Editor’s Note: This post originally appeared on First Round Review.
Massdrop launched in 2012 based on one idea: To use bulk buying to help people get lower prices on products they love. This idea became a platform, letting people band together across the internet to place large orders and unlock bargains on products no one needs in bulk: cameras, speakers, headphones, etc. Today, the company calls it “Community Commerce.” But Founder and CEO Steve El-Hage couldn’t have predicted how active that community would be in shaping the way Massdrop would work.
Today — with hundreds of thousands of users — the company’s look, feel, features, and future have been deeply impacted by customer feedback. El-Hage wouldn’t have it any other way. In this exclusive interview, he explains how Massdrop brought users into the product development process, the huge benefits that has yielded, and how other startups can do the same from the earliest stage.
In the last two years, Massdrop has implemented a number of features that started as user suggestions and became a reality. A lot of it has to do with listening in the right way. Below are the tactics they used to engage users in the development process, leading to a more useful, popular product.
Field Ideas Mindfully
At the beginning, Massdrop took orders from individuals wanting to buy specific products. The company told them that they’d need to gather a group of other interested customers to get the deal. Immediately, groups started springing up on email threads and online communities like Reddit and enthusiast forums. Most of them recorded group orders using unsecured, messy Google Docs. And a lot of the time, consensus and the sale fell apart.
“We knew what we wanted: A bunch of people to buy the same thing at the same time on purpose, but we didn’t know how hard that would be on the customer side,” says El-Hage. “We would comb the forums people were using — which we knew was unscalable — but we saw something important. People couldn’t get organized. A bunch of people would want San Pellegrino but others would want Perrier and it would get complicated. When orders did come in, we got to see how people structured these Google Docs to work out differences and try to make everyone happy.”
Watching users handle crowd-sourced demand prompted El-Hage and his co-founder to reach out to the leaders of groups — the influencers who were corralling users and submitting orders. “We asked them everything, starting with why they chose Google Docs. We asked them what we could build that would make the consensus-building process easier.”
“I spent a lot of time on live chat, not just for customer service, but to proactively learn from users.”
Massdrop added live chat provider Olark to its site to communicate regularly with active users. The system showed them which users were logged in and which pages they were looking at in real time. El-Hage wasn’t shy. He’d message people who had been browsing for more than five minutes and simply say: “How’s it going? I’m one of the founders. Do you have time to talk?” Nearly everyone had a lot to say.
In addition to asking about the Google Doc system they had witnessed, he would ask them a list of other, more basic experience questions: “What do you like about Massdrop? What don’t you like?” When talking to power users, he would say: “I see you’re on this page. We’re thinking about adding this feature, what do you think?”
When the Massdrop team got serious about building a group polling feature to help organize orders, they regularly communicated with influential users through live chat, lengthy email threads, and even on the phone. They wanted to understand how to nail the right functionality the first time around. Eventually, they found a successful group leader who was passionate enough to walk them through the rudimentary voting system he had built on Google Docs, and explain how they could do better.
“We took his suggestions exactly as if he had been an employee suggesting a feature,” says El-Hage. “We hashed it out together over a weekend, with this person participating every step of the way.” He would forward the entire chat log with the user to engineering so that they could go over it again and again. “From an engineering perspective we put together a very simple voting system, but it made everything immediately better. Users could create a topic like ‘soda water,’ vote to see who wanted what, then rally a group much more easily to take advantage of the discount.”
Hosting this polling process on the Massdrop site created two significant byproducts:
They were able to collect email addresses from all voters, not just the group leaders who had been submitting orders in the past; and
The system was secure, unlike Google Docs, so that no one could change other people’s votes or skew the results.
“We saw huge engagement right away. It was amazing,” says El-Hage. “That’s when our trajectory started to change. It’s also when more of our users started feeling comfortable sharing their ideas.”
