The Swipe Is Dying, and Bumble Is Ready

For over a decade, the swipe has been the universal gesture of online dating. A flick of the thumb. A match. A maybe. But that era is ending, and Bumble just made it official.

On Wednesday’s second-quarter earnings call, CEO Whitney Wolfe Herd didn’t just talk numbers. She talked about a future where the swipe doesn’t exist at all. And she teased that something new is already in the works.

“What will replace the swipe? I will be keeping under wraps a bit longer for competitive purposes,” she told analysts. “But it is designed to generate more immediate interactions and, most importantly, better outcomes, mimicking real life.”

That’s a big statement. And it lands right as Tinder announces its own pivot toward in-person events. The message from the industry’s two biggest players is clear: the swipe-free future is coming.

Plans: Bumble’s Bet on Group Socializing

The centerpiece of Bumble’s strategy is its newest app, Plans. Launched quietly last month, it’s designed to help people meet “IRL” in low-pressure group settings. Think small dinners, drinks at local spots, and shared interests — not one-on-one coffee dates with a stranger.

Wolfe Herd said the app showed “promising results” in early testing. The company is now leaning hard into what she calls “real-life experiences.”

The logic is simple. Gen Z is tired of endless swiping. They’re tired of superficial matches and ghosting. What they want, according to Bumble’s data, is organic connection.

“Group socializing is a real part of how Gen Z prefers to meet,” Wolfe Herd said. “And we believe Bumble represents a natural bridge from meeting to socializing to then dating.”

Why Groups Beat One-on-One

There’s a psychological comfort in groups. The pressure is lower. The stakes feel smaller. You can test chemistry without the awkwardness of a formal date.

Bumble saw this play out with Bumble BFF, its friend-making app, which gained real traction with young women. Plans takes that same concept and expands it — with the option for romance to develop naturally later.

Users on Plans can chat on Bumble’s platform without ever swiping or sharing a phone number. That’s a deliberate design choice. Bumble is marketing “no swiping” as a feature, not a limitation.

What Actually Replaces the Swipe?

Here’s where it gets interesting. Wolfe Herd didn’t reveal the new interaction model. But she gave hints.

“The core idea is a shift away from optimizing for swipe speed and velocity towards something more intentional,” she said. “Fewer, better, more considered signals.”

That language matters. The swipe was built for volume. It gamified dating, turning human connection into a slot machine. The replacement, according to Wolfe Herd, will be more deliberate. It will eliminate “the friction and delay” of current apps.

She promised the transition would be gradual, so as not to disrupt Bumble’s existing ecosystem. But the direction is unmistakable.

The Numbers Behind the Pivot

Bumble beat earnings expectations in Q2, but the company is still in turnaround mode. Revenue fell 15.2% year-over-year to $210.5 million. The Q3 forecast showed a decline in paying customers, and the stock took a hit.

These are not the numbers of a company resting on its laurels. They’re the numbers of a company that knows it needs to change — fast.

AI is part of the plan too. Bumble says it will deploy more AI tools under the hood to improve match quality. But Wolfe Herd was quick to clarify: this doesn’t mean Bumble becomes an “AI-driven experience.” The human element stays central.

A Generational Shift in Dating

This isn’t just a Bumble story. It’s a story about how young people date in 2025.

Gen Z grew up with swiping. They’re also the generation most vocal about hating it. Surveys consistently show rising fatigue with dating apps. The desire for authenticity is real, and it’s reshaping the entire industry.

Bumble’s bet is that the future belongs to IRL dating events and group experiences. Tinder’s recent moves suggest it agrees. The two biggest names in online dating are racing toward the same conclusion.

The swipe had a good run. It changed how millions of people met. But every era ends, and this one is winding down. What comes next is still under wraps — but it’s coming.

For now, Bumble is asking users to trust that the next iteration will feel more human. More intentional. More like real life.

That’s a promise, not a product. But it’s the most interesting promise in dating tech right now.

The new handshake: creators before checks

Venture capital has always been a relationship business. But the relationship used to start with a warm intro, a conference hallway, or a cold email that somehow landed in the right inbox. These days, it might start with an Instagram post.

Lightspeed Venture Partners just made that bet explicit. The firm hired Claire Zau, a seed investor with a serious following on social media, to help it reach founders who don’t move in traditional VC circles. It’s a signal that creator-led venture capital has moved from fringe experiment to mainstream strategy.

Zau isn’t a celebrity. She’s an operator-turned-investor who built her audience by actually explaining how seed deals work — the term sheets, the cap tables, the mistakes. Her followers aren’t just spectators. They’re potential founders, and Lightspeed wants them to think of her — and by extension, the firm — before anyone else.

