Hank Leber (@hankleber) header
Hank Leber (@hankleber) profile picture

Cofounder, @stanify_ai: AI-Powered Community Management ✨ Cofounded @giggsterHQ and ExecuteLA (acquired by Hawke Media)

Los Angeles, Ca
Joined March 2009
Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)Photo by Hank Leber (@hankleber)
Reddit just called its conversations an "extremely valuable corpus of data." "Extremely valuable corpus" is business-speak for gold. In its latest shareholder letter, Reddit says it's shifting to a logged-in-only experience. It starts with tests in select markets outside the U.S. Some logged-out access will stay, for people arriving from a shared link or search. That's a real reversal. Steve Huffman told CNBC: "Reddit is the last platform to provide basically unfettered logged-out access, and that's largely because of me. I've been holding that line." The letter's reasons: more signals for ads targeting, higher engagement from logged-in users, and this one. It "allows us to protect access to Reddit's extremely valuable corpus of data." The pressure is search. Per the letter: "During October, we have seen that search traffic has dropped by more than 1 million referrals on average per week." And on the earnings call, Huffman said: "The reality is, AI Mode does not drive traffic. Not for us, and not to anybody, as far as I know." So as the clicks dry up, part of Reddit's answer is to protect the conversation itself. Now look at your own comment sections. The same kind of material is sitting there. Customers describe their problems in their own words, ask questions before they buy, raise objections nobody puts in a survey, share product ideas, and complain in ways that tell you who's about to leave. And a lot of teams treat all of it as cleanup. Hide the spam, answer a few, move on. Reddit is building a business decision around the value of its comments. Plenty of brands haven't even read theirs. What I'd do this week: 1. Pull the last 30 days of comments across your channels and read them in one sitting. 2. Tag what you find: questions, objections, complaints, purchase intent, praise. 3. Take the five questions people keep asking and answer them in your next posts. 4. Write down the exact phrases customers use about your product. That's your copy, your FAQ, and the language you want people to see when they ask AI about your category.
3
1
94
Martha Martha Martha. Her statement said one thing. Her comment said another. Her apology said a third. All on the same Sunday. Martha Stewart's statement about her homeless-person Halloween costume Sunday morning: It "was intended to bring more awareness to the very serious subject of homelessness," and offending people "was never my intention." Then her comment later that day under JustJared's Instagram post about it: "My costume was designed to bring the homeless and the hungry to everyone's mind and boy it certainly did!!! But not in the correct way, because I did not explain it and I am so sorry..." It ended with "Do all of you???" Then her apology that evening, on her own feed: "It's hard to admit when you've really screwed up. My earlier statement didn't go far enough." And: "Wearing a costume like I did was insensitive and not funny." Three separate takes. Meanwhile, the feed kept going. The costume had been one slide in a Saturday carousel on her personal Instagram, captioned "My QVC candles are perfect for Halloween!!!" It was deleted quickly. Sunday morning brought a saffron post. On Monday came free pastries from a Brooklyn bakery, with "Sooooo good!!!" in the caption. One reply: "Guess she's not hungry anymore?" So we have three messages, on three channels: 1. a press statement 2. a media outlet's comments 3. her own feed. All of them in public, right next to the routine posts. This is a community management lesson for everyone (ie it applies to non-celebs too): 1. Pick one channel for the response, and own it there. 2. Don't explain yourself in other people's replies. Once there's a statement, everything else should point to it, not add to it. 3. When the heat is on, pause scheduled and routine posts until the comments calm down. 4. Read the dang comments before the next post. Founders and execs reply from their personal accounts all the time. If you / yours dropped a comment under a news post tonight, who on your team would know before the screenshots did? Figure this out now – not after a mess.