As the site has scaled, so has the number of users submitting ideas through multiple communication channels. If a startup has successfully inspired its users to contribute thoughts on the product, they will probably face a similar deluge. El-Hage has advice for triaging this influx of interest.
“When we get a reasonable idea from a user, we shoot it back at them and say, ‘Hey, this is great, can you write something up? Can you spec this out a bit and give it some shape? Nine out of 10 people won’t take that extra step. And that’s okay.”
“Make it clear that you will listen to users who are thoughtful about their ideas.”
“We want to hear from people who really understand our product,” says El-Hage. “And we do. We have people come back with seven, eight, nine paragraphs that get very specific on what we should build. People have drawn things out and made mocks. That’s what we want, but you have to put a good filter in place to get it. The same funnel applies internally. If an employee suggests something, we ask them to write it up and decide based on what they come up with.”
The most critical part of any documentation around a feature are the supporting arguments, he says. “Instead of saying, ‘It would be cool if you let users create their own polls,’ we encourage users to say, ‘It would be cool because of the following four reasons,’ with each one fleshed out so we can prioritize accordingly.”
After the Massdrop team decides that a submitted feature request makes sense, they ask the person who submitted it to rally interest from their community. “We tell them, ‘Okay, we’re going to start working on this. As a power user, your job is to gauge how much customers want this kind of feature.’”
This isn’t so much about determining whether to forge ahead as it is about getting people excited. The more an upcoming feature is discussed by various groups of Massdrop buyers, the more anticipated it will become, and the more it will get used once it’s built. “We want to start off with one person invested in a feature and eventually have whole communities invested in it as well. We assign them that piece.”
To add to the momentum, the Massdrop team will occasionally leak screenshots of what’s to come by emailing them out to just a few influential users at a time. “Even if what we’re working on is just starting to work and doesn’t look perfect, we’ll give them a sneak preview,” he says. “The vast majority of the time, people send us notes saying, ‘Wow, thanks so much!’ They feel like a part of the process even if they didn’t come up with the initial idea.”
“Make sure your users know that everything you do and everything you make is for them.”
“It’s amazing, if you involve just one person in making a feature happen, they’ll want to tell everyone: ‘That’s my feature!’ She’ll want to make sure everyone uses it.”
Overcommunicate Efficiently
To make sure the company doesn’t overcommit to something people won’t use or that won’t be sustainable in the long-term, El-Hage and the team set expectations very visibly. “We’ll tell people we’re pushing something new on X day, and if it doesn’t work according to a specific set of metrics by Y day, we’ll roll it back. Interestingly enough, we’ve never rolled anything back.”
Instead of removing products they’ve launched, Massdrop has used customer feedback to modify what’s already out there. For example, after they introduced the simple polling system and gave users the opportunity to create their own polls, they got a long email from a user pointing out all he ways the voting structure was flawed.
If 10 people are trying to figure out what to eat, and all 10 pick different things with the exception of 2 who want burgers, a simple voting system dictates that everyone should get a burger. This doesn’t work when the other 8 people have no interest in having a burger. The user explained how Massdrop could fix this problem by modeling their voting system after elections in Southeast Asia.
“We went back and forth asking him to explain more, and explain simpler,” says El-Hage. “He told us, ‘Let’s say that every one of the 10 people’s second choice is spaghetti. If you have insight into people’s secondary preferences, you can help a group make a better overall decision.’ We were fascinated with what he came up with. He shared videos — including one of a mathematician on YouTube explaining how to construct this type of voting system. After doing some experimentation on our side, we told him we were going to change polls to be built this way, and he was so excited.”
It seems like common sense that startups are working to better serve and delight their users, but it’s shocking how often this message gets lost, says El-Hage. “I think a lot of founders take this one thing for granted: They don’t tell their users how central they are to their strategy. Or at least they don’t do it explicitly.”
“Something flips in your relationship with your users when they know the product is all about helping them.”
