I just turned off the Bevy site and I thought this would be a good time to reflect on what worked and what didn’t.
Too many people try to hide their startup failures. I wish more founders would write down their learnings for others to enjoy.
Hopefully, some future founders will find my words helpful.
Background
For those unfamiliar with Bevy, here’s my description from a previous interview…
Bevy provides everything you need, but don’t want to own, for amazing experiences. We curate the best and easiest to use equipment in our online store, bring it right to your door when you need it, and take it back when you’re done. We actually own all the equipment so we can make sure that everything we send out is in great shape.
If you want to go camping this weekend you don’t have to spend $1,000 on camping gear. We will drop off the best tent, the best grill, the best sleeping bags, and cool things you wouldn’t even think of like headlamps and marshmallow roasting sticks for the ultimate camping experience. When you’re done we will take it all back.
Then the next weekend if you are having a friend over we will bring you a really nice inflatable mattress, towels, and linens to give your guest the hotel experience. And then the next weekend if you want to go to the beach we will bring you everything you need to have an amazing time there. All these products are curated into bundles for experiences to make it as easy as possible.
We are enabling people to live as if they owned all this specialized equipment and do amazing activities that weren’t really an option before.
I deeply believe in the macro trends around the business and believe someone will solve this at some point. Urbanization, access over ownership, and experiences over things are all here to stay.
How exactly to tackle the problem for rarely used and specialized items isn’t quite clear, but I have some ideas for how or when it could work.
Reasons for Failure
Asset Intensity — This is the biggest point of failure. There is a reason that high margin software businesses are so appealing and many investors won’t invest in any other business model. This was never a business that was going to be shooting off profits early on.
Raising money to prove out the unit economics was really the only route, but the business model risk made raising money early on difficult. If you are the former founder of Klout, like Joe Fernandez is at Joymode (also failed) you can raise a few million dollars with an idea and work to prove it out. Barring that, it is very difficult to get meaningful utilization numbers and prove out a business like this on a small scale.
Too Broad — We intentionally launched with a broader aim because we wanted to sell a lifestyle to repeat users instead of a once-a-year outdoor gear rental company. Proving out payback periods across many different categories is even more difficult. Focusing on one vertical and expanding would be an easier business to prove out.
We also never saw the repeat rate we expected. When we would pick up the gear, we would talk with the customers about how everything went, what else they are interested in, and what we could do better. They would have a great experience and would list 4–5 other offerings that they couldn’t wait to do, but they would rarely follow through on it.
Our plan was to move to a subscription model in time, requiring memberships or offering all-you-can-eat, but we never made it to that point.
Operational Intensity—We initially offered delivery in downtown Chicago which was a massive hassle. We limited delivery to one 3 hour window to create density which helped, but coordinating with many different customers was difficult.
Outsourcing delivery was also cost-prohibitive. It costs roughly 2x to outsource delivery compared to doing it ourselves. We eventually removed delivery and many people were happy to pick up items from the back porch during the day.
Moving things around in the physical world is always more challenging than shipping code. Most E-commerce plays can outsource to countless distribution centers, but we didn't have that luxury.
Market Timing—It isn’t clear to me that people are quite ready for access over ownership for smaller household goods. Some forward-thinking customers loved it, but many were surprisingly still reluctant to spend on experiences over things. People are comfortable renting outdoor gear, but folks seemed skeptical of most of our other offerings.
Customer Acquisition
What it is all about. Roughly in order from most to least effective.
Google Ads — By far our most effective channel. In Chicago, there are a lot of people googling “camping gear rentals” and no one offering it (REI only rents gear in ~10 states). These customers already had intent and converted at a high rate.
Content — Launching out of the University of Chicago, we had access to a lot of smart interns. The most successful use of interns was giving them an outline of a blog post and letting them flesh it out.
They would find 3–5 other posts on the topics and produce an ultimate guide with beautiful photos. Google loves long-form, authoritative content. It takes time to rank highly, but it can be a great enduring source of low-cost, high converting traffic. The interns would then walk away with a concrete example of work that drove X emails and X conversions.
Community—We had some success leveraging groups we already belonged to like neighborhood groups on Facebook and school groups through business school. If I were to start another e-commerce business, I would begin with content and community to drive low customer acquisition costs and dedicated customers.
Referrals — Incentivizing word of mouth was cheap and effective. There are some good plugins on Shopify to automate this and can auto-deny fake referrals, which quickly come up.
Press—With my podcast press connections, we quickly got some decent pieces from Chicago Inno, Built In Chicago, Voyage, and Dante32. This was nice social proof and led to some orders, although I am not quite sure how many. Press doesn’t take much time. It is basically just content that someone else is creating for you. When something actually shows up when your company is searched online, conversion improves too.
Events — We ran some all-you-can-drink beer pong tournaments that were a ton of fun, got some emails, and actually made a few hundred dollars each time. We only had one conversion from the hundreds of people that attended though. It really wasn’t worth the time it took to plan and setup. We also ran lobby events, which were not much more successful. We quickly stopped doing in person events.
Facebook Ads — They are expensive and we were never willing to spend enough to make them work. Facebook ads are great for figuring out what messaging resonates, but be wary of being dependant on them for customer acquisition. You want your own cheaper channels in the long run.
The Future
3rd Party Supply — My plan was always to move to a marketplace model once we got the business off the ground. Launching with our own supply allowed us to focus on building demand initially.
Most big marketplaces follow a similar route. Uber launched with their own black car fleet. Bird scooter launched with their own fleet before empowering small business owners to buy their own fleets. Omni has partnered with local rental shops to bring their inventory on to their platform, effectively acting as the backend software and lead gen for small rental businesses.
If you can convince existing players that your offering is worth it then this is by far the best approach.
P2P model — This is logistically tough to make work. It is basically double the work of selling something on Craigslist or Facebook marketplace, which itself can be tiresome. For rentals, you have request, approval, emails, texts, pickup, and then have to do it all over for the return.
This works for exceptionally expensive items like cameras or DJ gear where the average order value is high enough to put up with the friction for both parties involved. Fat Llama had early success building liquidity with DJs in London and now with high end electronics and campers. There is a huge graveyard of failed startups in this space.
Delivery — In the near future, self-driving delivery vehicles will drop the cost of deliveries by ~80%. This makes local delivery of small goods feasible.
The mailed package model of Arrive and others is intriguing as well. Seasonality and utilization are less of an issue when you can service the whole country.
Drop Spots — Another model that we dabbled in was using lockers for drop spots and holding inventory. Many high-rises now have lockers for package deliveries. These can easily be used for item pickup and return. We have proved that people are happy to save money and pick up locally so I suspect they would love to pick up from lockers at their convenience.
What’s Next for Me?
I have shifted to the other side of the table (venture capital), which I’ll talk more about soon. It has been great helping out and learning from entrepreneurs building all different kinds of businesses instead of being laser focused on one. I’m sure I’ll get back to operating in time.
If I can leave you with one thing, don’t be afraid to take risks and go for it.
There is no better way to learn than by doing.
Whether it works or not, you will learn a ton.
If you are building something cool, are raising money, or just want to chat you can find me at @ColinKeeley or colin at colinkeeley dot com.
Thanks again for all the Bevy team members that helped out along the way, Polsky at the University of Chicago, and, of course, all our customers that believed in the mission.
Own less, live more.
Colin