Josh Gregory

Once Meta or Google budgets pass roughly £25–30K a month, they say, return on ad spend falls, acquisition costs climb, and results swing unpredictably.

• Typical reactions — changing bids, testing random creatives, duplicating campaigns — treat symptoms and leave the cause untouched.
• They name three places money leaks at scale: no funnel path for cold audiences, so efficiency drops as more cold traffic arrives; Google and Meta planned in isolation even though they overlap, producing wasted clicks and broken journeys; and no lifetime-value or offer strategy, so paying more to acquire a customer only works if the spend is recovered later.
• Their proposed fix is structural: weekly creative testing cycles, margin and customer-acquisition-cost checks before budgets rise, offers built for first-time buyers, a mapped path from awareness to purchase, and landing pages designed for cold clicks.

The clearest statement is the post that lists the £25–30K threshold and three leak points, then contrasts common reactions with a structural fix.

Read original · Original 2

In Josh Gregory's Performance Max example, four distortions are named: a large share of conversions come from branded searches and inflate results, repeat customers are counted as new revenue, display and YouTube clicks raise volume without always adding sales, and limited visibility hides where budget actually goes. They do not argue for dropping the campaign type, but for managing it — exclude branded terms so results reflect true performance, add negative keywords to cut waste, track new versus existing customers to see real impact, and run Performance Max as one part of a wider mix with Search and YouTube.

The Performance Max post sets out four distortions and four corresponding controls, explaining why a campaign that looks like the top performer may not be adding as much growth as its reported numbers suggest.

Read original

In one case, they describe a brand spending £80,000 a month across Google and Meta that closely tracks click-through rate, cost per click, and return on ad spend, while the product feed is messy, tracking is inaccurate, and the creative has not been refreshed in months — leaving, in their estimate, roughly half the budget wasted.

• Their point: scaling without structure is expensive guessing.
• The response is to clean up accounts, fix tracking, and rebuild the creative flow.
• A companion post makes the same point about budget increases: spend raised too quickly makes the platform broaden targeting, dilute the best audiences, and depress conversion rates.
• They say the problem is usually how the budget is spent, not how large it is.
• Their recommended controls are refined campaign structure, audience signals and exclusions to keep targeting sharp, and search terms, product feeds, and creative kept aligned with intent.

The £80K-a-month account is the concrete case: healthy headline metrics alongside a broken feed, inaccurate tracking, and stale creative, followed by unglamorous repair work.

Read original · Original 2

Small budgets, they say, prove profitability, and larger ones extend it.

• Agencies too often dismiss small accounts or use them to explain poor performance instead of building the right structure and following the data.
• This complements, rather than contradicts, the argument about scaling: the same structural requirements apply, but at a low budget they can be tested before real money is committed.

Gregory says a client started at £700 a month and reached an 8x return on ad spend within a few weeks, after which spend increased. Ask for supporting case data to verify the result.

Read original

In the scaling post, this appears as the missing lifetime-value or offer strategy: paying more to acquire a customer only makes sense if that spend comes back later.

• Offers need to work for first-time buyers, and margin and acquisition-cost checks belong before any budget increase.
• The same logic underlies the recommendation to judge campaigns on incremental rather than reported revenue, since repeat purchases counted as new growth flatter the numbers behind scaling decisions.

The two arguments meet in the Performance Max post, where repeat customers counted as new revenue are identified as a distortion to correct with new-versus-existing customer tracking.

Read original · Original 2

It’s the clearest statement of the argument running through Josh Gregory’s work: the usual reactions to a scaling problem — changing bids, testing random creatives, duplicating campaigns — miss the cause.

Josh Gregory describes what happens when Meta or Google budgets cross roughly £25–30K a month: return on ad spend falls, acquisition costs rise, and results swing unpredictably.

• The default response is to change bids, test creatives at random, or duplicate campaigns — none of which helps if the system behind the ads is not built for scale.
• Three specific places money leaks are named: without a funnel path for cold audiences, efficiency drops as more cold traffic arrives; planning Google and Meta in isolation creates wasted clicks and broken journeys, even though the channels overlap; and without a lifetime-value or offer strategy, higher acquisition costs only work if that spend is recovered later.
• The fix is structural rather than tactical: weekly creative testing cycles, margin and customer-acquisition-cost checks before budgets go up, offers designed for first-time buyers, a mapped progression from awareness to purchase, and landing pages built for cold clicks.

For brands planning a campaign

Use it to scope a paid-media engagement. It recommends sequencing the work: cover the funnel for cold traffic, plan Google and Meta together, then build a lifetime-value and offer model before approving any budget increase.

Josh Gregory argues that Performance Max can look strong on the surface — healthy return on ad spend, conversions arriving — while the underlying numbers tell a different story.

• Four distortions are listed: branded searches account for a large share of conversions and inflate results; repeat customers are counted as new revenue; display and YouTube clicks raise volume without always adding sales; and limited visibility obscures where budget is going.
• The conclusion is not to abandon the campaign type but to manage it: exclude branded terms so results reflect true performance, add negative keywords to cut wasted spend, track new versus existing customers to see real impact, and run Performance Max as one part of a wider mix alongside Search and YouTube.