“Now all of our users know that if they can rally support and want to change something about the site, we’ll do our best to represent them and make it possible,” says El-Hage. “Amazingly, as soon as anyone steps out of line or conversation becomes less than constructive on the forums, they police themselves and say, ‘Hey, get out of here, we’re serious about this.’”
When you do find a user who is serious about co-developing a feature with your company, it’s important to give them the respect and the room that they deserve. “If someone truly has a great vision, we basically make them an unofficial mini product manager. We ask them to walk us though all their thinking. We ask them what they would do if they had more resources. We ask them what they would do after shipping the feature to optimize its capabilities. We don’t just take their proposal at face value — we tell them, ‘This is yours. You own it. You’re inventing it.’”
The Massdrop team continued to hold in-depth live chat discussions about features until they hit 100,000 users. That’s how committed they were to keeping lines of communication open with as many people as possible. When it crossed that line, they had to think about how to scale not only their service, but their culture of involving customers in product decisions.
“The amount of time we spend talking to our users is infinitely outweighed by the amazing results of those conversations.”
Find Your Heroes
Today, with the site growing rapidly, El-Hage and his team rely on a list of 10 to 15 active customers that they call “hero users.” By and large, these are people who have suggested and helped develop features that are now core to the company.
“They aggregate thoughts from much larger groups of people and send them along to us whenever it’s relevant,” he says. “I also email them on a regular basis to see what’s going on, or if they have any ideas.”
These hero users also play a major role in setting reasonable expectations for the communities they represent. El-Hage knows there’s risk in getting users hyped up over a feature and then not delivering. It’s easier to communicate challenges in their engineering pipeline and other considerations to a smaller group of people who know the company inside and out. These ambassadors can spread the message thoughtfully.
“They know how important it is to us to have a transparent process,” he says. “So we can tell them, listen, we’re working on this checkout system first because of X, Y and Z, and then we have some bugs to squash and then we’ll take on that thing you brought up. When you give people the full reasoning, they tend to understand and calm down.”
“Tell people exactly how many requests you’re getting, and never make promises you can’t keep.”
As with any engineering organization, things fall behind. Some steps always take longer than you think they will. Some products will need to be scrapped to make room for more pressing projects. The key is to make sure people aren’t left hanging. They should always know what’s going on, and having trustworthy ambassadors embedded with your users makes that much easier.
“We actually have a policy that no one from Massdrop is allowed to comment directly within the communities our heroes preside over,” El-Hage says. “We want our users to disseminate information themselves so that we’re never spamming, and we never come off as defensive or promotional. Customers are very sensitive to that these days. Our goal is to equip our most outspoken users with the tools they need to represent our side of any argument.”
Hero users hail from every corner of the internet. Some are ringleaders in forums focused on camera equipment or speakers. Others are regular Redditors or Facebook group moderators. Even though these conversations take place outside of Massdrop, the team still listens in. El-Hage recommends that startup leaders figure out all of the places where users are or may be discussing them and check in from time to time.
“No matter what kind of business you are running, there are people who feel very strongly about it — that’s a fact,” he says. “The best thing you can do is figure out who they are and then enlist them to help you make your business better. That starts with asking them why they care so much. How do they articulate it? There’s a lot to learn from that.”
Make Sacrifices to Build Trust
Like many other marketplace startups, Massdrop serves two categories of customers: the people who actually buy products off the site, and the vendors who provide the inventory they buy. Both groups have heavily influenced the shape of the service. Especially when it came to getting more vendors to work with the company, making changes to build trust became paramount.
“When we were first talking to vendors, the most common question we got was about privacy,” says El-Hage. “Many of them didn’t want people to see that they were offering the same products for different prices on Massdrop than on other sites. That could seriously impact their business. It was the number one reason suppliers didn’t want to work with us. Fewer vendors meant fewer products available to our consumers.”
So the team went back to the drawing board to come up with solutions to the privacy problem. They posed two options:
They could make every user on Massdrop create an account to connect their identity in real life to their identity on the site; and
They could de-index product pages from Google so they wouldn’t be searchable. Vendors liked both ideas; but both demanded enormous tradeoffs for Massdrop.
