Embracing Marketing Mistakes · 2026-07-02 · 4 min
Key moments - from our scoring
Substance score
32 / 100
Five dimensions, 20 points each
Speaker_01 dissects a counterintuitive marketing pitfall: negative social proof, where campaigns inadvertently encourage the exact behavior they're trying to prevent. The episode explores how marketers misapply social proof by highlighting undesirable actions, accidentally making them seem normative. Three case studies illustrate the problem: a British university's binge-drinking awareness campaign that backfired by stating "the average student drinks eight pints a week" - making light drinkers feel they should increase consumption; Robert Cialdini's Arizona petrified wood study showing that signs saying "people steal 100 tons yearly" actually increased theft until simplified to "please don't steal"; and Wikipedia's donation messaging claiming "only 1% donate," which discourages giving rather than encouraging it. The conversation emphasizes how behavioral psychology principles like social proof can be weaponized unintentionally, turning cautionary messaging into permission-giving. Marketers, behavioral designers, and anyone running awareness or fundraising campaigns will find actionable insights on how to flip these dynamics by reframing social proof positively - using localized, inclusive messaging like "4,000 Bristol residents donate monthly" instead of highlighting the non-compliant majority.
Negative social proof occurs when marketers highlight that many people engage in an undesirable behavior, inadvertently making it seem normal or acceptable. For example, saying "the average student drinks eight pints a week" to discourage drinking actually makes light drinkers feel they should drink more to fit in, and heavy drinkers feel validated.
Cialdini found that park signs stating "people steal 100 tons of petrified wood yearly" actually increased theft, because visitors thought a small piece wouldn't matter. When the signs were changed to simply "please don't steal the petrified wood," theft decreased significantly.
Instead of saying "only 1% of people donate," Wikipedia could use positive social proof by stating something like "4,000 Wikipedia lovers in Bristol donate monthly - could you join them?" This approach makes giving feel like joining a community of supporters rather than being in the rare minority.
By stating the average student drinks eight pints weekly, the campaign made students who drank less feel like outliers who should increase consumption to be normal, while heavy drinkers felt validated, causing overall drinking to increase rather than decrease.
Our reviewer’s read on each dimension, with quotes from the episode.
The episode conveys one legitimate behavioural science concept - negative social proof - with three illustrative examples in four minutes, which is a reasonable ratio, but there is only a single idea and no novel extension beyond what Cialdini already published. Filler ('It's hilarious', 'Yeah, I know, I know') eats into a short runtime.
negative social proof is definitely a mistake I see a lot of companies make
by saying the average student drinks eight pints a week will make everybody who drinks less think, oh my God, I need to drink more
Every example and the core concept come directly from Robert Cialdini's widely read 'Influence' - the petrified wood study is arguably the single most cited Cialdini field experiment - making this almost entirely a retelling of canonical, well-circulated material with no new angle or first-principles extension.
Robert Cialdini, who's the author behind the book Influence, he went into this park
This is um is another example as well
The guest ('Phil') demonstrates solid familiarity with behavioural economics research and applies it to marketing contexts, but the transcript reveals no practitioner credentials, scale of work, or personal test data - he is summarising published academic studies rather than sharing hard-won operational experience.
There's this concept of negative social proof
Robert Cialdini, who's the author behind the book Influence, he went into this park
The guest cites concrete numbers (eight pints a week, ~100 tons of wood per year, 1% donation rate, 4,000 Bristol donors) which is above average for a short episode, but the sourcing is vague ('I think it was a European university… let's assume it was a British one') and the numbers are presented loosely rather than with precise citations.
the average student drinks eight pints a week
people who enter this park steal one point or something like a hundred tons of this wood every year
The host poses one broad, show-title-prompted question ('What are the biggest mistakes you see marketers making?') and contributes only a single short interjection for the rest of the segment; there is no follow-up probing, no pushback on the absence of precise citations, and no attempt to connect the examples back to B2B marketing practice.
What are the biggest mistakes you see marketers making with behavior in in your area?
And so he's my piece doesn't matter.
Computed from the transcript - who did the talking, and the words that came up most.
Behavioural science meets marketing mishaps back in episode 46 with Phill Agnew. As a specialist in testing what really works in ads, Phill exposes how well intentioned campaigns often push audiences in exactly the wrong direction. He unpacks the idea of negative social proof, where highlighting how many people binge drink, steal from parks or ignore donation prompts can actually signal that this behaviour is normal and acceptable. Phill shares stories from university alcohol campaigns, an American park full of disappearing petrified wood and even the familiar fundraising banner on Wikipedia to show how this plays out in the real world. He then explains how a small shift in wording, audience framing and local proof can dramatically change response rates. For marketers who love data but still rely on gut feel when choosing headlines, this conversation offers a smarter way to use social proof without accidentally encouraging the behaviour you are trying to reduce. Here's the link to the full episode 46. Send us Fan Mail Is your strategy still right in 2026? Book a free 15-min no obligation discovery call with our host: [ Book your call with Chris now ]
Transcribed and scored by The B2B Podcast Index.
