Last updated: 4 May 2026

What Is Real Buyer Growth? Definition, Importance, and the Keigen Methodology

By Helen Chen, Managing Director, Keigen Technologies UK Limited

TL;DR

Real buyer growth is the portion of reported ecommerce growth that survives evidence review — the new-customer wins, conversions, and revenue lifts that hold up after bots, promotion abuse, and acquisition pollution are filtered out.

Most growth dashboards measure activity. Real buyer growth measures what deserved to count. The distinction matters because four forces are now widening the gap between reported and real numbers: bot traffic at the campaign layer, promotion abuse at the offer layer, AI-powered buying agents arriving on retail sites at industrial scale, and CFO scrutiny on marketing spend.

This article defines real buyer growth in working terms, separates it from adjacent concepts (incrementality, attribution, ad fraud detection, AI agent identity), names the five contamination patterns we see most often in mid-market ecommerce, and shows how a Real Buyer Growth Evidence Review verifies which portion of last campaign deserves to be scaled.

What real buyer growth means

Real buyer growth is a category of ecommerce performance measurement that asks one question: of the growth your dashboard reports, how much was driven by genuine commercial demand from genuine buyers?

It is the layer beneath reported growth. Where reported growth counts everything the analytics stack records — clicks, sessions, new accounts, redeemed codes, attributed orders — real buyer growth counts only the portion that survives evidence review against bot signatures, promotion-abuse patterns, attribution-window artefacts, and identity-deduplication checks. The output is a verification rate and a narrow set of reason factors that explain the gap between the two numbers.

For ecommerce operators, the cost of conflating reported with real growth is rarely a single missed sale. It is misallocated next-campaign budget, distorted CAC and ROAS calculations, retargeting audiences populated by traffic that was never going to convert, and finance conversations in which marketing claims do not survive a CFO review. Real buyer growth is the language layer that lets a merchant say to finance, “this portion of last quarter held up; this portion did not.”

Why this matters now

Four forcing functions are widening the reported-vs-real gap in 2026 in ways that did not apply in 2023.

Bot economics have inverted at the campaign layer

TrafficGuard reports that ecommerce advertisers can now lose 15–30% of paid media spend to invalid traffic before a click ever reaches checkout, with damage compounding through polluted attribution and weakened retargeting audiences. The MRC Global Fraud Report finds 43% of merchants flag coupon, discount, and refund abuse as their most common fraud category. Forrester research cited across the promotion-integrity field reports that merchants exposed to promotional abuse lose approximately 31% of annual marketing spend.

Promotion abuse is now industrialised

Kount's merchant survey found that 42% of businesses allow customers to abuse promotions through repeated-account creation, referral-loop exploitation, and code-stacking patterns that look like organic acquisition on dashboards. Liminal's 2025 Link Index reports only 20% of merchants enforce promotion-code restrictions automatically — meaning four in five rely on after-the-fact reconciliation that arrives weeks after the budget is committed.

AI buying agents are arriving on retail surfaces at scale

McKinsey's October 2025 research projects the US B2C agentic-commerce opportunity at $900 billion–$1 trillion by 2030, with global projections of $3–5 trillion. Adobe data cited in BCG analysis recorded a 4,700% year-over-year increase in generative-AI-driven traffic to US retail in July 2025. Shopify's Universal Commerce Protocol announcement in January 2026, co-developed with Google, signalled that native shopping inside Google AI Mode and the Gemini app is now production infrastructure for tens of thousands of merchants.

CFO scrutiny on marketing spend has hardened

The LexisNexis True Cost of Fraud study reports merchants now lose $4.61 in downstream costs for every $1 of direct fraud, a 32% increase since 2022. Boards and finance functions are asking marketing teams to defend campaign claims with evidence, not narrative.

The gap between reported and real growth is no longer a quality-of-data inconvenience. It is a budget-defence problem.

How real buyer growth gets verified

Verifying real buyer growth means running a structured review of one campaign window — paid traffic, landing pages, attribution events, promotion redemptions, order data, and identity signals — against a defined set of contamination tests. The output is a written verdict, a verification rate, and a small set of reason factors.

