Growing an AI company to 1,000,000 Users and the Art of Finding Your Own Path: A Chat with Pebblely’s Co-Founder Alfred Lua

‍Aspire: It's been about over a year since you quit your startup marketing job and plunged into entrepreneurship with a schoolmate who "rekindled your dream". How did you know it was time to shed your "golden handcuffs" and start your own business?

Alfred: Growing up, there was a lot of inspiration around me to start my own thing. My parents and sister are entrepreneurs. I also grew up in a period where companies like Facebook and Twitter got really popular. It was amazing to see the global impact of tech companies.

I tried a lot of different business ideas, but they were always on the side. A few factors came together and I thought this was the right time to give it my full attention. The first is that my co-founder, Swee Kiat, came back to Singapore because of the pandemic, and we started catching up again. He is someone I trusted, and I knew I could learn a lot from him. The other factor: I was turning thirty, and was planning to start a family. I had this window of opportunity to take risks — and in the worst case, I would find a job. I didn't think it'd be that hard.

‍

Aspire: Before Pebblely, you talked about iterating with five different ideas and being rejected by Y Combinator twice. If your entrepreneurship journey to founding Pebblely were a book, what would the major chapters be?

Alfred: Swee Kiat and I actually reconnected because I asked him to give me AI tuition! We would print research papers and review them every weekend. This was before all the cool stuff was launched — Open AI had just launched GPT-2, which is two versions before what we have now. It was pretty unknown, nobody really talked about it.

We were meeting weekly, to the point we decided, “If you're interested, let's start a company”. We quit our jobs without having a particular idea. We just knew we wanted to start a software company together.

The next big chapter was when we landed on Dashibase, a no-code tool for app-builders, after a few iterations. We had good traction from the start, and investors were interested even though we didn't really have a solid product, or much revenue. We were quite encouraged by that, so we just kept going and building. 

But after seven months, we realized it wasn’t growing as fast as we'd like, and we made the decision to kill it. We had enough runway for at least two years from our own savings to keep working on Dashibase. But we didn't want to work on something that wasn't growing fast when there could be more promising ideas.

So, back to square one. We had no idea what to do next. We were inspired by Pieter Levels, who launched 12 startups in 12 months. We thought: let's launch an idea every month, and if the idea doesn't get any traction within a month, we'll just kill it and move on. Pebblely was the first idea in that series. Things worked out pretty well. 

‍

Aspire: It seems the sixth time's the charm. Pebblely recently crossed a million sign-ups since its launch eight months ago. How did the learnings from year one shape Pebbley's genesis and growth?

Alfred Lua: One, find fast-moving water. Previously, Dashibase was in the developer tooling space, it was an exciting area with a lot of money flowing in. But, we struggled to build the right product. Then we noticed that eCommerce had been on the rise since the pandemic. Generative AI was advancing rapidly as well. We were fortunate that Swee Kiat did his master's thesis in AI image generation, and he had been following the space. He said, “It seems like we can do something about this”. So it’s crucial to identify the right market and trends. We continue to look for interesting trends as we build up Pebblely.

The second: launch quickly and iterate. When we were iterating our first five ideas, we launched each idea within a week. It was either a very basic prototype or just a landing page. We did the same with Pebblely, we built really fast and launched it. We realized that the only time we really get honest feedback is when we put our products in people's hands, people who are paying for it — and not family and friends who’re too polite to tell us it's terrible.  

The third lesson: be close to your customers. We talk to our customers every day, either on a video call or via emails, learning how they've been using the product, how they find out about us, what made them sign up, what they hate. Based on that feedback, we improve the product as we go. 

‍

Aspire: It’s also about finding fast moving waters you can swim in, right? Any learnings on finding which ideas fit your team?

Alfred: One thing we always come back to — doing what we are interested in. If we’re interested, we'll be more curious to dig in and figure things out. Problems become interesting challenges, instead of insurmountable obstacles. AI has been a big part for us, and for me, it’s marketing. Figuring out how to help businesses grow, that’s something I've been doing for over seven years.

We have this funny criteria for deciding what to focus on: “If we were to go to a party, will we be excited to tell other people about what we're working on?”  

‍

Aspire: So, how did your experience translate into building a product for marketers?

Alfred: Marketers are good at jumping into new stuff. I learned that from my experience with Buffer and ReferralCandy. We often have to prove the ROI, and it's not like with other teams, like engineering, where you might have critical technology that companies need to invest in. Marketers have to be resourceful and creative. We have to figure out things. “How do I shoot this video using my iPhone, or use certain apps to improve it?”

So we know the challenges that marketers face. How can our solution meet that demand? AI can help with the skill gap, but how do we make it easy? Recently, I was talking to a customer who said that spending two hours in Pebblely gave her two months worth of content. That's our goal, helping marketers get the content that they need, quickly and cheaply, compared to a traditional photoshoot that you might spend a few thousand dollars and a few weeks on.

‍

Aspire: We definitely resonate with that. You've discussed the hotly-contested generative AI space in other interviews. Knowing that many startups struggle from the lack of differentiation, what’s Pebbley's edge? 