Requiring every user to sign up created accountability for vendors and allowed them to apply their standard e-commerce policies. Before that, they weren’t sure how to work with the platform. On the flipside, this barrier to entry would reduce the number of signups on Massdrop by a nontrivial amount. El-Hage and his team had to run a cost-benefit analysis. “We called the last five vendors that had rejected us and told them we were going to build a signup flow. All five switched to yes immediately, and we could reasonably assume others would feel the same way.”
More happy vendors meant more great products on the site, which would inevitably draw more users — that was the calculation they made. It turned out to be true, but they had to deal with some disgruntled customers in the interim.
“We responded with maximum transparency, telling our hero users how mandatory accounts would secure products that were not available on the site before — or any sites, for that matter,” says El-Hage. “They eventually got used to it, and signups rebounded.”
De-indexing content on Massdrop from Google was an even bigger decision. Anyone who saw what happened to Rap Genius when it was de-indexed earlier this year understand just how important direct traffic from the search engine is to a business. Removing product pages from search results would make it that much harder for browsers to find their way to Massdrop.
“At the same time, the more private the site became, the more comfortable vendors became, which meant a better experience for users in the long run,” says El-Hage. “We tested the waters by mentioning the idea, and our vendors were thrilled. But it did mean a pretty big traffic hit at first. Every day after we pulled the trigger, we watched traffic drop and drop for about seven days.”
Today, 95% of Massdrop’s pages are not available on Google, but the company succeeded at getting even more rare vendors and products on the site as a result. They also saw a spike in social activity about Massdrop as inventory improved. “The ratio ended up being about a 7 to 1 increase in traffic. For every one user we lost from direct search, we gained 5 to 7 new people coming in purely through social channels.” These channels included Twitter, Facebook, Google Groups, online forums, discussion boards and Reddit. Sharing went through the roof.
“We told our vendors, ‘Look, we removed everything from Google for you. We really care.’”
“Deciding to wall off the site and disappear from Google were very early moves we made influenced by vendors. In the end, it improved the entire experience for everyone across the site, so we’re happy we listened to them and were willing to take risks.”
Along these same lines, Massdrop does not require vendors to sign contracts guaranteeing inventory like almost every other commerce marketplace. “We try to keep it more friendly, more casual: ‘Look, Joe, please put aside 200 units of that one product for the group if you can. Here’s the price we agreed on. We’ll call you in a few days.’ That has sent a very strong signal that we not only want them to trust us, but we also trust them, even if it does make us unusual.”
The takeaway for other startups is that it’s generally worth it to make short-term sacrifices to gain long-term trust. Don’t get discouraged by knee-jerk reactions. A certain percentage of users will be angered by any change you make. If you stick to the course that has the best chance of improving existing users’ interactions with the site, you’re in good shape, El-Hage says.
Rely on Users to Surface Their Pain Points
Right after Massdrop had walled off its site and was beginning to see more interesting products offered by vendors, they realized that very few users were actually making purchases. It was unclear why. “The products were exactly what they wanted at good, unprecedented prices. It was a major concern and we had to dig into it,” he says.
Looking at Google Analytics, the team saw that over 50 people were simply loitering on a product page that required 50 people to buy to get the lowest discount. They just kept refreshing but didn’t take any actual actions on the page. This went on for a couple of days.
“So we went back to live chat, pinged the people we knew best and asked them what was up,” says El-Hage. “They told us that they were all waiting for other people to jump on the bandwagon so that it was guaranteed the price would drop. Essentially, 50 people were just standing there, waiting for each other to make a move. We started telling them directly over chat that other people were going to buy too so one of them should just decide to be first. Still, nobody wanted to be first.”
This got the Massdrop team thinking about an incentive structure that would get people to act, but that wouldn’t add complexity to the site. After communicating the issue to some of the more engaged users, one stepped forward proposing a “commit to buy” feature.