For brands planning a campaign

A useful diagnostic for brands that treat Performance Max as their best channel. The four checks — branded-term exclusion, negatives, new-versus-existing customer split, and mix position — give an audit a concrete, testable starting list.

Josh Gregory describes a brand spending £80,000 a month across Google and Meta and closely tracking click-through rate, cost per click, and return on ad spend.

• But the product feed is messy, tracking is inaccurate, and the ad creative has not been refreshed in months, wasting roughly half the budget.
• The line that carries the piece: scaling without structure is expensive guessing.
• The response is unglamorous: clean up the accounts, fix the tracking, and rebuild the creative flow.

For brands planning a campaign

Put feed hygiene, conversion tracking, and a creative refresh cadence ahead of bid or budget changes. For a brand already spending heavily, start with an audit of those three foundations, not a new campaign structure.

Josh Gregory explains that when ad spend rises too quickly, the platform broadens targeting, the best audiences are diluted, and conversion rates fall.

• The problem is usually how the budget is spent, not its size.
• Three responses are given: refine campaign structure to control where budget goes, use audience signals and exclusions to keep targeting sharp, and keep search terms, product feeds, and creative aligned with intent.
• The post closes by offering a small number of free Google Ads audits to ecommerce brands seeking a second opinion on their account.

For brands planning a campaign

Relevant when a brand wants to increase its budget immediately.

• It supports a gradual scaling plan with audience signals and exclusions in place, plus checks that search terms, feeds, and creative still match the targeted intent.
• Ask for evidence that the proposed changes are grounded in account data; the free-audit offer at the end is a service promotion, not a research finding.

It’s the quickest way to see how Josh Gregory approaches testing before scaling.

Josh Gregory describes a client that started at £700 a month on ads and reached an 8x return on ad spend within a few weeks, at which point spend increased.

• Agencies too often dismiss small budgets as too small to matter or use them to explain poor performance, the argument goes, when the right structure, proper testing, and following the data can make modest spend perform well.
• Small budgets prove profitability; larger ones then extend it.

For brands planning a campaign

A test-first approach for brands with limited budgets: use a small spend to establish that the structure and creative can be profitable, then scale on that evidence rather than assumption.

Examples include cold-audience funnel gaps, disconnected Google and Meta campaigns, weak landing pages, and offers that do not work for first-time buyers. Diagnose the system before treating a budget increase as the solution.

The posts connect account structure to funnel design and unit economics, and recommend recurring checks such as margin and customer-acquisition-cost reviews alongside creative testing.

For brands planning a campaign

For a brand seeking profitable growth, assess the path from first impression to purchase across both platforms. Map cold and warm audiences, align channel roles and landing pages, and check whether the offer and customer economics support acquisition before increasing spend.

Examples: Source 1 · Source 2

In their Performance Max example, branded searches, repeat customers counted as new, and activity from Display or YouTube may inflate apparent results. They recommend evaluating those sources rather than assuming reported return on ad spend reflects incremental sales.

Gregory proposes exclusions and negative keywords, customer-status tracking, and comparing Performance Max with Search and YouTube as parts of a mix.

For brands planning a campaign

A brand evaluating Google Ads should review branded demand, new-versus-existing customer reporting, search-term quality, and the role of each campaign type. Use those checks to decide whether to expand Performance Max or shift budget within a broader channel plan.

Examples: Source 1 · Source 2

Josh Gregory recommends refining campaign structure and audience signals while keeping search terms, product feeds, and creative aligned with intent.

The advice combines account hygiene with structured creative testing and gradual, data-led budget decisions.

One post offers free Google Ads audits for a limited period. This is a call to action in that post, not evidence of a standing service or partnership format.

For brands planning a campaign

Before increasing spend, audit conversion tracking and product data, review audience controls and exclusions, and establish a regular creative refresh and testing cycle. A smaller-budget example describes a client moving from £700 per month to an 8x return before scaling.

Examples: Source 1 · Source 2 · Source 3

Martin Davies advises using an impression share report by day to work out whether a sales dip comes from lower market demand or competitor activity.

• Josh Gregory argues that reported campaign performance must be checked against what actually changed in the account.
• Both also discuss brand traffic inside Performance Max and how excluding it affects reported return on ad spend.

Martin Davies stays close to Google Ads diagnostics, including a third-party script for separating brand and non-brand performance. Josh Gregory makes the same scaling argument across Google and Meta, devoting more of each post to the structural and creative fixes that follow the diagnosis.

Read original · Original 2

Ujla Shakoor argues that a single winning ad becomes the account's biggest risk and buyers should always ask what is being tested next. Josh Gregory recommends weekly creative testing cycles and a creative flow that keeps pace with spend.

Ujla Shakoor writes from a hands-on Meta media-buying perspective and reflects on their own career and direction. Josh Gregory takes a scaling-and-structure approach spanning Google and Meta, with little personal reflection.

