There is no right way to run your business. One of the reasons there are thousands of books on marketing and sales and managing people is that we all look to the entrepreneurial community to figure out how to make improvements. While the mistakes are helpful to learn from, it’s also important and useful to reflect on what has worked.
Here’s how we got to $4 million in sales in 2015 for Project Repat- in just three full years of being in business.
(Editing by Ross Lohr- the only way our business can work)
Before we became t-shirt quilt salesman
Breaking up with Flash Sales
There is a cyclical component to every e-commerce business. Christmas makes people shop more, while sales can drop abruptly in January. Since 2013, we have scheduled a flash sale during January and February, and it helped keep us in business. During the holiday season (4th quarter) our production partners ramp up, and their weekly output increases, but in the new year, people shop less. The flash sales helped keep the production partners humming, but there just seemed to be no way to sustain the same level as the holiday season. In February, 2015, we went back to Living Social and ran a deal for a month, and sold about 5,000 vouchers. It was good, but not a game changer. By comparison, we sold 14,000 vouchers in the Summer of 2014. It appeared that we had tapped out the flash sales.
Our groupon from Jan 2013
In recent years, there has been a general shopping trend of shoppers moving away from daily deal sites like Groupon, Living Social, Fab, and Gilt Groupe and shopping in a different way. While flash sales worked really well in the years following the financial crisis, Amazon prime has since changed the way customers shop online. When we ran our last flash sale in February, 2015, we noticed that customers had waited to buy with a second flash sale rather than buying directly from our website. This defeated the purpose of using the flash sale as a customer acquisition tool since their fees were so high. It made sense to break even and use Living Social as a discovery tool when we first launched our company, but only if people were making a second purchase directly from our website. When we asked Livingsocial if they could give us data on new customers versus repeat customers, they told us they could not.
Not only did we see a drop off in the success of the deal, but we could not accurately figure out which customers were discovering us for the first time versus using the flash sale site to make a second purchase instead of our own website. This was a major problem, and when Living Social asked to do another deal, we declined. While flash sales were a great customer acquisition tool for a bootstrapping company, we had to break up with flash sales in order to further grow our business. It’s tough giving up your biggest acquisition method, but we had to leave our crutch for walking alone so we could sprint ahead.
Facebook advertising. Agency vs in-house.
98% of our facebook content is user generated
Where were we going to find new customers without flash sales? Google did not allow us to go after the ‘blue ocean market’ — those customers who fit our customer profile but did not necessarily know they were our customer yet. Only a small percentage of Americans are searching for “t-shirt quilts” on google, and ads spent on more generic search terms (such as ‘graduation gifts’) did not give us a good return. Email marketing was working really well, but with only 10–15% of our list opening an email and less than 1% of recipients buying, the only way to grow was to increase our list size.
In 2014, we started experimenting with boosting posts on Facebook and noticed that posts with small giveaways did really well. We wrote, “Mention someone in comments section below who has too many t-shirts to win a FREE t-shirt quilt.” This gave people an incentive to share the post, and all the comments gave us more organic traffic. If a post started to do well, which meant a few dozen comments within 30 minutes, we would boost it by putting money behind the ad.
This was Facebook ads on training wheels, but while still riding around on our tricycle, the Facebook ad dashboard was approving. We were getting more interesting data after our ad spend was over, and tried a third party marketing platform called Sprucemail. It was a Software as a Service (SaaS) that gave us the ability to incorporate our email list into our Facebook ads. We could re-target our own Facebook lists, and then use the Facebook social graph to identify our target customer to grow our list. Their pitch was that only a fraction of your email list actually opens your email, but many more look at Facebook more frequently. If we set up an email campaign, Sprucemail would find our email subscribers on Facebook and show them ads.
We tried Sprucemail out and saw our sales spike. We thought that if we hired an “expert” on this — or at least someone we perceived to be an expert — we could run better ads that would increase our ROI. We canceled Sprucemail, and paid the agency a percentage of our total Facebook spend. We wanted a lot of ‘new clicks’ — or new traffic coming to our site — and then to have a strong conversion rate when re-targeting those new clicks. The agency told us they could do it so much better than everybody else, and that we were just wasting our time trying to do it ourselves.