The RealBuyerGrowth method runs four checks in sequence.

Claim. Reported new customers in a campaign window are filtered against repeat-account, repeat-device, and identity-overlap signatures before being counted as clean acquisition.

Evidence. In RealBuyerGrowth specimen reviews, between 18% and 35% of accounts a Shopify-stack merchant counts as “new” share device fingerprints, payment instruments, or address graph nodes with prior accounts under different email handles.

Limitation. Identity overlap alone does not prove abuse — household sharing and legitimate gift purchases produce overlap signal too. The reason factors flag overlap; the verdict requires corroborating signal (promo-code reuse pattern, referral-graph topology, or short-window repeat-redemption).

Source. RealBuyerGrowth Evidence Review specimen file, aggregated across UK Shopify-stack engagements 2026 Q1.

Date. As of 2026-05.

The second check tests paid-traffic quality at the campaign layer — separating sessions consistent with human commercial intent from sessions that match bot, scraper, or invalid-traffic signatures before the click translates into attribution credit.

Claim. Paid-traffic sessions whose engagement signature falls below a defined commercial-intent threshold should not contribute to verified-acquisition counts even when the session technically converts.

Evidence. TrafficGuard data on UK ecommerce ad spend reports 15–30% of paid-media budget can be lost to invalid traffic before any checkout event, with the damage compounding through polluted attribution and weakened retargeting audiences. RealBuyerGrowth specimen reviews on Shopify-stack stores typically locate 8–22% of paid-traffic conversions in this category.

Limitation. A session falling below the threshold is not automatically illegitimate — slow-loading pages, mobile-first browsers in low-bandwidth conditions, and accessibility-tool users can produce thin engagement signatures. The reason factors flag the pattern; the verdict requires corroborating signal from referrer chain, IP reputation, or session-time distribution.

Source. TrafficGuard 2025 published merchant data, cross-referenced against RealBuyerGrowth specimen file 2026 Q1.

Date. As of 2026-05.

The third check examines promotion redemption patterns for the five named contamination types (below). The fourth check reconciles attribution-window artefacts: orders attributed to a campaign whose conversion path does not survive a 14-day stability test.

The output document — the Real Buyer Growth Evidence Review — runs to 12–18 pages. It contains an executive verdict, a verification rate (the percentage of reported growth that survived review), the top three reason factors driving the gap, and a recommended next-campaign separation: which portion of reported growth is ready to scale, and which portion should be quarantined from paid budget until a follow-on review.

What does and does not count

The category boundary is the most-asked question we receive. Real buyer growth is a measurement category, not a tool category. It does not replace your ad platforms, attribution stack, fraud screen, or analytics warehouse. It produces an evidence layer that sits above them.

Counts as real buyer growth	Does not count
Full-price orders with stable conversion path over 14-day window	Discount-driven first orders with 0 follow-on engagement
New accounts that survive identity-overlap and device-fingerprint checks	“New customer” entries that share device, payment, or address graph with a prior account
Paid-channel traffic with engagement consistent with human commercial intent	Click-through traffic that fails session-quality tests at the landing-page layer
Referral-driven acquisition where referrer and referred have independent identity graphs	Referral chains where multiple referred accounts trace to a single device or payment instrument
Promotion redemptions consistent with merchant-defined eligibility windows	Redemptions enabled by code stacking, eligibility-window arbitrage, or restored-account abuse

The principle: a buyer counts when the dashboard claim about that buyer would survive a CFO review against four corroborating evidence layers. If the claim fails any one of identity-stability, traffic-quality, promotion-eligibility, or attribution-stability, the buyer enters the gap, not the verified column.

The five contamination patterns

Most reported-vs-real gaps trace to five named patterns. These are the patterns RealBuyerGrowth Evidence Reviews are scoped to detect on a Shopify-first stack.

New-customer recycling

A merchant offers a first-order discount to attract acquisition. Existing customers create new accounts — different email, same device or payment instrument or address — to claim the offer again. The campaign reports strong new-customer acquisition; the underlying reality is a discount given to existing customers under a new label. Kount's merchant survey finds 42% of businesses allow this pattern to operate without active control.

