Nextdoor’s Nirav Tolia thinks he can turn things around with AI; he has a lot riding on it

Social media is often about scale, but Nextdoor bet long ago on something different: that it could grow a big business off smaller, local communities.

For years, the plan worked. The 15-year-old company has long been a dominant platform for neighborhood-based conversations, connecting users for everything from lost pet alerts to local business recommendations. Then at some point, growth stalled. Users found less reason to engage with the app beyond transactional use cases. What also turned them off: misinformation, racism, and petty bickering on the platform, perpetuated by contributors who either didn’t know better or didn’t care.

Now, Nirav Tolia is on a mission to turn things around. He has a strong motive to prove himself capable. Tolia founded the company and led it until 2018, when he was reportedly ousted by the board over a disagreement about a potential acquisition. In 2021, Nextdoor went on to go public through a special purpose acquisition vehicle at a $4.3 billion valuation, but facing flattening growth and dwindling advertiser interest, the board last year asked Tolia to return.

Tolia also has a lot at stake financially. He says he’s the company’s largest individual shareholder in Nextdoor, whose market cap now hovers around $1 billion. Benchmark, one of Nextdoor’s earliest venture backers, is the company’s largest institutional shareholder. Tolia claims that “neither of us have sold since the IPO.”

Tolia says he has confidence in the company, thanks largely to the pieces he is putting in place. He’s making progress, too. Nextdoor reported 45.9 million weekly active users as of the end of September, up 13% from the same period in 2023. One of those most valuable pieces, seemingly, is the proprietary data that Nextdoor generates from the platform – data that, unlike Reddit, for example, Nextdoor has no intention of selling to an OpenAI or Google.

Still, with a long climb ahead of it, and given the challenges facing a company of Nextdoor’s comparatively small size – including the inability of larger institutional investors to get involved as shareholders – some obvious questions include how much time Nextdoor has for this turnaround, and whether it might be better off as a privately held company.

We talked with Tolia about these things, as well as his relationship with longtime board member Bill Gurley of Benchmark, in a conversation that you can hear here. Meanwhile, excerpts from that chat follow below, edited for length and clarity.

You’ve been on a bit of a media tour lately. Why?

In the first few months when I came back, the most important priority for me was to speak with people internally, but as we work internally to get our plan together, and to think about how we start to achieve our potential, it is and will be increasingly important for us to tell that story externally as well. Because one of our challenges is that Nextdoor is a company that’s been around for 14 years and has 100 million verified neighbors [registered on the platform]. So it’s not small. It is a company that we believe doesn’t have the relevance that it can have and isn’t achieving the potential that it should be achieving.

You’ve said that you’re planning to change the product.

We believe very deeply that there are so many ways our neighbors can help us. We also believe very deeply that staying connected to our local community is something that is important to keep us informed, keep us safe, and keep us smart about purchase decisions that we’re making in the local community. But we realized when we looked at the product that we didn’t have enough of the relevant information you need to stay informed . . .and when we did have that information, we didn’t always deliver it the right way. And so creating a new product – which we’re calling next internally, the “next Nextdoor” – is all about making it more powerful for users to feel like they’re connected to their local communities.

What are some of the specific transformations that you’re going to make?

Nextdoor has always been a place where UGC and neighbor created content is 99% of what you find. But we realized over time that neighbors don’t always have the broad range of information that you’re looking for, and there are lots of entities in your neighborhood, whether it’s schools or local businesses or influencers or organizations that you rely on, and those could be local news publishers as well. They all have content that should be shared with neighbors. And so one of the big initiatives [is to bring in] new content from new sources that can broaden the value proposition when you’re reading the content on Nextdoor.

I stopped using Nextdoor because I was a little alarmed by what was [being said by neighbors]. A lot of your challenge is making certain [users] aren’t going off the rails, without also being accused of suppressing free speech.

It truly is one of the big challenges that all social media platforms have. I’ll say two things. One is, we can always do better at ensuring that the mood and the sensibility of the conversation is more constructive and positive. The other piece though . . .I believe that with the advent of AI, there is going to be more technology leverage than there ever has been. And so the combination of us reminding our neighbors that this is about community and coming together and then using new and innovative AI tools…I’m very, very optimistic that…we can improve [users’ experiences] in a pretty significant way.

Would you ever use AI to dampen political conversations?

Our official policy is that we don’t allow for national political conversations. Of course, people will still bring things up every so often. One thing is, we can create dedicated groups for the political conversation – which is something that we do – because it turns out that some users actually do want to have those discussions, and in some ways they get even more angry when we say, ‘This is not a place to have the political discussion. Go have that on X [or] on a different platform.’ So part of it is shunting it into an area where it’s not part of the mainstreaming experience, but it’s still accessible. And then part of it really is . . . [applying more AI].