Each day they would spend hours fine-tuning every Facebook ad, trying to show us how efficient the system they could be, and we would always hear how in a few days, the system would work even better. They brought on SEO and Facebook ads specialists and kept changing and revising the ads, but we did not see the same web traffic. We wanted to grow the audience, and cared more about casting a wider net than focusing on getting the cost per acquisition (CPA) as low as possible. Herein lies the problem with bringing on consultants. They were so concerned about using our money as efficiently as possible to show the lowest CPA that they did not focus on what we wanted them to do — drive as much web traffic as possible. They knew we were measuring them based on sales vs amount spend, and were worried that if they spent a lot on new traffic without seeing an immediate increase in sales, we would not see their value.
There was not anything particularly wrong with the agency we worked with — it’s more of a problem with the model. The Facebook ad platform is intended to be as simple as possible to use. They want small businesses without large marketing budgets to spend more money using Facebook rather than spending money to pay a consultant who will then spend that money. If you have to pay for both placing the ads and the ad, you will spend less on the ad. Thus, Facebook created a very simple ad platform product. This is great for Facebook and the small businesses who use it, but bad for the industry consultants. Once we figured out that we could run the kind of ads we wanted without an agency help, we could dedicate a lot more funds to it.
The reason we wanted as many online visitors as possible is that we knew that more traffic coming to our site meant more people seeing our retargeting ads. These were the ads that people would see in their newsfeed after they had already clicked on our site. It was like a slight nudge to come back to our site. They were thinking about the product, but maybe they just needed a coupon to actually make the purchase.
When the agency saw that our monthly sales numbers did not increase dramatically solely based on Facebook ads, they added an aggressive email campaign. We had always been frequent emailers — we know that we have a product that people find helpful, and maybe just needed some more reminders to buy. The greatest thing the agency did to boost our sales was show the effectiveness of an end of the month multiple email barrage that saw our monthly sales numbers go from $200K to $275K. The last few days of May we started earning $20K in revenue a day with an expiration date. It was mimicking what worked for the flash sales — time constraints with a discount code, and it worked really well. It was not rocket science, but just basic e-commerce marketing, and it was very effective.
Focusing on E-Mail Marketing
Our Klaviyo dashboard. We sent a lot of emails!
Over the first six months of 2015, we saw the power of Facebook as an ad platform. We paid money to target our customers, and Facebook would bring visitors who would either buy or sign up to be on our email list. Nothing else really worked as effectively. We tried adroll, pinterest, reddit, twitter, and banner ads on other sites, but nothing drove traffic like Facebook.
For the first quarter of the year, we collected about 10,000 new emails by having a pop up that would ask for your email. If you were on our site for a few seconds, a pop-up would appear. This was a standard practice, but it wasn’t until we implemented an exit pop-up that we saw our list expand dramatically.
An exit pop-up worked similar to a regular pop up, but it would not appear until someone hovered their mouse to click out of our site. At the last moment, the customer would see an offer to save 20% before they left. By the end of May, we had collected almost 10,000 emails in that one month alone, and we started to notice a pattern. If we pumped money into Facebook spend- a few grand a day, we would receive 300–600 new emails per day. Several of those people would buy right away, but most of our new subscribers needed 5–10 emails before they bought. It was a combination of re-targeting them on facebook and also sending them frequent reminders via email. Again, it was not a massive business partnership, but simple, yet effective marketing.
The flash sales were used early in our business made us addicted to the gigantic sales day for $100K, but our new formula started to lead to a lot of days earning $20–30K in revenue when we sent out an email with a deadline and a coupon code. It wasn’t the flashiest marketing efforts, but it worked. We were able to boil down the formula into: (Facebook spend) + (exit pop-ups) + (klaviyo segmented email marketing) = $700K month in November, and $300K in December — resulting in $1 million in sales during the holiday season and hitting our goal to double last year’s holiday sales. We had figured out how to drive traffic to our website and capture emails, which led to a lot more sales. More importantly, by shopping with us, our customers upcycled over a million t-shirts this year, and created almost 50 jobs in America.