Read original · Original 2

Mark Tran's two-account example—new sales that were mostly returning customers, and the fix of separating new from returning customers and excluding past buyers from prospecting—makes the same case as Josh Gregory's Performance Max post about repeat customers being counted as new revenue. Both also track platform rule changes that affect ecommerce creative, such as Meta's AI labelling of generated images.

Mark Tran concentrates on Meta campaigns and platform policy shifts, including what brands should audit after AI labelling arrives. Josh Gregory covers a wider paid-media mix, with more emphasis on account structure, product feeds, and budget pacing.

Read original · Original 2

The two posts show a diagnostic method for exactly that situation: run an impression share report by day across the past 30 days and compare it with the period before sales dipped.

• Steady impression share points to lower market demand; a falling one points to competitor aggression.
• Change strategy accordingly.
• On Performance Max, the post offers a concrete way to answer the brand-inflation question a scaling brand will face: track brand versus non-brand clicks, conversions and share of brand revenue over time.
• Where brand takes a high share of PMax traffic, move it into search campaigns and add it as a negative in PMax, accepting a short-term return-on-ad-spend drop for overall business growth.
• Josh Gregory's posts name branded-search inflation inside Performance Max as a leak to fix; Martin Davies supplies the day-by-day test to establish whether it is happening in a given account, along with the demand-versus-competition check a brand scaling on Google Ads needs before blaming its own structure.

The posts are narrow and Google-only.

• Both cover Google Ads: impression-share diagnosis and the PMax brand decision.
• There is no Meta planning, cross-channel funnel map, offer or lifetime-value work, or creative testing cycle, so the Meta half of a scaling campaign stays uncovered.
• The PMax method depends on a script created by Smarter Ecommerce and shared with Martin Davies by Ben Fisher, so the brand relies on a third-party tool it does not control rather than a process Martin Davies built.
• The demand-versus-competition advice draws on one season of slower UK ecommerce sales, so it travels best to a similar market and pattern.
• Ask about scope, price and availability.

Examples: Source 1 · Source 2

The post describes the failure precisely: spend rising from $1K to $10K behind a single hook, angle and message, then cost per acquisition climbing and sales slowing as the creative tires.

• It also describes a habit of always asking what is being tested next, with tests across different hooks, angles, customer problems and formats.
• For a brand whose scaling plan keeps stalling at the creative layer, that pipeline discipline is more specific than anything in the Josh Gregory posts, which set a weekly creative testing cycle as a rule but stay at the level of the rule.
• Ujla Shakoor is the better call when the brand already has a Google structure it trusts and needs the Meta creative engine rebuilt to carry higher spend.

The tradeoff is scope and proof.

• Only one of the two posts gives campaign advice; the other is a personal reflection on five years of work, a deliberate pause and a clearer direction afterwards, with no transferable method.
• Neither post covers Google Ads, tracking fixes, product feeds, Performance Max, funnels, offers or lifetime value.
• A brand needing Google and Meta, funnel structure, creative testing and tracking fixes would need other help for most of that work.
• The Meta example is one account narrative at dollar budgets, with no client outcome data, account figures or commercial terms.
• Ask for those before treating the approach as a demonstrated result.

Examples: Source 1 · Source 2

The two-account post is the most detailed measurement case among the alternatives: two ecommerce accounts, a cricket store and a tennis store, each running one Advantage+ campaign with broad targeting, campaign budget optimisation and no exclusions for existing customers.

• They reported strong ROAS, low acquisition cost and strong ad performance, but closer inspection showed that a large share of “new” sales came from existing customers cycling back.
• The platform optimises for the cheapest conversion, often someone who was going to buy anyway.
• The prescribed fix addresses the tracking seam in a scaling campaign: separate new from returning customers, exclude past buyers from prospecting, push retention into email and SMS, and accept a lower reported ROAS while measuring new-customer volume.
• The post also names its outside sources — Cody Wittick's breakdown of a $71K version of the same problem and audits of Australian ad spend estimating up to 44% wasted or misattributed — giving a brand a way to test the claim against other work rather than take it on faith.
• The second post covers AI disclosure: Meta auto-labelling AI-generated ads where background, image generation or animation is used; auditing the creative pipeline; checking live ads for the “AI info” label; and briefing designers and media buyers.
• It also notes Google's own AI ad labels and the EU rules as the direction of travel.
• A brand scaling paid social with AI-assisted creative should read that alongside its scaling plan.

The posts focus mostly on Meta and platform policy.

• There is no Google campaign structure, impression-share diagnosis, feed or search-term work, funnel map, offer or lifetime-value modelling, so the brand gets less help on the Google half of a scaling campaign than from Josh Gregory or Martin Davies.
• The 44% waste figure is second-hand and describes the wider market, not the brand's own account; the cricket-and-tennis example is described without account data.
• The disclosure post states that tools that do not write metadata may slip through detection for now, while Meta runs its own detection and the gap is expected to close.

Examples: Source 1 · Source 2