Why trust is the new due diligence

The logic is straightforward. A founder’s first interaction with a VC firm used to be the pitch. Now, for a growing slice of the startup world, it’s a DM, a comment thread, or a saved post that gets watched three times.

That changes the power dynamic. Founders are doing their own diligence on investors, and they’re doing it publicly. They want to know who they’re dealing with before they give up a chunk of their company. A creator who’s been transparent about the ugly parts of fundraising — the rejections, the bad advice, the deals that fell apart — becomes a trusted voice in a sea of polished marketing.

Lightspeed isn’t alone in seeing this. Andreessen Horowitz acquired Erik Torenberg’s Turpentine podcast network, a move that gave it direct access to a massive audience of founders and operators. OpenAI snapped up TBPN, the tech podcast network behind some of the industry’s most-listened shows. These aren’t vanity acquisitions. They’re distribution plays.

The Turpentine and TBPN pattern

Look closer at those deals. Turpentine produces shows that founders actually listen to — not just for entertainment, but for signal. TBPN does the same on the AI and tech side. When a firm owns the channel, it owns the relationship. It gets to be in the ear of the next generation of founders months or years before a term sheet is ever discussed.

Lightspeed’s hire of Zau is a lighter-touch version of the same playbook. You don’t need to buy a network. You just need the right person who already has the audience’s trust.

What Claire Zau brings to Lightspeed

Zau’s background is a mix of operator and investor. She’s worked in product and growth roles, then moved into seed investing, where she built a reputation for being accessible and direct. Her social presence is less about hype and more about education — breaking down the mechanics of early-stage funding in a way that demystifies the process.

For Lightspeed, that’s the point. The firm has deep pockets and a long track record. What it needs is a bridge to founders who might not have a Stanford email address or a Y Combinator alum in their contacts. Zau is that bridge.

• She reaches founders where they already are — on Instagram, not just in boardrooms.
• She translates VC jargon into something actionable.
• She models the kind of transparency that younger founders expect from investors.

The creator economy comes full circle

There’s a poetic loop here. The creator economy was once the domain of consumer startups — the platforms, the monetization tools, the influencers-turned-entrepreneurs. Now the venture firms themselves are borrowing the playbook. They’re becoming creators, building audiences, and treating content as a core part of their deal flow.

It’s not hard to see why. The cost of acquiring a founder’s attention through traditional channels has gone up. Conferences are expensive, intros are scarce, and every firm is fighting for the same handful of hot deals. Content flips that equation. It compounds. A good post from a year ago is still bringing in inbound interest today.

This also changes what a “warm intro” means. In the old model, you needed a mutual connection. In the new model, a founder who has followed an investor for two years and learned from their content already has a relationship. It’s parasocial, sure. But it’s real enough to get a meeting.

What this means for founders

If you’re raising a seed round, this trend matters more than you might think. The investors who are building in public are the ones who are easier to approach. They’ve already shown you how they think. You can tailor your pitch to their interests, their pet peeves, their investment thesis — because they’ve told you, in public, on the record.

That’s a huge advantage. It levels the playing field for founders who don’t have the “right” network. You don’t need a cold intro if you can write a thoughtful comment on a post and follow up with a sharp email.

But a word of caution: creators aren’t pushovers. Zau and others like her have built their audiences by being discerning. A follower count doesn’t get you a check. It gets you a conversation. The bar for the actual deal is still high.

The risk of the creator-led model

There are downsides, and the smart firms know it. A creator-investor can become a bottleneck if every founder wants a piece of their time. There’s also the risk of style over substance — a firm that looks great on camera but doesn’t have the operational chops to help founders when things get hard.

And there’s a subtler issue: the audience itself. A large following doesn’t mean the right following. A firm that hires a creator with a million followers in the fitness space won’t help its enterprise SaaS deals. The fit has to be genuine, not just performative.

Lightspeed’s bet on Zau looks like a fit. She’s in the seed stage, she’s focused on founder education, and she operates in the same world as the startups the firm wants to back. But the proof will be in the deals — and in whether the founders she reaches actually become Lightspeed’s next big wins.

The bottom line

Creator-led venture capital isn’t a gimmick anymore. It’s a distribution strategy, a trust engine, and a filter for finding founders who don’t fit the old mold. Lightspeed’s move is a clear signal that even the most established firms see the value in having a real person, with a real audience, doing the work of building relationships before the money ever changes hands.

For founders, the takeaway is simple: the investors you follow are watching you too. And the ones who create content are the ones who are easiest to reach. Use that. It’s the closest thing to a warm intro you’ll ever get.