Referral farming

A referral programme rewards introductions. Operators create referrer/referred pairs from a single identity graph, harvesting referral bonuses on synthetic introductions. Healthy referral data shows referred accounts diverging quickly from the referrer's identity neighbourhood. Farmed referral data shows the opposite — tight clustering, shared addresses, sequential creation timestamps.

Entitlement leakage

A loyalty tier, member discount, or partner-rate entitlement is meant for a defined buyer group. Codes leak into channels they were not scoped for — affiliate sites, deal forums, employee-only codes circulating publicly. The merchant sees redemption volume against a code; the underlying buyer set is no longer the one the entitlement was designed for.

Code stacking abuse

Promotion codes meant to be exclusive are stacked through ordering tricks, browser-state manipulation, or sequential-redemption windows. The order looks legitimate; the margin economics do not survive review. Liminal's 2025 data showing only 20% of merchants enforce promotion restrictions automatically maps directly to this pattern.

Promo-code leakage

Codes intended for warm-list audiences appear on coupon aggregators, browser extensions, and deal sites within hours. The merchant attributes redemptions to the original campaign; the actual traffic source is the aggregator surface. The reported attribution and the real attribution diverge by the share of redemptions captured at the aggregator layer.

A Real Buyer Growth Evidence Review names which of these patterns is most likely affecting a campaign window, with worked evidence for each named pattern.

What this is not

Real buyer growth is not anti-bot security. The objective is commercial-decision evidence, not network-layer protection. Anti-bot tools sit upstream of where RealBuyerGrowth operates.

It is not attribution modelling. RealBuyerGrowth does not replace MMM, MTA, or last-click attribution. It produces an evidence layer that tells finance and marketing whether the attribution stack's outputs deserve to be acted upon.

It is not chargeback management. Disputes, returns, and refund-fraud sit in the post-purchase risk stack and are addressed by chargeback platforms.

It is not AI agent identity verification. As AI buying agents arrive on retail surfaces at scale, the question of which agent is transacting on whose behalf becomes a separate problem with separate infrastructure. Initial deployments of RealBuyerGrowth produce campaign-window evidence on human-driven acquisition; advanced features for AI-agent-segmented traffic verification are scoped for higher-assurance deployments inside the Keigen portfolio.

The verification rate and the escalation principle

The verification rate is the operational metric the Real Buyer Growth Evidence Review produces. It expresses, as a percentage, how much of reported growth survived review.

A specimen output: a campaign reports 1,842 new customers; review finds 1,311 survive verification; the verification rate is 71.2%; the gap of 531 is decomposed into reason factors (top factor: repeated device reuse) with an at-risk incentive spend figure of £4,980. The merchant now has the language to tell finance: 71% of last campaign held up; 29% should be separated from next-campaign budget pending follow-on review.

Verification rates above 85% indicate a campaign whose reported-vs-real gap is narrow enough to scale with confidence. Rates between 65% and 85% indicate a gap large enough to require campaign-level separation before scaling. Rates below 65% indicate a campaign window where the next move is investigation, not investment.

The review operates on three escalation levels.

Level 1 — mid-market DTC and Shopify-stack merchants

A standard Real Buyer Growth Evidence Review covers paid social, influencer, and new-customer-discount campaign types at average order values typical of beauty, apparel, supplements, and lifestyle categories. The fixed-scope output answers next-campaign separation questions.

Level 2 — high-AOV jewellery, premium goods, and limited-edition launches

When average order value rises and inventory is constrained, the contamination patterns shift toward returns abuse, dispute manipulation, and chargeback-pattern review. The Extended Campaign Evidence Review (from £3,500 + VAT) adds returns-and-disputes evidence layers.

Level 3 — enterprise commerce integrity

For retail-media networks, sponsorship reconciliation, and audited-evidence-before-payout requirements, the work escalates beyond campaign-window review into structured commerce-integrity infrastructure. The Enterprise Commerce Integrity Review (from £7,500 + VAT) connects into the broader Keigen portfolio, where evidence-before-value-release governance is the parent category.