Alfred Lua: When people say that AI companies don't have differentiation, I think it’s because a lot of people are building a simple UI over technology, which may or may not solve the problem that customers have. 

Pebbley is different because we do not simply build a UI over an API or an open-source model. We research the technologies and train our own AI models so that we can generate better images. We also listen closely to our customers to understand their needs. Several features were inspired by customer conversations, such as generating images with multiple products and reusing a generated background.

The other is brand differentiation. One example we talk about is Canva. There are multiple clones of Canva, but you’ve probably never heard of them before. It's not technically impossible to build a Canva, because we have seen people do it, but it's really hard to build the company and brand that they have. For us, this applies to all the marketing that we've been doing. 

‍

Aspire: It's interesting how your short-form videos are reaching a group of people who may not really know about new technologies. Can you share your insights on marketing to marketers on different channels? Especially with an AI productivity tool that’s game changing for this industry?

Alfred: If the key is to get results earlier, figure out the channels that bring you results faster. For us, we needed to validate our idea very quickly—or kill it. So we picked social media, especially TikTok, because it’s possible to go viral without having many followers. 

The right channel also depends on the product. Our product is very visual, so we're able to ‘show and tell’ on social media platforms, like TikTok. In the past, I was marketing a social media scheduling app, and then an eCommerce referral app. Besides the app’s interface, there weren’t many visuals to show.

One last thing about TikTok: there are a lot of entrepreneurs and small business owners there. And there’s TikTok Shop, where a lot of eCommerce merchants sell their products. That makes it a perfect place for us to reach them. 

With LinkedIn, we’ve built a brand around the company, not just the product. It’s brought in a lot of partnership opportunities, such as our partnership with HubSpot.

Something that we should have deprioritized was SEO. The technology is so new that people aren’t actively searching for it. We could have targeted adjacent keywords, like “How do I take product photos?” But it didn't feel like the right thing to have spent time on, given that the idea was invalidated, and we didn’t even know if the company was going to be around in six months. Investing in SEO was a waste of time at that point. 

‍

Aspire: Whatever you’re doing seems to be working, because Pebbley has been rapidly gaining traction worldwide. It’s even been featured as a tool to learn by marketing leaders like Hubspot. Any tips for founders wanting to launch SaaS solutions globally?

Alfred: I repeat this a lot. It’s important to understand your customers. A lot that we've done wouldn't have been possible if we didn't talk to our customers.

When we started going viral at the start, we noticed that most of our users are using our product on mobile. There were a lot of people in Africa, Brazil, and Indonesia using our product, they’re just not sitting at a computer. 

So we quickly made Pebblely mobile-friendly and created content to teach people how to generate AI product photos on mobile. Then, we saw more people sharing about Pebblely on social media.

As entrepreneurs, I think it's just fun to talk to someone using your product. We set up several automated emails to reach out whenever customers hit certain milestones when using Pebblely, so that they'll schedule a call with us. 

Stay close to your customers. You’ll learn a lot of product and marketing insights from them.

‍

Aspire: Moving forward, what's the game plan for Pebblely? Are you gearing up for your third Y Combinator attempt, or something else?

Alfred: We applied to YC at the start of this year. That was when Pebblely gained a lot of traction, but we didn't get the interview. 

Thankfully, Pebblely has grown pretty well. For now, we're deciding to stay bootstrapped. We want to enjoy every day as we build a great product. We don't want to rush. We don't want to raise money for the glamour around successful fundraising. We have enough to last for a while. 

We also want to keep the team lean for as long as possible. We're not planning to grow the company for the sake of having a big office or a big team. A small team allows us to move fast and adapt to trends. That should also help us keep close to customers as we build our product.

For now, we're not thinking much about the managerial part of building a business. We’re optimizing for what we enjoy daily. It's cliche, but if I don't enjoy my day-to-day life now, I might as well work at a big company and make a lot of money. We quit our jobs because this is something we enjoy more.

The dream might not be an end goal, but an end state. If we can keep doing this, building something great that people want to use, that’s it—that’s our dream.

‍

Aspire: How do you find that joy amidst all the hustle?

Alfred: This is hard. It’s important to have courage to find your own path. Starting a company is quite a lonely journey. To a certain extent, you're always looking around to see what others are doing and searching for validation.

One thing my co-founder has taught me is to think from first principles. It's a lot easier to think by analogy. If ten startups are doing it, we would feel compelled to follow. It doesn't help that most media will cover companies that raise money. But raising money isn't the only way to build a successful business.

It can be dangerous to rely too much on an outsider's perspective, rather than what you truly think. 

We fell into that last year when Dashibase had good early traction and several investors were interested. We started listening to the investors more than ourselves. “I'm sure they're smart people and they saw that this space has a huge opportunity, maybe we should spend another month building this.” But we had the numbers internally, we knew it wasn’t going the direction we wanted. Eventually we pivoted, and probably lost investment opportunities because of it.

Find out what you're doing this for, and have the courage to double down. 

For us, it is enjoying every day and building things that people love to use.

Learn more about how your business can create beautiful product photos in seconds at pebblely.com