1
81
If you're a global brand that needs to apologize, do it in the right places. 🌎 Three apologies about China this week. Three very different choices about channel and language. Sanrio ended its Hello Kitty merch tie-up with the Japanese drama Vivant. Its China subsidiary apologized late Oct 28 on Hello Kitty's official Weibo. Chinese users then asked why it hadn't been posted on Sanrio's overseas platforms. The next day, the Japanese parent posted a statement in Japanese on its corporate site. As of today, the apology exists on Weibo and in Japanese on the corporate site, but there's nothing in English and nothing on Sanrio's global social accounts. The words barely changed. The story kept changing because of who could see them. Meanwhile, Mbappé had a snafu involving China, and he posted once on X – in Chinese, French, AND English. Real Madrid put its statement on the club site in at least five languages, and on its Chinese Weibo. ^ It came after three sponsors had already dropped him. But nobody had to hunt for a version they could read. The ops lesson for brand and social teams is to know your audience, then pick the channel and language before you write a word. ↳ Where do the people you hurt actually read? ↳ Who else will screenshot it and ask why they didn't get one? ↳ Which of your accounts stays quiet, and can you defend that? To the audience that gets left out, an apology posted in one region reads like containment. 🙅‍♀️
113
I work on a product that writes AI replies. And this week I mostly agree with the people trying to get them banned from X. Pieter Levels wants zero tolerance: post one AI reply and your account gets suspended for life. He calls AI replies an "existential threat to the future of this platform." Ethan Mollick says each of his posts gets "dozens of obvious and numbing AI slopreplies." Worse, when real people reply (his example: someone from Anthropic), X's ranking "does not surface them first." So empty replies cost you twice. They waste your time, and they bury the reply that mattered. But let's be real. AI didn't ruin replies. Empty replies did. You know the ones: ↳ restate your post back to you ↳ call it a "great insight" ↳ end with a question nobody asked A person can write that. Plenty did before LLMs. AI just made it free to send ten thousand. So the test isn't "did a human type this?" It's "does this add something a human would?" Nikita Bier, X's former head of product, wrote this week that every anti-spam fix is "a bandaid for 3-12 months." Platforms will always be chasing bots. Brands shouldn't play that game. The job is to know the rules that already exist and win under them. X's own automation rules: ↳ Any AI reply bot needs X's "prior written and explicit approval." Allowed, not banned. ↳ Auto-replies only to people who "opted in." "Sending automated replies to posts based on keyword searches alone is not permitted." ↳ "Advertisers, publishers, and brands using auto-response campaigns must request approval from X." What brands must do: 1. Keep a human in the loop, and use AI to scale their humanness not automate careless actions. 2. Reply where people talked to you, not wherever your keyword shows up. 3. Ask yourself: will the commenter be glad they read this reply? Simple. That's how we think about it at Stanify. AI reads busy threads and tags each comment by intent, topic and urgency, so the ones that matter don't get buried. The brand's human team decides what goes out. This is fighting fire with fire. And that's the game at hand – not cat & mouse. I keep saying it: AI is leverage for human-ness – not a replacement for it. Where do you sit? Is an AI-drafted reply fine if it actually helps? #ai #socialmedia #communitymanagement
1
2
110
Two mascots got roasted by the internet this month. One nailed it. ✅ The other made it worse. ❌ Meet Liberty Biberty. Liberty Mutual's fuzzy yellow puppet, built by Jim Henson's Creature Shop and named after an old ad where an actor flubs the company name. Plenty of people hate it. A Facebook group devoted to hating it has 40,000+ members. John Oliver called it "Big Bird's sleep paralysis demon." On Oct 25, several dozen people protested at Liberty's Boston HQ with signs reading "If you give me Biberty, give me death." The 19-year-old organizer called it "the next Boston Tea Party." This is all earned media most brands would need to pay to get. Now Kiwibank. Its sheep mascot Buck was so cute it got thirsted-over online. What a weird f-ing world... One fan's post about it passed 4M views. The bank filed DMCA takedowns, and the posts multiplied. On Oct 12 it posted a video comparing the posts to deepfake abuse, and the backlash got even bigger. And so did their efforts to take it all down. The lesson isn't "leave everything up." Threats, scams and harassment don't belong on a brand page. Def hide them fast. But a joke isn't harm. Dissent isn't harm. Someone loudly hating your puppet means people are talking about your brand. Know the difference before you moderate. Attention is the scarcest resource today. Embrace it.
207
Bacon and egg soda. 🥓 Garlic bread toast airplane. ✈️ Cheese roll sushi. 🧀 ^^ All after a "manual check." Just Eat said every AI menu photo was reviewed before it went live. Love to meet that reviewer. They paused the trial. Apologized. Started pulling the images. The lesson isn't "AI is bad at food." It's that a poorly built human-like reviewer isn't a control. It's process theater. AI is leverage for humans. Not a replacement for them. #ai #socialmedia #marketing
69
Still amazed how many brands don't know this stuff is possible
Hank Leber says 97% of comments on brand pages never get a reply. His fix: an AI that hides the trolls without them knowing, answers in the brand's voice, and turns comments into insights. @stanify_ai on The Index with @afkehaya
1
2
85
Poetry. I’m enjoying this more than any vertical scrolling slop brain candy video out there.
