Are dark patterns illegal in the EU? Some are, categorically: a fake countdown timer is on the UCPD blacklist of practices unfair in all circumstances. Others are assessed case by case as misleading actions or omissions. Online platforms carry an additional prohibition under the Digital Services Act, and a dedicated regime, the Digital Fairness Act, is being drafted with dark patterns as a named target. When the European Commission and national authorities screened 399 online shops, they found manipulative practices on 148 of them: fake countdown timers on 42, visual or linguistic steering toward more expensive choices on 54, and important information hidden or de-emphasised on 70.
For an acquirer, though, the legal question is downstream of a commercial one. A target that converts well above its category benchmark is presented as evidence of product-market fit, brand strength, and merchandising skill. Sometimes it is exactly that. Sometimes the conversion rate is the aggregate output of design decisions that a new owner cannot keep making.
This piece is the synthesis of the GP3 pillar. The three articles preceding it each describe a single manufactured conversion signal: the reference price that makes the price look like a bargain, the review base that makes the product look trusted, and the support structure that makes leaving harder than staying. Dark patterns are the general case, and the conversion rate is where they all report.
The Taxonomy a Screen Actually Encounters
Five families recur often enough to be worth naming, because each leaves a different observable trace:
- Manufactured scarcity and urgency. Countdown timers that reset on reload, “only 2 left” counters unconnected to inventory, “17 people are viewing this” figures generated rather than measured. The design goal is to shorten deliberation, which is precisely what the law objects to.
- Drip pricing. The advertised price is not the payable price. Service fees, mandatory insurance, or delivery surcharges surface deep in the flow, after the consumer has invested effort and formed an intention.
- Preselected extras. Warranties, subscriptions, expedited shipping, or donations ticked by default, so that the consumer pays by not noticing rather than by choosing.
- False hierarchy. Visual and linguistic weighting that steers toward the option better for the trader: the cheaper delivery greyed into near-invisibility, the decline link rendered as plain text beside a saturated button, confirmshaming copy that makes the negative choice feel foolish.
- Asymmetric exit. Frictionless entry, effortful exit. One click to subscribe, a support queue to cancel. This is the pattern the customer-service piece maps in detail.
None of these require a manipulative intent to be documented. They are design states, observable by anyone who walks the flow.
Where This Sits in EU Law: Three Layers, and a Fourth Being Drafted
The blacklist. Annex I of the Unfair Commercial Practices Directive lists practices unfair in all circumstances, needing no case-by-case assessment. Point 7 covers falsely stating that a product will be available only for a very limited time, or only on particular terms for a very limited time, to elicit an immediate decision. A countdown timer counting down to nothing sits squarely inside it. The Omnibus Directive added the review entries to the same list, which is why manufactured reviews and manufactured urgency are legally closer relatives than they look.
The general clauses. Most other patterns are handled case by case under UCPD Articles 6 and 7, as misleading actions or misleading omissions. Drip pricing is the classic candidate: whether a given implementation misleads depends on what was disclosed, how prominently, and at what point in the flow. Case-by-case does not mean unenforced, as the sweep numbers show; it means the assessment is contextual rather than automatic.
The consent rule. Article 22 of the Consumer Rights Directive requires express consent for any payment beyond the agreed price for the main obligation, and states that such consent cannot be inferred from a default the consumer has to reject. That is a direct answer to preselected extras, and it carries a refund consequence where it is breached.
The platform layer. Article 25 of the Digital Services Act prohibits providers of online platforms from designing or operating interfaces that deceive, manipulate, or materially distort a user’s ability to decide freely. Two boundaries matter for diligence. It binds platforms rather than every trader, so a brand’s own webshop is governed by the UCPD rather than by Article 25, while the marketplaces it sells through are governed by both. And it explicitly does not cover ground already occupied by the UCPD or the GDPR, so it supplements rather than duplicates. For a marketplace-dependent target, this connects to the enforcement dynamic mapped in the DSA piece: the platform polices its own interface obligations by policing its sellers.