“He knew exactly what people were thinking — no one wants to get in early, they all want to pile on at the end to get the discount, so whatever gets built has to play into that. The result was a feature that would allow members to commit to join the group purchase at the lowest price level. It guarantees them a spot, but not before that price is reached. If a product never hits that price, they won’t be charged,” says El-Hage. “It turned out to be a much cleaner solution than anything we had in mind, and now it’s one of our most popular features on the site.”
“If a user can solve a real problem for us, and explain how, we will do it. That’s our promise.”
“We’ve gotten a lot of non-obvious answers to problems from our community,” says El-Hage. “A lot of startups don’t want to air their problems for their users to see, but if we hadn’t, we wouldn’t have arrived at the right answer. It’s worth being vulnerable to get to the right thing as fast as you can.”
Notably, Massdrop got zero pushback from users when they launched the commit to buy feature — an unusual occurrence. The team attributes the smooth deploy to the fact that it came out of real pain points that the user base could only identify from their perspective, and because the company moved so fast to course correct based on this feedback.
The most important step after implementing user advice is to thoroughly reward those who were involved. “The people who have helped us build the site and the service don’t work for free — they work for glory. They work for pride and reputation among groups that matter to them.”
“People really like helping you build your product. They like being heard. They like being creative.”
“There are so many parts of Massdrop where when we reflect, we realize we didn’t come up with them ourselves. The community came up with most of this, in fact, and we were around to enable it.”
Motivate Your Users to Build with You
So how can other startups galvanize their users to behave this way? How did Massdrop condition a fleet of helpers to get them to where they are today? According to El-Hage, any founding team can do the same thing to gain an edge.
“First, I think it’s incredibly important to take care of your earliest adopters — especially valuing their buy-in to your product development,” he says. “The only way to make it clear that this is important to you is to start doing it. Then tell everyone about it. Be vocal that users came up with features and helped make them possible. Give away credit. Don’t worry about keeping any credit for your company at all — building up your supporters is way more important.”
“It doesn’t matter if you’re sitting in a room thinking really hard all day. All that matters is that you create a great product.”
“Tell your user base stories about the heroes on your site that made things happen. That’s how you’ll get even more of them. And once you have more of them, you can build a process around it to source and filter for the very best ideas at scale. Tell your users, ‘Talk amongst each other, figure out what you need, give us all the reasons why we should do it, and our developers will be on top of it.’”
Once something is built, recognition is key. “Send it out to everyone saying how Joe came up with it and walked the team through how it should look. Joe will feel amazing, and it will send the message that this is how your company works, and that anyone can make an impact.”
As further encouragement, El-Hage makes sure that every feature proposal gets some form of response from a developer on the team no matter what. “Even if it’s a chore, it’s worth it to us to say we’ll look into something, or give a reason why we can’t. As a startup, you absolutely can’t afford to blow people off, especially if they cared enough to provide feedback.”
The biggest advantage to having customers mold your product is that you’re much more likely to tackle ambitious step changes that can make a major difference for your business, he says.
“If you wait until a product is pretty much finalized and then open it up for suggestions, you’re going to get tiny tweaks — like it would be cool if this button was blue instead of green,” he says. “If you call for blue sky proposals, you’ll get things like launching full-scale, experience-altering features. Having big structural changes come out of the community makes us much more comfortable pursuing them — it’s incredible validation.”
According to El-Hage, Massdrop will change rapidly over the next months to a year for the better because of this framework for user involvement.
“So much will be different, but we’re not worried about it even a little bit,” he says. “Most Series A or B companies spend a ton of time thinking about what they should be developing next, and they still end up rolling back features after wasting two months of time and four developers. Since we’ve had this system in place to make user ideas a reality, we’ve had zero failed features. It’s impossible to get that kind of insight and conviction just sitting in a room with the same people every day brainstorming.”