1 - > SPEAKER_00: What I like about your content that you share, 2 - > Phil, is like you share, you do all your own tests as well, 3 - > don't you? 4 - > So you'll say, like, we tested this against this, and this was 5 - > the di this was the difference. 6 - > And it's usually just one picture of one ad one way and 7 - > one picture of an ad another way. 8 - > What what are the biggest?
9 - > Obviously, my show is called embracing marketing mistakes. 10 - > What are the biggest mistakes you see marketers making with 11 - > behavior in in your area? 12 - > You know, you do you spot them a lot? 13 - > SPEAKER_01: The classic one, the really funny one is because 14 - > we've spoken about social proof already.
15 - > There's this concept of negative social proof. 16 - > So social proof is very powerful. 17 - > We do follow the actions of other people. 18 - > But sometimes marketers get confused and they share that 19 - > lots of people are doing an action which the marketer would 20 - > consider undesirable, and they put that in their campaign.
21 - > And that actually influences the wrong behavior. 22 - > So let me give an example. 23 - > This was a study in universities, I think it was a 24 - > European university, that wanted to reduce the amount of binge 25 - > drinking on campus. 26 - > In fact, if it was binge drinking, it must have been a 27 - > British university.
28 - > So let's assume it was a British one. 29 - > They wanted to reduce the amount of binge drinking. 30 - > And so they said the average student drinks eight pints a 31 - > week. 32 - > This is way above the uh suggested amount.
33 - > You need to reduce the amount you're drinking. 34 - > But actually, if you understand social proof, you realize by 35 - > saying the average student drinks eight pints a week will 36 - > make everybody who drinks less think, oh my God, I need to 37 - > drink more. 38 - > I'm such a loser. 39 - > And maybe those who drink more think, oh yeah, cool.
40 - > I'm above average here. 41 - > God, I'm that's quite cool, isn't it? 42 - > So you're you're encouraging the wrong behavior. 43 - > And then that study, they actually found that drinking 44 - > increased when they showed that.
45 - > This is um is another example as well. 46 - > Yeah, I know, I know. 47 - > It's hilarious. 48 - > And there's another example from an Arizona park where people 49 - > were stealing this petrified wood, which is crystallized 50 - > wood, really, really precious.
51 - > And people were stealing it. 52 - > And Robert Cialdini, who's the author behind the book 53 - > Influence, he went into this park and he looked at the signs, 54 - > and the signs say, people who enter this park steal one point 55 - > or something like a hundred tons of this wood every year on 56 - > average. 57 - > Please don't. 58 - > This is a really bad thing to say.
59 - > SPEAKER_00: And so he's my piece doesn't matter. 60 - > SPEAKER_01: Yeah, well, exactly. 61 - > So if I just take a little one from my daughter that weighs 62 - > about two grams, it won't even make a difference. 63 - > SPEAKER_00: Yeah.
64 - > SPEAKER_01: When and then when he changed the signs that said 65 - > just please don't steal the petrified wood, the amount of 66 - > people who stole actually decreased. 67 - > And I see this all the time. 68 - > If you go on Wikipedia today, probably you will see a message 69 - > that says something along the lines of only 1% of people 70 - > reading this message will donate. 71 - > Please donate to keep Wikipedia running.
72 - > As we the studies suggest that shouldn't work. 73 - > Because what people will think when they read that is, oh, 74 - > nobody really gives anyway, and Wikipedia is quite successful, 75 - > it's not going to go anywhere soon. 76 - > They probably don't need my help. 77 - > Whereas if they said something like, if they wanted to use 78 - > social proof, um, I'm in Bristol, if they said something 79 - > like 4,000 Wikipedia lovers in Bristol donate on a monthly 80 - > basis, could you join them?
81 - > That should, according to the sites, be much more effective. 82 - > And I'd love to see Wikipedia testing that. 83 - > So negative social proof is definitely a mistake I see a lot 84 - > of companies make.
Other episodes covering the same guests and topics, from across The B2B Podcast Index.
How to (Actually) Use Intent Data Effectively - Real Use CasesDemand Decoded: Demand Generation & Business Growth · features Phil60 / 100
Eight Psychology Experiments for MarketersThe Digital Marketing Podcast · on Robert Cialdini92 / 100
How Sales Reps Can Use the Consistency Principle to Close More DealsSales Leadership with Fexingo · on Social Proof78 / 100
How Salesforce Used Trailhead to Build a Marketing EmpireProduct Marketing with Fexingo · on Behavioral psychology78 / 100
How to Stop Guessing and Start Testing with Casey Hill, DoWhatWorksThe Content Cocktail Hour · on Social Proof78 / 100
The Behavioral Science of Buying: Driving Predictable Revenue with Nancy HarhutRevenue Science · on Robert Cialdini74 / 100EP 117: 70% of Teams Are Using AI With Zero Oversight. Here’s What Happened80 / 100How a Transparency Slip Sparked a PR Firestorm57 / 100EP 116: "SEO Is Dead. It's an $85 Billion Market and It's Over"80 / 100The Hangover That Landed Us A Major Client58 / 100
EP 115: The Moment Claude Co-Work Replaced 4 Days of Work in 30 Minutes85 / 100