The regime being drafted. The Commission’s Digital Fairness Act names dark patterns, addictive design, and unfair personalisation as its targets, with a proposal expected during 2026. Nothing in it is law yet, and its final scope is genuinely unknown, so no diligence conclusion should rest on it. The direction of travel, though, is not ambiguous: the practices below are moving from patchwork coverage toward dedicated regulation, and a target whose conversion depends on them is positioned against that current rather than ahead of it.
The Fake GM3 Problem
The mechanics are the ones the GP3 pillar has now described three times, which is the point of naming the general case. Manipulative design lifts conversion. Higher conversion lowers cost per acquisition. Lower CAC reads as marketing efficiency, and GM3, the layer that measures exactly that, records the lift as skill. Nothing in the P&L records the cause, because interface design is not a cost line.
What makes it a repricing question rather than an ethics one is what happens to that lift under new ownership. Three routes lead to the same place. An acquirer’s counsel reviews the checkout and requires changes. A national authority acts, whether from a sweep, a competitor complaint, or a consumer body. Or a marketplace enforces its own interface obligations against the seller. Each route removes a mechanism the conversion rate depends on, and conversion reverts toward what the product earns unaided. The lost volume then has to be repurchased with media spend, which is visible, or foregone, which is worse. A target with a category-normal conversion rate and clean flows has less headline appeal and a number that transfers intact.
There is a second-order effect worth pricing separately. Patterns that manufacture consent to extras, or that suppress cancellations, do not merely inflate conversion; they inflate average order value and retention too. Those are the inputs to the lifetime-value assumptions a buyer underwrites the deal against. The manipulation compounds through the model rather than sitting in one line of it.
Why This Slips Through Due Diligence
The checkout is nobody’s assigned territory. Commercial diligence examines the conversion funnel as a set of metrics rather than as an interface, and metrics cannot show why a step converts. Legal diligence reads terms, policies, and disclosures, which are usually correct in text and irrelevant to the question, because a dark pattern is a matter of prominence, sequence, and default rather than of missing text. Financial diligence has no line item to interrogate. And the target’s own team may sincerely describe these patterns as conversion-rate optimisation, because that is what the tooling and the vendor case studies call them.
The practice is visible only by using the site as a customer, in sequence, and observing where the interface stops informing and starts steering.
What Outside-In Analysis Can Detect Before the Data Room
- Walking the full purchase flow to a completed order, recording every screen: what is preselected, what is disclosed and when, and where the price changes
- Reloading pages with countdown timers and scarcity counters to see whether the numbers reset, persist, or move independently of any plausible inventory
- Comparing the advertised price with the final payable total, and identifying at which step each additional charge appears
- Testing the exit paths that mirror the entry paths: cancellation, unsubscribe, return initiation, and account deletion, timed and counted in steps against the equivalent entry action
- Checking consent architecture for defaults that require rejection rather than selection, including cookie and marketing consent, which is the subject of a related exposure in its own right
- Reading reviews for the language of surprise: “was charged for,” “didn’t realise I had signed up,” “price changed at checkout,” “couldn’t cancel”
- Benchmarking the target’s stated conversion rate against category norms, treating a material positive gap as a prompt to examine the flow rather than as a conclusion in itself
None of this establishes a breach; that assessment belongs to counsel and to authorities. What it establishes is how much of the target’s conversion advantage is attributable to design that a new owner would have to defend, and how quickly the flows could be brought to a defensible state.
The Pre-LOI Question Every PE Fund Should Ask
An above-market conversion rate is the number a seller most wants examined and the one least often interrogated, because interrogating it feels like arguing with good news. The question that separates the two explanations: if every manufactured signal were switched off tomorrow, the timers, the defaults, the deferred charges, the asymmetric exits, what would this target’s conversion rate be, and how much media spend would it take to buy the difference back?
A target that converts on product, price, and genuine reputation is selling an advantage that survives the transaction. A target that converts on interface design is selling an advantage that belongs to the design, and the design belongs to whoever is willing to keep running it.
This analysis is part of Tronvik’s GP3 Waterfall methodology, focused on GM3 marketing-efficiency mapping. Nothing in this article constitutes legal advice. To initiate an outside-in dark-pattern screen on a specific acquisition target, contact info@tronvik.com.