Product Variations: How We Did It and Why They Matter

When I started Native, the first 50 customers included executives from Facebook, Casper, and Harrys. In addition to these executives, many of my friends bought a bar of Native to show support.

For the first few weeks of the business, I was focused on acquiring new customers. I had bought a 100 bars of Native, and then followed up that purchase order with another for 500 more. As a result, I had filled up my brother’s living room with sticks of deodorant, Uline boxes, and navy blue crinkle paper.

After a few weeks of focusing on new customers, I started reaching out to my original customers. I would email them under the alias “Julia Colbert” (the Native customer service team still uses this alias I think and if you ever see me in person, ask me how I came up with that name). Julia would ask every customer what they thought of Native, and how we could improve the product. The results were clear: the original formula wasn’t going to outsell Dove Deodorant anytime soon. These emails were so disheartening that eight years later, I still remember them vividly. One of my friends who purchased two bars (her initials are AJ in case she is reading this) wanted to return the products for a refund. I still think of those emails when I see her, and I make sure we split the bill if we are getting a meal.

Aside from the emails I got through Julia, I started looking at two main metrics in order to assess the quality of my product: reviews and repeat purchase rates. Reviews were great because I could ask for them quickly after you got your product. Repeat purchase rate was more important, because nothing told me whether you liked my product more than you paying for it again, but it was a lagging indicator as compared to reviews. That is, I could ask for a review within a month of delivering the product, but couldn’t ask you to purchase another bar for 2-4 months after your initial purchase.

Within the first six months of our business, I knew the world needed an aluminum-free deodorant like Native. I knew that with our existing formula, we had the chance to make a small but meaningful business and take on Tom’s. I knew that if we could develop a better formula, we would be the next Dove and that was a much bigger prize.

Creating a new deodorant formula wasn’t easy. In fact, it was super frustrating. For the first year of Native, I was the only employee and lived out of my brother’s house (follow him here). I remember telling him my dilemma, and after 30 years on Earth together, no one knows who to push me to the point of working hard without being upset like him. I remember him telling me that I had to formulate a deodorant better than Unilever and P&G. “Well, they spend several billion dollars in R&D each year, and I’ve got about $500, so I’m not sure how to do that,” I responded (he kept pushing me, and without him, Native wouldn’t have become the company that it was). At the time, I had no idea, but the reality was I had better resources than those huge companies did.

While the term is product/market fit, I think the term should be product/marketing fit. Basically, you need those elements for your ecommerce business to succeed. Early on, I knew that the marketing piece would take care of itself or rather, I could take care of it. The world needed + wanted an aluminum free deodorant that wasn’t Toms of Maine. Toms was such a piece of shit and still is. I’ll die on that hill, and I hate how they created a reputation that better-for-you products had to be grungy and ineffective. Fucking corporate robots had a three decade head start to me and not only could they not get it right, they ruined the ecosystem of aluminum free deodorants for years. Seeing people bad at business is like seeing an animal on the road that has been run over by a car. You look away and shake your head.

But in any case, I knew I had a product problem, and it wasn’t just one problem. Native’s original formula could stain your clothes. It wouldn’t rub on easily, and for men, it would pull on their armpit hair and often pull it out. It might leave little pieces of deodorant on your skin or hair in tiny balls, and no one wanted to scoop those parts out of their armpits. The problems were massive and fundamental. I knew all these problems, because I was still sending out those emails from Julia and as the only employee, I was responding to all of our customer service issues.

And so I started working with my contract manufacturer (CM) to produce a better product. The hard part is, I had no idea how to judge whether something was better or not. So I made up random ways to test the product:

A. I would send out samples to our best customers, asking them to compare a new formula against the formula they were using and tell me which one they preferred.

B. I would rub the deodorant on tshirts and then put them through the washer/dryer to see if the stains would get out. See below for a photo from 2016 of me doing this.

One day in May 2016, we realized that we had a problem with our existing deodorant formula. It was falling apart. I realized this when I went to our logistics facility in San Leandro randomly. It was about a 30 minute drive from San Francisco if you left at 7AM, and a 90 minute drive if you left at 9AM. So I got there really early for a reason I can’t remember and started looking at our products. I pulled off the cap to a bar of Eucalyptus & Mint, and it didn’t look right. I could push my finger all the way to the bottom of the tube right through the deodorant. I took another bar and the same thing happened. So I messaged our CM who randomly was in town and told her we had a problem.

Here is the exact photo I sent my CM:

We regrouped the next day at TinyCo’s offices (I lived with Suleman and worked out of his office too!). We both knew there was a problem. While we knew it was a huge problem - the cost wasn’t insane since we had just in time inventory.

We scrapped all the inventory sitting in stock. My CM called the factory that day from SF and said “stop making Native right now - we are going to make changes right away to our formula.” Any existing customers who ordered would have to wait until we launched a new formula.

I didn’t realize it at the time, but we were super lucky for a few reasons:

1. I had a great relationship with our CM, and so there wasn’t a blame game. Something was wrong and we would figure it out later. Right now, lets fix the issue.

2. We didn’t have a ton of inventory that we lost. The CM and I ended up splitting the cost on this. Another way we built trust.

3. We had already been working on a new formula that we thought was better, so we weren’t starting from scratch here. We were close with two formulas that we had experimented with - Version 7 and Version 8. Version 8 had jojoba oil while V7 did not, and so that made V8 more expensive. We went with V7.

4. Most importantly, this event made me make a change. I was still thinking about what Suleman told me. I had to make a better deodorant than P&G and Unilever. I was going to spend a long time on product development, and I wasn’t sure when we would launch an improved formula. This event forced my hand, and so by June 2016, we were selling V7.

V7 of Native was incredible. It didn’t stain + it rolled on easily + it didn’t create a mess under your arm or on the stick. Our reviews went up within the first 30 days of launching it. Our repeat purchase rate would go up from the 20-30% range to over 40% when the year was over. Simply put we had a winner.

We got really lucky with V7. Had we created something better than P&G and Unilever could create? Probably not I thought, but we did create something as good as we needed to (it turned out we did create something better than P&G could have but that story is for another day).

And what I realized then was that R&D budgets mattered a lot less than I thought. We were scrappy, and had a direct line of communication with our customers. We didn’t have red tape, or any dogma about formulas. We didn’t care if the formula would work on equipment - we were pouring our deodorants by hand. We had one goal - increase our repeat purchase rate and happiness - and that was a winning formula.

My goal with this is that you try to improve your product as well. Remember: it is product/marketing fit, and in ecommerce, we spend so much time working on the marketing that we lose sight of the fact that good products matter.

We aren’t the only company to play around with our product. Allbirds has made more than 30 changes to its shoes in order to prevent holes. Tuft & Needle took feedback from customers and made changes to their mattress foam in order to improve the product. In 2020 alone, they said they made 12 major changes to the mattress and 30 minor ones.

Native seems to have played around with its formulas more since I left. And that’s great! I think they still have the culture of using data to make products products better, and I’m proud of that!

But I still prefer V7, and I have our old CM make me my own bars!

If you enjoyed this letter, follow me on twitter here. Finally, check out NumeralHQ. This newsletter isn’t sponsored (they don’t know I’m writing this), but I am am an investor in them. They help you automate your sales tax for your ecommerce store. It is a great solution - far better than TaxJar or Avalara in my opinion. I’m biased but it is true!