UPDATE ON GERALD AND THE CROW SITUATION: Things have escalated. I repeat. THINGS HAVE FUCKING ESCALATED. For those of you just joining us, yesterday I accidentally entered into diplomatic relations with a gang of crows after Gerald stole my good spoon. Well. This morning I walked outside and Gerald was waiting for me. Alone. Standing directly in the middle of my driveway. No screaming. No hopping around. Just staring at me like a tiny feathered loan shark who had come to collect. I said, “Morning, Gerald.” He tilted his head. I gave him a walnut. He did not eat it. Instead, this motherfucker picked it up, walked three steps to the left, PUT IT DOWN, and stared at me again. Message received. The offering was insufficient. So I went back inside and got more walnuts. When I returned… THERE WERE NINE CROWS. Nine. Yesterday there were seven. Gerald is fucking recruiting. And they were arranged in a SEMICIRCLE. This isn’t a bird friendship anymore. I’m funding organized crime. I distributed the walnuts because apparently I pay protection money now. Everything seemed fine until one of the smaller crows tried taking two. Gerald immediately screamed at him. The little one dropped the second walnut. I have accidentally created an economy. Gerald controls distribution. There’s a hierarchy. There are consequences. I’m pretty sure I witnessed a disciplinary hearing before breakfast. Then Gerald flew away. I thought we were done. WE WERE NOT DONE. Twenty minutes later I looked outside and there was something sitting where Gerald had been standing. A bottle cap. A shiny rock. And… MY FUCKING SPOON. HE RETURNED THE SPOON. Ladies and gentlemen. I believe we have successfully negotiated a trade agreement. Unfortunately, there is now a crow sitting on my mailbox. So I still cannot get my fucking mail. I tried approaching. He puffed up. I backed away. That mailbox belongs to the organization now. At this point I’m just a tenant. Anyway, Gerald has officially been promoted to General Gerald, Commander of the Driveway, First of His Name, Keeper of the Good Spoon, Lord of the Power Line, Collector of Walnuts, and Regional Manager of Whatever The Fuck Is Happening Here. I’ll keep everyone updated. But if I suddenly stop posting… Gerald found out I told you.
3
3
162
Hank Leber retweeted
292
554
312
5,475
2,965,057
Hank Leber retweeted
We're buying 100+ acres of land and developing it into an innovation district for entrepreneurs to live and work. It will be in New Hampshire and it's called Freedom Village. We're unveiling our plans for the first time at Freedom Summit Aug 20-21 where we will gather builders across science, energy, startups, AI, housing and community. Join us to live free and build the future in New Hampshire. Comment BUILD and I'll DM you the link. @jhitchco
99
45
28
353
69,709
💯
A CEO from one of our portfolio companies shared this with their team. I’m re-sharing it with their permission, because it resonated and reflects what all founders and CEOs should be communicating. -- We are living through a period of compounding change. And in moments like this, the biggest risk is no longer making the wrong decision. It is moving too slowly while the world moves around you. There are two paths. We can play defense: - Protect what we have - Optimize what works - Wait for clarity It feels safe. It isn’t. Or we can play offense: - Learn faster than the environment changes - Use new tools to solve old problems in better ways - And create entirely new strategies and businesses That’s where the opportunity is. Challenge yourself to do things faster and better than you have ever attempted. Stay uncomfortable. Stay on the front foot.
1
2
120
I’m desperately trying to start and fund Trade School 2.0 please DM me if you’re that cracked founder
Plumbers and electricians seeing AI replace everyone who went to college
296
85
24
1,855
435,663
Lol fair..
AI Made the Best Anti-AI Meme So Far
57
Smart Substack for GTM & growth (esp. fintech & SaaS): aidansteinbach.substack.com/…
51
Hank Leber retweeted
You have 10K followers. 100k. 10M. Most are ghosts. That's not a community; it's a list. The only question that matters: Who actually wants to talk to you? Stop counting followers. Start making conversations. 1 engaged member > 1,000 silent ones. open.substack.com/pub/stanif…
1
2
89
word of advice: If you’re going to pay $7.5M for a #SuperBowl spot, pay the extra few bucks for scalable servers.
1
1
181
We are in a really dark place as a country. This is civil war type of stuff...
URGENT. FYI. After Charlie Kirk, on the sky app there are people asking for the next person to be targeted. These include JK Rowling, Matt Walsh, Trump, Ben Shapiro, Libs of TikTok, Elon Musk, and Andy Ngo. I took as many screenshots as I could.
2
194
I saw this video this morning on my feed and I think it so eloquently explains what Charlie Kirk was put on this earth to do. “When people stop talking, that’s when you get violence, that is when civil war happens”
1,809
47,037
2,929
217,794
16,844,250