Just to give you a very specific example, we have something called a kindness reminder, where, before your post goes to the site, we run it through an AI technology that looks for heated language or expressions that we might describe as unconstructive. And before we allow you to post to the site, we hit you back with a message that says, ‘Hey, you may want to reframe the words that you used in these ways.’ We can do so much more on this front.

Is that AI developed in-house?

Definitely in-house. This is something I feel really strongly about. When it comes to AI, we actually have the three ingredients that you need as a technology company to really do great things. The first thing is, you need engineers, so to your point, we’ve developed it all in house. The second thing is, you need your own content. You can go buy content; licensing content is what a lot of companies do today, but it’s much better if you’re creating your own proprietary content, which is what’s happening at Nextdoor. And then the third thing is you need an audience that you can test your LLMs against, because that’s how they get smarter. And with 100 million users, we obviously have a very large fish bowl.

Would you ever license that AI?

It’s something that we get asked all the time by investors. I have a very strong point of view on this, [which is]: if you can find Nextdoor content, whether it’s Google or OpenAI or Microsoft Passport, why would you come to Nextdoor? So my personal opinion is if there’s a platform that has a proprietary database of content that is the reason their users visit them, it’s very, very, very dangerous for them to then give that to one of the places where, frankly, consumers would rather go first. I would never say never. But the strong point of view I have is: if you want the value of Nextdoor, I want you to come to Nextdoor.

You’ve known [famed VC] Bill Gurley for years. He’s on the board of Nextdoor. According to The Information, he engineered your ouster, then called you back after growth had flattened. Is that an accurate retelling?

No, it’s not accurate. But look, there was nobody around the company that was excited about its trajectory. I mean, we had gone public and our stock was down 90%. It wasn’t a Bill Gurley thing or a Nirav Tolia thing or any one individual. It was clear that we needed to do something. Bill is someone who I’ve known since 1996. I’ve worked with him since 1999 when he funded my first company. I don’t know that there’s been anyone who’s been more influential in my professional career than Bill. I think extremely highly of him. We don’t always agree, but we’ve had an amazing working relationship through three different companies, and he’s invaluable to me, and he was invaluable when I left Nextdoor and he was invaluable when we were on the board together, and none of us were happy. It wasn’t just me and Bill. It included Sarah Friar, who was CEO at the time. We all want to try to make this company great, and we’re just trying to find a way to do that.

Do you want to do it as a public company? I wonder if there’s any benefit, with all that you want to do, to taking Nextdoor private again. Have you talked to any [private equity] buyers about that?

We don’t comment on those kinds of things. But [it’s] a really interesting question, which is, theoretically, would we consider going private. Just going back to Bill, I had many conversations with Bill about this, because obviously Bill has primarily been a private company investor and I’ve primarily been a private company CEO, so that’s a comfortable place to be, because we’ve done it for many decades.

When you’re a public company, the quarterly results matter a lot. When you’re a public company, your score, which is your stock price, is something that you’re seeing in real time every single day, whereas as a private company, as long as you have a good relationship between your investors and the management, you can take long-term bets, and you can do things that may affect short term metrics without really feeling the pain of an instant external reaction at the same time.

What Bill and I talked about is, if you really want this company to be great, you need to do it as a public company. It’s kind of the equivalent of, do you want to play in the Major Leagues or the minor leagues? I mean, private companies are still minor leagues. When you eventually go public, there are more responsibilities. You are tracked more closely. You do have more skepticism around you. And there are people who care deeply about short-term performance as well as long-term, but those are all muscles you need to build anyway. I [also] think the cost of going private and then slowly but surely building the thing back up and going public eventually in the future, that would have just felt like walking in a big circle.

Who owns Nextdoor at this point? I know you had a sizable position, as did Benchmark. I saw that Cathie Woods’s Ark Investment Management recently brought up something like 4% of your Class A shares.

Yeah, absolutely. That’s all public information. So I am the largest individual investor. Bill and Benchmark are the largest institutional investors. Neither of us have sold since the IPO, so you should know that we feel very, very, very confident that the potential of Nextdoor is greater than what is seen today, and we’re working every day to make that a reality, not just for all of our shareholders, but that includes us as the two major shareholders.

Cathy is someone who we feel very, very fortunate to have taken a large position in Nextdoor, but there’s an interesting thing that’s gone on in our industry: We are a subscale company, currently at a billion dollar market cap. Everyone kind of understands the challenges of being subscale. One of them is most big investors, institutional investors – they’re not even allowed to buy positions in companies that are sub $3 billion or $4 billion in market cap. And so we want more investors, long-term investors, marquee investors, but we’ve got to do it slowly but surely because we’re subscale at the moment, and until we prove ourselves a little bit more, we won’t have that opportunity.

It just goes back to the same thing that we were talking about a few minutes ago, which is: we’ve got to prove our value, and that’s something that starts with a better product that proves more value for our users, which then allows our advertisers to have a better experience, which allows our financial metrics to be better, and ultimately, all that stuff will take care of itself. But right now, we’re in a place where pretty much everything we’re doing has to do better.