Maxwell Finn

The maker is too close to his product and sees in his methods only the ordinary. That's Claude Hopkins in his 1927 memoir explaining why almost nobody does what he did for Schlitz and I think it's still the reason most ads say "high quality" when the real proof is sitting in the building. A couple more ways to find yours on top of the steps in the video: 1. Ask your newest hire what surprised them most in their first week, because they haven't gone blind to it yet. 2. Go through your support inbox for the "wait, you actually do that?" replies, because that's usually your best ad angle sitting there unused. His line on it: "Few who dominate a field have any exclusive advantage. They were simply the first to tell certain convincing facts."

In 1951 Yale researchers gave students the same article and told some of them it came from Oppenheimer and others that it came from Pravda. Right after reading, 23% moved toward the Oppenheimer version and only 6.6% toward the Pravda one. Four weeks later the gap had almost completely closed and on the submarine article the Pravda version went up 13.9 points, because people remembered what they read long after they forgot who said it. This is called the sleeper effect and the catch later research found is that it only shows up when the reason to doubt comes after the argument and not before it. TLDR: if your video ad opens on the logo or the pitch, you're handing people the reason to doubt before they've heard the argument. Test a version that makes the strongest point first and judge it on 7-day click conversions, not day one.

Most people decide whether their cost per customer is good by comparing it to what someone else says is normal, but the number that actually matters is the one your own margin can pay for. Imagine two brands that both sell a $100 product and both pay $45 in ads to get a new customer. The first one keeps $60 of every order after the product, shipping, payment fees and returns, so each new customer puts $15 in its pocket on the first purchase. The second one only keeps $30 after those same costs, so it loses $15 every single time the ads do their job. Same price and same ad results, but one of them should be raising the budget while the other one loses more money the more it spends. Before you scale, take your average first order and multiply it by the percentage you actually keep after those costs. That's the most you can pay for a customer and still break even on the first purchase. Then compare it to what you're paying right now and see how much room you have, because what you pay per customer usually goes up as you spend more and start reaching people who are less ready to buy. If repeat purchases are a real part of your business you can afford to pay more than that, but only once you've actually counted how many first-time buyers come back.

If you're planning a price increase this year, study Qwikster first because Netflix did pretty much every part of it the hard way. The apology was honest (Reed Hastings wrote that he "slid into arrogance based upon past success") and it still backfired because the same email announced another change. Netflix's own shareholder letter said the Qwikster split was "relatively minor" compared to the price change, so the real damage came from a price hike that landed without a reason people believed. If you're raising prices tell existing customers before it hits their card with the actual reason and if the price and how people buy both need to change, ship them far enough apart that you can tell which one people are reacting to.

In 1982 about 50 towns canceled trick-or-treating because someone laced Extra-Strength Tylenol capsules with cyanide and seven people in the Chicago area died. J&J didn't stop at the two bad lots. It pulled every capsule in the country (it told Congress 31 million bottles came out of stores) even though the FDA tested 1.7 million capsules and found no cyanide outside Chicago. The recall cost about $100 million before taxes. On November 11 Tylenol came back with three seals you could check before you ever took a pill and by February 1983 it was back to about 28% of the market and still climbing. This is called costly signaling and the basic idea is that when people can't judge quality for themselves, they believe the signals that would be too expensive to fake. Most landing pages do the opposite and slap on a "satisfaction guaranteed" badge anyone could put on any page. Try putting the actual refund steps (how to ask, how fast, who handles it) next to the buy button and compare checkout conversion rate and refund rate against the badge.

Before a difficult client call, I'd write down exactly what I think we're disagreeing about, then ask the client what they expected to see by now so we can work on the actual gap instead of defending everything at once.

If you sell anything with a bonus attached, watch which part of the offer people get excited about, because it might not be the part you think you're selling. My favorite detail in the Wrigley story is that most of the country barely chewed gum when he started giving it away, so he wasn't chasing a trend, he was following what his customers kept picking. By 1971 the company offered no premiums or prizes at all and his son said they "used to do about 17 brands and we weren't getting anywhere" before focusing almost everything on three. If you run a bonus stack, look at which bonus people bring up on sales calls or in reviews and test an ad that leads with it instead of the main product. The truth is your customers are telling you what to sell all the time, you just have to watch what they pick when you aren't forcing the choice.

Charging even a few dollars for a trial usually gets you fewer signups, but the people who pay have already decided the problem is worth money to fix, which makes them a different buyer than someone grabbing a free login. Free still wins when people have to use the product before they can see why it's worth paying for.

If you're not the biggest company in your market, one of the smartest things you can do is say it before your customer does. Avis did it in 1963 and what I love most is how specific the copy got, all the way down to saying their people should know where you get a good pastrami sandwich in Duluth. Most smaller brands I see do the opposite and try to look bigger than they are, which nobody believes anyway. If you cost more than the big guy, say it first and then tell people exactly what the extra money pays for. P.S. The photo in the original ad is a hand holding up two fingers, a pretty genius way to say number two without saying it.

If you want to become more useful to a client, ask what marketing work keeps getting pushed back because nobody owns it, then pick one specific job you can do consistently before adding more services to the proposal.

