas."
• KPI scope: own only what the agency controls. Paid media alone should not be held to a blended metric it cannot fully influence.
• Brief gate: no ad brief without a reference and a reason. State the persuasion logic and awareness level before production.
• Receipts standard: show the decision, not only the number. A failed test and the pivot can be stronger proof than a winning screenshot.
• AI operating stack: Pulse analytics, an always-on agent, and creative AI tools support the work. AI amplifies craft. It does not replace founder footage or clear thinking.

Proof#

• VIBAe: Penang increased monthly revenue to $250k within 30 days. The brand became an 8-figure company in two years, with 90%+ of online sales driven by advertising. Eddie later bought shares, and the co-ownership is stated openly.
• Buckley Belts: grew from tens of thousands to multi-7-figure monthly revenue in four months, expanded across non-English-speaking Europe, and reached $1M+ monthly profitable ad spend in 2024. Eddie has never owned the brand.
• H.H Cosmetics: 6x daily new customer acquisition, 4x ad spend with improved reach and ROI, and stable accounts within three months. Co-founder Ko Hamer said Penang quadrupled the agreed performance numbers over four months.
• Bellissimo Hats NY: CPA down 30%, ad budget doubled, and an average ROAS of 3 sustained for more than 12 months.
• Portfolio proof approved for use: $5M+ per month in profitable spend managed, average partnership over 12 months, and tens of millions in ad spend data managed.

Every public number must come from Eddie's fact bank or an approved published case study. Historical results can be stated as evidence, never promised as an outcome.