CRO / Trust
Do Trust Badges Actually Help Conversion?
// a badge helps most where the brand is least known
By Fracto Solutions
October 27, 2026
The short answer
Nobody has shown convincingly that trust badges lift conversion, and one published test found a small decrease. Use a seal your certificate authority actually issues you, place it near the payment fields, and treat it as support rather than proof. Never display a badge you are not entitled to.
Do trust badges really work?
We want to answer this properly rather than repeat the usual claim. There is no adequately powered, independently verifiable A/B test isolating a trust badge’s conversion effect, the one published negative result found a small decrease, and the lift figures you will find online come from companies that sell trust seals.
Here is what actually exists. GoodUI aggregates seven tests on the pattern and states that its own cumulative power target has not been met, so by the aggregator’s standard the result is not yet conclusive. WiderFunnel published a test where adding a McAfee badge decreased conversions by 1.6%, which its author attributes to placement rather than to the badge itself. And the figures that circulate most widely, from Holabird Sports, Decor Steals, Scrubs and Beyond and similar, are TrustedSite case studies; TrustedSite is the former McAfee SECURE badge business and sells trust seals.
The academic work cuts against the premise. Kirlappos, Sasse and Harvey, “Why Trust Seals Don’t Work” (TRUST 2012), tested 60 experienced ecommerce users across six sites with and without seals. 38% failed to notice any seal present. Among those who did, the authors found a dangerous belief that a seal “automatically legitimizes any website”. Their conclusion was that seals give inadequate protection against fraudulent sites.
We should be even-handed: we also found no rigorous demonstration that badges hurt conversion by creating anxiety where none existed. That is CRO folklore in the other direction. The defensible position is that the direction of the effect is unestablished and probably depends on placement and context. This is the detail under the trust pillar.
What does the Baymard seal research actually say?
Less than it is usually quoted as saying, and it is old. The survey was fielded in January 2013, it measured stated trust preference rather than recognition or behavior, and 49% of respondents answered “don’t know” or had no preference.
Two further details matter. An independent CXL survey in October 2016 with 2,100 US respondents produced a different ranking, putting PayPal first, and its authors note that how people say they would behave and how they do behave are not the same thing. And in a later round, Baymard found a homemade seal outperformed every established SSL seal bar one.
That last finding is worth sitting with. If a seal somebody made up performs like a real one, seals are measuring visual familiarity, not security. It is the strongest reason to treat any badge as decoration that may reassure, rather than as evidence of anything.
Which badges should I use, and which should I never use?
Use the seal your certificate authority actually issues you, plus the payment and wallet logos you are licensed to display. Never display a badge you are not entitled to, which is a per-se offence in the EU and UK and a trademark problem in the US.
We used to name a specific security seal here, and we have removed it. The Norton Secured Seal was discontinued and automatically replaced by the DigiCert seal on 16 to 17 October 2023. It is not a product you can buy any more, so the only way to display one today is to use an image you have no license for, which is precisely the practice described below.
Badge type
What the evidence says
Rules
Seal from your own CA
Effect unestablished; may reassure
Licensed to you, so safe to display
Payment and wallet logos
Familiarity reassures; untested
Follow the network's brand guidelines
Money-back guarantee
Reduces perceived risk
Must be real and honored
Self-made trust graphics
Beat real seals in one Baymard round
Fine if they claim nothing untrue
A seal you don't hold
Not applicable
Banned outright in the EU and UK
Note the last row. Displaying a trust mark without authorisation is unfair in all circumstances under UCPD Annex I point 2 and DMCCA Schedule 20 paragraph 3.
The legal position deserves spelling out, because it is stronger than “a trust risk”. In the EU, Annex I point 2 of the Unfair Commercial Practices Directive bans “displaying a trust mark, quality mark or equivalent without having obtained the necessary authorisation”, and Annex I practices are unfair in all circumstances, with no need to prove any consumer was harmed. The UK carries identical wording at Schedule 20 paragraph 3 of the DMCC Act 2024, in force 6 April 2025, where the CMA can now fine directly, without going to court, up to 10% of global turnover. In the US you face FTC Act §5 deception and, separately, trademark infringement under Lanham Act §§32 and 43(a), which means the mark owner can sue you directly. General information, not legal advice.
Where should badges go?
Near the payment and card fields, where security anxiety peaks, rather than buried in the footer. Baymard’s testing found that perceived security is driven by gut feeling and visual cues, and that reinforcement “must be distinct and unique for the credit card fields”.
Be precise about what that finding is: it comes from moderated in-lab usability sessions published in 2016 and updated with 2023 data, and it measures perceived security, not conversion. We know of no study testing badge placement against conversion. Our practice is to put any reassurance where the worry lives and make the payment section visually distinct, which also supports reducing checkout abandonment, but we are offering that as a sensible default rather than a measured one.
One boundary worth naming. The same Baymard research says users cannot assess your encryption. That means “looking secure” and “being secure” come apart, and advice to make a page look safe is only honest alongside actually using a reputable hosted checkout. Simulating security is the version regulators treat as deceptive.
What can a badge not do on its own?
They cannot replace real proof, and on the current evidence they probably cannot do much on their own either. A badge on a site with no reviews, no policies and a scruffy checkout is not going to carry it.
Build the fundamentals first: reviews, clear and findable policies, honest contact details, professional design, a checkout that behaves. Those are also the things the law requires in several cases, which is covered in About and policy pages. Then add a seal you actually hold, near the payment fields, and measure whether anything moves. If you cannot detect a difference, that is consistent with the published evidence and not a sign you installed it wrong.
