Have you ever scrolled through a “best of” list and wondered, Did anyone actually try these things?
Nine times out of ten, the answer is no.
Recently, I looked at who was ranking for a handful of “best product” searches in a category I know well. One review site was run by a company selling a direct competitor to the product it was reviewing. That detail was tucked into a disclosure most people would probably never read.
Another site said right in its own copy that it hadn’t personally tested a single item on the list. The rankings were “research-based.”
As a shopper I totally get how easy it is to miss these things. I’ve been fooled myself. That experience led me to write about how to tell when a product—or the hype around it—is fake or AI-generated.
It’s something I keep coming back to as AI becomes a bigger part of how we find and buy things: looking trustworthy has become remarkably easy. Earning that trust still takes work.
An algorithm knows whether a page is optimized. That doesn’t tell us whether the product is good. AI can put together words that sound honest. That doesn’t tell us whether the founder behind them is.
For those of us trying to find something worth buying, that’s a pretty meaningful gap. And I think it makes human curation more valuable than ever.
Grommet’s been doing this since day one
Grommet launched in 2008, just weeks after the financial crash, with a simple idea at its heart: somebody should personally vouch for what’s being sold online. By finding great products and sharing the stories behind them, we could help Makers get discovered and help Shoppers find something worth bringing home. Grommet was born–thanks to co-founders Jules Pieri and Joanne Domeniconi.
Five days after launch, a French blogger reviewing the site described the bigger question as choice vs. relevance. Do you want hundreds of options to sort through yourself? Or do you want to trust someone who has already done that work for you?
Eighteen years later, that question still gets to the heart of what we do.
I joined the team in 2009. I can attest that testing a product and sharing honest feedback was a job requirement for everyone, across all parts of the business. Since 2023 I’ve personally touched, tested, and evaluated every product we’ve launched. For me, that means a promise that I’ll get beyond the spec sheet and listing photos and actually spend time with the thing in my real world.
It’s a promise I take personally.
Every weekday, our team decides which products we feel good about sharing with our shopper community. Along the way, we also make plenty of decisions about what isn’t ready for that introduction.
My garage is proof: not everything makes the cut
If you’d like a behind-the-scenes look at that process, my garage would be a good place to start.
It’s full of products that have launched and some that never made it to launch. Just ask my husband, who can’t park the car in there anymore.

A pile of this week’s deliveries in my garage to be tested and possibly launched on Grommet.
Some of those products are genuinely great and get a spot on our launch calendar. Some are things I personally liked, but they didn’t quite meet our bar, weren’t the right fit, or needed more work before they were ready.
That’s one of the hardest parts of this job. We want to help every Maker get discovered. We also know we can’t help them all.
We consider thousands of submissions each year. Historically, only about 3% make it to launch.
Grommet has a different job than Amazon. We’re here to narrow the choices—to help you find something we’d feel comfortable recommending ourselves, because it cleared every standard we have.
That means saying no a lot more often than we say yes.
Try calling a big-box retailer with a real question
Try calling a big-box retailer with a detailed question about a product sometime.
Does this material off-gas? Will it hold up for the way you plan to use it? How long does it actually last with daily use?
Once you get beyond what’s printed on the website, you’ll likely reach the end of what the person on the phone can tell you. They haven’t touched the product. They’re working from the same information you are.
Reach out to Grommet with a question like that, and you get me. Yes, really me!
If I don’t know the answer, I’ll go find the product (there’s a decent chance it’s still in my garage) or call the Maker directly. Sometimes I’ll connect you with the Maker so you can ask your questions yourself.
You don’t need a special service tier for that. It’s just how we work.
We know these products. We know the people who created them. And we care about what happens after the package arrives at your house, when you actually start using what we recommended.
Your friend in the know — features are earned, never bought
Grommet’s always acted as your “friend in the know”. You can see that in our early product videos. Our team would tell you about a product, explain why we selected it, and share what we liked about it. And often, who it’s NOT for. (Radical? Maybe. But honest.)

One of the emails sent to our 3 million subscribers personally recommending a gift I love to give, Sayso Cocktail Mixers.
There’s something familiar about that kind of recommendation. It’s the same impulse that makes you text a friend and say, “You have to try this.”
That’s the responsibility we take on when we feature a product. We’re putting our recommendation behind it.
And here’s a detail that feels especially relevant today: Makers cannot buy a listing on Grommet. There’s no paid placement. The right keywords or a bigger ad budget won’t get you a feature.
A product earns its place on its own merits, in front of a team that has actually used it.
From there, our feed is 100% driven by Shopper support. When someone in the Grommet community upvotes or purchases a product, that creates a trust signal and helps move it up in the feed.
That’s the only way it happens. Our Shoppers decide which discoveries rise.

A screenshot of Grommet’s Trending Feed which changes in real time based on real, verified Shopper input
Human curation vs AI: Why this matters more than ever
AI and algorithms will keep getting better at sounding trustworthy. As they do, I think we’ll care even more about who has actually done the work behind a recommendation.
Who tested the product? Who was willing to pass on something that didn’t measure up? Who took the time to answer a hard question honestly?
Those are the people and companies I want to hear from when I’m deciding what to buy.
I’ve spent close to two decades doing this work, and I can tell you it asks more of us than an algorithm can do. It takes time, judgment, and a willingness to make difficult calls.
It’s worth it. And the fact that Grommet is thriving 18 years after our first launch tells me it still matters to shoppers, too.
That’s why I still believe in human curation. The more our internet fills with automated recommendations, the more I appreciate being able to say, “I’ve tried this. I know the person who made it. Here’s why I think you’ll like it.” If you’re into that, be sure to sign up for our email at thegrommet.com. I personally write every email we send (and though biased), I think you’ll be into them.
If you’re a Maker wondering how this applies to launching a product today, I wrote more about that here.