The Problem: Everyone’s AI Art Looks the Same

Open ChatGPT, type “a coffee cup on a wooden table,” and you’ll get a photo-realistic cup. Ask a friend to do the same, and you’ll get a nearly identical cup. That’s the dirty secret of AI image generators: they pull from the same training data, so they default to the same aesthetic.

It’s a real headache for marketers. You need visuals that match your brand, not some generic stock-style look. But here’s the good news: you can break out of that sameness with a structured approach to prompting.

The trick isn’t just adding more adjectives. It’s building a 7-pillar prompt framework that forces the AI to consider every dimension of your image—from subject to style to technical specs.

Pillar 1: Define the Subject with Specificity

Start with the obvious: what’s in the image? But go deeper than “a coffee cup.” Specify the type, the angle, the context.

• Be concrete: “a matte black ceramic espresso cup on a rustic oak table, top-down view”
• Add context: “with a small brass spoon and scattered coffee beans, soft morning light”
• Remove ambiguity: “no text, no logos, no other objects”

The more you narrow the subject, the less the AI falls back on its default mental image.

Pillar 2: Choose a Style That’s Not “Default”

AI defaults to a generic photographic look—think Instagram filter meets stock photo. To stand out, you must name a specific style.

Try art movements, photography genres, or even film types. “In the style of a 1970s Kodachrome photo,” or “mimicking the brushwork of impressionist painting,” will shift the output dramatically.

Don’t be shy about combining styles. “Minimalist product photography with a surreal twist” gives the AI a richer target than just “minimalist.”

Pillar 3: Control the Color Palette

Color is a brand signal. If your brand uses specific hex codes, you can describe them in the prompt.

Instead of “vibrant colors,” try “palette dominated by deep teal and warm coral, with high contrast.” The AI will respect named colors far better than vague ones.

You can also specify mood via color: “muted pastels for a calm feel” or “neon accents for energy.” This is where your brand guidelines become a prompt asset.

Pillar 4: Set the Lighting and Mood

Lighting changes everything. A product shot in harsh studio light feels different than one in golden-hour sun.

Use terms like “hard shadows,” “soft diffused light,” “backlit,” or “volumetric light through fog.” These aren’t just technical words—they’re creative levers.

Mood is equally powerful. “Moody and dramatic” versus “bright and airy” will give you two completely different images from the same subject.

Pillar 5: Specify Composition and Camera Angle

Don’t leave composition to chance. Tell the AI where the camera is and what’s in the frame.

• “Close-up macro shot”
• “Wide-angle from below, emphasizing height”
• “Over-the-shoulder view”
• “Symmetrical composition with the subject centered”

Even simple instructions like “rule of thirds” make the output feel more intentional. And if you want a specific aspect ratio, say it: “16:9 landscape” or “4:5 vertical for social.”

Pillar 6: Add Detail and Texture

Generic images lack texture. The fix is to describe surfaces and materials explicitly.

“Rough linen fabric,” “polished marble,” “weathered metal”—these words give the AI something to render. Don’t stop at the main subject; describe the background and props too.

One trick: add a phrase like “high detail, sharp focus on the subject, soft bokeh background” to guide depth and clarity.

Pillar 7: Use Reference Images and Multi-Model Tools

Prompts alone can only get you so far. For real brand consistency, you need reference images.

In ChatGPT, you can upload a brand logo or a past campaign image and ask it to generate new visuals in that style. The AI will mimic the composition and color scheme more reliably than any text description.

But don’t stop there. Tools like Magnific let you upscale and reimagine images with more creative control. You can take a rough AI output and refine it, add detail, or change the mood—something ChatGPT alone doesn’t allow as easily.

Putting It All Together: A Sample Prompt

Here’s what a full 7-pillar prompt looks like:

“A matte black ceramic espresso cup on a rustic oak table, top-down view, with a brass spoon and scattered coffee beans. Style: minimalist product photography with a surreal twist. Palette: deep teal and warm coral, high contrast. Lighting: soft diffused light, gentle shadows. Composition: centered, rule of thirds, 16:9 landscape. Details: rough linen napkin, polished marble surface, sharp focus, soft bokeh. No text, no logos.”

That prompt beats “a coffee cup” every single time. It’s not about being wordy—it’s about being specific.

Why This Matters for Your Brand

Generic AI images can actually hurt your brand. They signal “cheap” and “lazy,” even if you spent hours on them. By using this 7-pillar prompt framework, you create visuals that feel intentional and unique—which is exactly what your audience expects.

Start small. Pick one pillar you’ve been ignoring (probably lighting) and apply it to your next batch of images. Then add another. Within a week, you’ll notice your AI art doesn’t look like everyone else’s anymore.

For more on AI image generation for marketers, check out our other guides on brand-specific visual prompts and using reference images in AI tools.