Nice! I’ve always wanted an excuse to look like a mentally ill crazy person in public. This also might win an award for the most dystopian tweet of the year.

The average American doesn't realize a huge percent of the political fundraising emails and texts they get bombarded with are not actually being sent directly from the candidate. They are being sent by businesses who own lists as affiliates and are being paid up to 50% of the donations they generate. This means for a lot of people only $50 of that $100 you donated to a candidate actually went to their campaign and the rest went to some person or company.

Every December hundreds of millions of people post an ad for Spotify and all Spotify does is show them their own listening data. Year in Music launched in December 2015 and Spotify says more than 5 million people used it. It got renamed Wrapped in 2016 and by 2018 everyone got a share card built for Instagram, Twitter and Facebook. From Spotify's own shareholder letters: 28 million people visited the site in one week in 2018 (vs 20 million for all of 2017), the first 24 hours of 2020 Wrapped beat all of 2019's engagement and last year it hit 300 million+ people and 630 million shares. Apple copied it with Replay in 2019 and YouTube Music followed in 2020. It works because people aren't posting about Spotify, they're posting about themselves (identity signaling, from a 2007 consumer research paper). High IQ move for any brand with customer data: send each customer a recap of their year with you on a Stories-sized card with a tracked link, then measure shares and new customers from those shares.

Some simple advice for business owners running ads… Do not add another agency or contractor to your ad accounts without telling your current marketer and explaining why. 1. They can see this. 2. Without any context the obvious conclusion they will come to is you’re planning to replace them. 3. It’s disrespectful to not be transparent and honest with someone who has worked incredibly hard to grow your business. 4. Don’t be shocked if they stop giving your account 100% effort.

When founders brief a new ad they usually want it to say everything they love about the product, so you end up with fast shipping, a great guarantee, better ingredients, a discount and a bunch of reviews all fighting for the same quick glance in someone's feed. The problem is that every extra reason makes the others harder to notice. Someone scrolling isn't going to read five claims and add them up, they're going to catch one thing (if you're lucky) and decide in that moment whether it's for them. It also makes the ad pretty useless as a test. If an ad that says five things wins, you have no idea which of the five people actually cared about, so the next ad you make is a guess too. Before you brief the next one, I'd finish this sentence: "This ad needs to convince [a specific person] that [one specific thing]." Then make everything in the ad (the visual, the headline and the first line of copy) work to prove that one thing. The other benefits aren't wasted. Each one can become its own ad and the landing page is where the people who already care can read about all of them. Run a few of those single-reason ads side by side and the results will start telling you which reason people actually buy for, which is the thing you really wanted to know in the first place.

One of the most underrated copywriting tricks has nothing to do with what your line says and everything to do with how it sounds. The researchers behind this study say rhyme works a lot like repetition and they named their own paper "Birds of a feather flock conjointly" which is a famous saying with the rhyme taken out on purpose. Give the rhyme to the one line you want people repeating back to you, like your offer or your guarantee, repeat that line more than once and run it as its own ad next to the plain version. Most of us obsess over the words in a headline and almost never read it out loud, which is the first thing I'd do with any line I wanted people to trust.

If I were building an AI content routine, I'd start with real customer questions and examples of posts we'd actually publish, because "make something engaging every day" isn't enough direction to get useful output.

Complaining, coming up with excuses and bringing up problems is easy. It is also incredibly unproductive and helps makes things better for no-one. A policy we have had at Unicorn since the early days is before your bring a problem up to the team you need to have attempted at least 2 solutions on your own.

As someone who has worked with clients paying $1k a month all the way up to 6 figures a month throughout my career I can confidently say this is not the reality. The small few grand a month clients are the biggest headache and resource drain by far. They require the most time and CS by far and are the quickest to freak out after a bad day of performance. They also are magically also experts at everything despite having relatively small businesses and second guess you/have opinions on everything you do. I promise you with 10 low paying clients like this you are not taking Fridays off and operating some low stress kick your legs up business. The final thing this post is missing is expenses. He makes it seem like you just bank $20k profit a month, but that is definitely not the case. Unless you have a solid social presence, speak at lots of events and/or incredible referral system you’re going to be investing in marketing. Smaller clients like this churn at the highest rate, so you’re going to need to consistently be marketing to stay at that 10 active client number.

Arm & Hammer had a product that almost never gets used up, so they told people to put a box in the fridge and then gave it an expiration date. In 1972 the brand started telling people to open a box and leave it in the refrigerator to soak up food smells and one 1970s spot opens on a tired old box saying "Remember me? I'm the box of baking soda you left in the refrigerator" and ends on "What took you so long?" when the fresh one shows up. Then it was one for the litter box, one for the carpet and "One in the freezer. One in the fridge." and by 1994 the company said over 80% of American homes had a box in the fridge. This is called extending the product life cycle and the basic idea is that a mature product keeps growing when the people who already buy it use it more often and in more ways. If your product gets used up slowly, put a replace-by date on the package plus a reorder email timed to that day and compare second-order rate and 90-day LTV against a holdout. A lot of brands chase new customers when sometimes the easier growth is sitting in the ones they already have.