The 2027 AI buying agent bridge

The category becomes more important as AI-agent commerce develops. Here is why.

AI-agent commerce is moving from theory into payment, search, browser, and commerce-platform roadmaps. The exact adoption curve will vary by market, but the operating question for merchants is already visible: when a buying action is mediated by an agent, assistant, browser, or delegated workflow, the dashboard still needs to distinguish trusted buyer demand from activity that should not be counted as growth.

Two things follow. First, the contamination patterns expand: a sixth pattern — illegitimate-agent traffic — joins the original five. AI agents are not bots in the classical sense, but the merchant's question stays the same: which portion of reported growth was driven by buyers whose intent and identity I would defend in front of finance? Second, the merchants who already operate a verification layer in 2026 are positioned to extend it cleanly into agent-segmented growth verification in 2027. The merchants who do not have this layer enter 2027 with no language for what is happening to their dashboards.

Real buyer growth verification is the foundation on which AI-agent-segmented growth verification is built. The merchants who close that framework gap first will be better placed to defend campaign numbers as agent-mediated commerce grows.

Frequently asked questions

What is real buyer growth?

Real buyer growth is the portion of reported ecommerce growth that survives evidence review against bots, promotion abuse, identity overlap, and attribution-window artefacts. It is a measurement category, not a tool category. The deliverable is a verification rate and a written verdict that names which portion of last campaign deserves to be scaled.

How is real buyer growth different from incrementality or attribution?

Incrementality asks whether a campaign caused conversions that would not have happened otherwise. Attribution asks which channel deserves credit for a conversion. Real buyer growth asks an upstream question: of the conversions the dashboard reports, how many survive an evidence review against contamination patterns? The three layers stack — incrementality and attribution remain useful when the underlying conversions are real. Real buyer growth verifies the underlying conversions before incrementality and attribution interpret them.

What does the verification rate measure?

The verification rate is the percentage of reported growth — typically reported new customers in a campaign window — that survives identity-overlap, traffic-quality, promotion-eligibility, and attribution-stability checks. A specimen output of 71.2% means 71.2% of reported new customers held up against review and 28.8% entered the gap as either contaminated or unverifiable.

What signals count as real buyer growth?

Full-price orders with 14-day stable conversion paths, new accounts that pass identity-overlap and device-fingerprint checks, paid-channel traffic with human-commercial-intent session quality, referrals where referrer and referred have independent identity graphs, and promotion redemptions consistent with merchant-defined eligibility windows.

What does not count as real buyer growth?

Discount-driven first orders with no follow-on engagement, “new” accounts sharing identity-graph nodes with prior accounts, click traffic that fails landing-page session quality, referral chains tracing to a single device or payment instrument, and redemptions enabled by code stacking, eligibility arbitrage, or aggregator-channel leakage.

How long does a Real Buyer Growth Evidence Review take?

A fixed-scope review covers one Shopify-first store and one campaign window — live, upcoming, or recent. Engagement timeline is 5–7 working days from data access to written verdict, followed by a 45-minute findings call. The price is £1,250 + VAT for the standard review.

Does RealBuyerGrowth replace ad fraud detection or attribution platforms?

No. Ad fraud detection operates upstream at the network layer; attribution platforms operate downstream at the credit-allocation layer. RealBuyerGrowth produces an evidence-layer verdict that sits above both, telling finance and marketing whether the outputs of the upstream and downstream layers deserve to be acted upon for the next campaign.

Get my growth evidence review

The Real Buyer Growth Evidence Review is a fixed-scope written review of one Shopify-first store and one campaign window, delivered in 5–7 working days at £1,250 + VAT. The output is a verification rate, a written executive verdict, the top three reason factors driving the gap between reported and real growth, and a recommended next-campaign separation.

Applications are reviewed before payment. We reply within 24 hours.

Get my growth evidence review

For teams not yet ready to commission a review: read the RealBuyerGrowth method for the methodology behind the verification rate, or read the AMS framework for the parent operating model under which RealBuyerGrowth and the rest of the Keigen portfolio sit.