We use recognizable badges where they help, on top of real trust.
We put the right security and payment cues where anxiety peaks at checkout, skip the meaningless graphics, and make sure the fundamentals, proof, policies, and a reassuring checkout, are in place first, then measure whether conversion actually moves.
It’s the same measurement-first approach behind our conversion optimization work, where we rebuild the buying path and measure the result .
// we would rather show up in your reporting than in your inbox
Frequently asked questions
Do trust badges increase conversion?
There is no adequately powered independent A/B test showing that they do. One published test found a 1.6% decrease, and academic work found 38% of experienced shoppers did not notice seals at all. The lift figures circulating online come from companies that sell trust seals, so treat them as marketing rather than evidence.
Which trust badge is most effective?
Use the seal your certificate authority issues you, plus payment logos you are licensed to display. The seal most often recommended in older guides, the Norton Secured Seal, was discontinued in October 2023, so displaying one now means using an unlicensed image.
Is it illegal to display a badge I don't have?
In the EU and UK, yes, outright. Annex I point 2 of the Unfair Commercial Practices Directive and Schedule 20 paragraph 3 of the UK’s DMCC Act both ban displaying a trust mark without authorisation, with no need to prove harm, and the CMA can fine up to 10% of global turnover. In the US you face FTC Act section 5 and a trademark claim from the mark owner.
Where should I put trust badges?
Near the card and payment fields rather than the footer. Baymard’s lab testing found perceived security is driven by visual cues and that reinforcement must be distinct around the credit card fields, though that measures perception rather than conversion, and no study tests badge placement against sales.
Can trust badges backfire?
They can if they are fake or unlicensed, which is a legal problem rather than a conversion one. On the conversion side the evidence is thin in both directions: one test found a small decrease, but we found no rigorous demonstration that badges create security anxiety where none existed.
Sources
Do badges lift conversion? No good public evidence. We found no adequately powered, independently verifiable A/B test isolating a trust badge’s conversion effect. GoodUI aggregates seven tests and states its own power target is not yet met. WiderFunnel published a negative result: a McAfee badge decreased conversions by 1.6%, which its author attributes to placement. Kirlappos, Sasse and Harvey, “Why Trust Seals Don’t Work: A Study of User Perceptions and Behavior”, TRUST 2012, Springer, pages 308 to 324: 60 experienced ecommerce users across six sites with and without seals; 38% failed to notice any seal present, and those who did tended to believe a seal “automatically legitimizes any website”. The lift figures circulating online (Holabird Sports, Decor Steals, Scrubs & Beyond and similar) are TrustedSite case studies, and TrustedSite sells trust seals. We also found no rigorous demonstration that badges reduce conversion by creating anxiety, so we do not assert that either. https://discovery.ucl.ac.uk/1353955
Trust seals: Baymard Institute site-seal survey, fielded 9 to 11 January 2013 via Google Consumer Surveys, 2,510 US adults with 1,286 answering on actual seals. The question asked which badge gave the best sense of trust, which is stated preference rather than recognition or behavior, and 49% answered “don’t know” or had no preference. A later Baymard round in 2023 found a homemade seal outperformed every established SSL seal bar one. An independent CXL survey in October 2016 (2,100 US respondents) put PayPal first instead, and its authors note that how people say they would behave and how they do behave are not the same. https://baymard.com/blog/site-seal-trust
Why we no longer name a specific seal: the Norton Secured Seal, for years the most commonly recommended badge and the leader in Baymard’s 2013 survey, was discontinued and automatically replaced by the DigiCert seal on 16 to 17 October 2023. It is no longer a product you can buy, which means any Norton badge on a site today is an unlicensed image and therefore the exact practice banned by UCPD Annex I point 2 and DMCCA Schedule 20 paragraph 3. Recommend the seal your certificate authority actually issues you, and nothing else. https://techdocs.akamai.com/cps/changelog/oct-17-2023-digicert-site-seal-replaces-norton-site-seal-1
Displaying a badge you are not entitled to: banned outright in the EU by Annex I point 2 of the Unfair Commercial Practices Directive, “displaying a trust mark, quality mark or equivalent without having obtained the necessary authorisation”, unfair in all circumstances with no need to prove harm. The UK carries the same wording at Schedule 20 paragraph 3 of the Digital Markets, Competition and Consumers Act 2024, in force 6 April 2025, where the CMA can fine directly up to 10% of global turnover. In the US, FTC Act §5 deception plus trademark infringement under Lanham Act §§32 and 43(a), which exposes you to a private suit from the mark owner, not just a regulator. https://www.legislation.gov.uk/ukpga/2024/13/schedule/20
Perceived security at checkout: Baymard Institute, “Perceived Security of the Payment Form”, published 5 October 2016 and updated with 2023 data, from moderated in-lab usability sessions. The average user’s perception of security is “largely determined by their gut feeling”, directed “to a large extent” by how visually secure the page looks, and “the visual reinforcement must be distinct and unique for the credit card fields”. Qualitative observation of perception, not a measured conversion effect, and it predates universal HTTPS and hosted wallets. https://baymard.com/blog/perceived-security-of-payment-form
Stated as Fracto’s position, not research: that badges belong near the payment fields rather than the footer, and that they support trust built elsewhere rather than creating it. Baymard’s placement finding is about perceived security in lab sessions, not a measured conversion effect, and we found no study testing badge placement against conversion. This is our practice, offered to test.