Listen on Apple Podcasts
AO: So what’s one unconventional growth tactic you’ve run that worked surprisingly well?
Darius Contractor: So when I was working at Dropbox, one thing we were trying to do is activate Dropbox business users and get them up and going with Dropbox. And so for us activation meant having them upload files to share with their other team members in Dropbox and then also install the Dropbox client and also invite other team members to Dropbox and have them join in and, shared files themselves. So, this is kind of our goal and the hack is in that context. So, what we did is actually looked at the users who were upgrading and realized that many of them were actually normal Dropbox users.
When I say normal kind of like individual, like maybe free or a pro account Dropbox users that were then upgrading to Dropbox business, which is now a team platform. And one thing we did is when they did that upgrade, we actually just kept all their files in a private directory, because we didn’t know which they wanted to share with the larger group. So one realization I have is that there’s probably a lot of people who are upgrading, planning to share all the files or most of them that they have in their current Dropbox with their team. they might expect that that’s the behavior. Now we were hesitant about defaulting that behavior because of the people that were expecting the opposite behavior of their files to be private, they would be very unhappy if all their files suddenly went public.
But what we realized when we took a look at those users, it wasn’t super obvious that you actually needed to share those files out. People had different expectations for what the default behavior was. So what we did is something that I call a minority report, where we looked for large but not majority cases that are worthy of localized optimization. And so often it is people like this, where you have this very high intent user group that’s coming from another service or coming from another part in your service that could be super served by either migration flow or something that speaks directly to their needs.
However, it isn’t the majority of cases. It isn’t every case. And because it’s not every case, oftentimes teams don’t focus on optimizing it. Plenty of people just signed up brand new for Dropbox business, from the web or they just put in their company email and made a brand new account. And so it wasn’t something that we immediately thought about as far as optimizing this kind of migration flow from a Dropbox personal account to a Dropbox business account. But as we dug in, we found a lot of people who we’re doing this, upgrading from one to the other, and then many of those people had files and many of them weren’t sharing those files.
So what that led us to do is put as part of the upgrade flow, for the people who had files in their personal account and were upgrading them, a simple chooser of which folders in their personal account do they want to share with everyone in their new business account. And so if this was a business account already, and it had invoices, customer information, designs and mocks, they might just check everything off and hit next. And then boom everything they previously were using Dropbox for in a work context, probably on their personal, would just be shared with work or by default they would be unchecked.
And you could say, no, I don’t wanna share this. And they could hit Next. And what we found was that a very high percentage, maybe somewhere around half of people would actually check one or more boxes. So that was a lot of files that were going from private to shared within the team context. Of course, all these are private to the world, they’re only shared within the team. And so what that gave us is a double digit increase in the percentage of activated users based on this activation metric we had of shared files, etc. So, it was surprisingly a big win even though it was localized to this one set of users.
AO: Very interesting. What is the strategy behind this that you could share in terms of how you got there? Because I imagine this would be easy to overlook.
Darius Contractor: Yeah. And as I was bringing up earlier a little bit, I think we tend to, in growth, optimize that every man case. Optimize pretty much a blank start that someone just coming to this thing for the first time, and they’re kind of clicking through the obvious flow and they may be inviting people from scratch, et cetera, et cetera. But I think what we overlook sometimes is the very high intent, maybe subset of those users, that we could give a particular experience that would then massively accelerate them in such a way that the effect would actually be seen globally in the numbers.
Because obviously you don’t want to micro-optimize a small segment and have a quarter percent increase. It doesn’t make sense. Another example of this is Gmail users. If someone comes in and says, I have a Gmail email address, you might say, okay, great. Do you want to import your contacts? That’ll make it easier to share. Whereas here they come in with some other services you can import contacts from or some arbitrary email, you might not give them that example. You might not give them that prompt because you don’t have an easy way of importing their contacts. So there’s a number of these where you can find this kind of large, but not every user and kind of super charge them and get wins that are big enough you see them for the whole site. Cause if you can take half the users and increase them by 20%, you’ve increased the whole site by 10.
AO: Very neat. How are you prioritizing these groups? Is there any framework that we can apply? Because the big one everyone’s focusing on, but sometimes we have these other segments. Is there any prioritization that you could share?
Darius Contractor: One way to look at it for a number of different cuts you can do on the data. You can cut it by a browser type. You can cut up by email type. You can cut up the size of the company. Then look at those first, maybe the four biggest batches in your pie chart, so to speak, because it’s not going to be like 5% sliced for the most part, it’s going to be a relatively large batch usually. And then maybe layer on top of that a multiplier of how high intent or activatable are these people. For instance, if you run some service that does catering or something, maybe you have some of your customers come in brand new.
They’d never gotten catering before and other of your customers are transferring from a very popular catering service. So they might already have menus. They might already have expectations. They might already have a schedule with that existing catering service. So could it be that if that’s 30% or 40% of your customers, you can super serve them by saying, Hey, just upload your menu or your approach from this other service. And we’ll take care of everything. And that can increase activations for that subgroup by like 30%, if that’s 30% of your users, that’s a 10% macro win and that’s worth it.
So I cut it by these different companies or slices and think to yourself more or less, how accelerated is this group or how high intent they are? How much does their background, especially technologically accelerate them? Like the Gmail user case, where you can actually do an integration with email contacts, as you can do this one off technical acceleration, that actually has macro ongoing gains.
AO: Very neat. So it sounds like you’re really doing a lot to understand what’s happening before. You’re understanding the before, so that you can know the velocity of what they’re going to be doing. Are there any research techniques that you’ve employed, qualitative interviews, surveys to help learn about this segment that you’ve mentioned at Dropbox?
Darius Contractor: Yeah, you can do that as well. I always recommend qualitative just to see how people are using the site. Definitely emailing 100 users, and talking to like 10 to 20 of them. One good way to do it, is to do a diff of your expectations versus their reality. If you expect them to take these 10 actions, you can put together an interview asking if they took each one of those 10 actions and it’s probably one or two of those actions that very few people took because they missed it or they didn’t think it was important. And that can make it easier to reassess how you do onboarding or like anything it can be onboarding, it could be upgraded, it could be churn, just kind of deep diving on how the customer thinks about it.
AO: Oh, that’s a great idea. Could you give me a little bit more on the diff, if I wanted to go do it right now?
Darius Contractor: Yeah. For instance, say with Airtable you might have a simplistic level expect people to come in and then first want to learn how your table works. And then secondarily want to learn what parts of their business to apply it for. Next, try to find the best template as a starting place. Then find a way to put in their data, so they can start to onboard and use our table for their particular use case. But then as you talk to people, you might find that no, they actually show up with a very specific use case and they’re just like, how do I do this one thing? And they’re not actually in that exploratory mode, I’m giving a very specific example.
The reality of it is that you have a range of users who are in different modes and some are exploratory and some are very high intent and specific. But as you talk to 10 or 20 of them, you find that, okay, maybe there’s a segment of these users that I haven’t been thinking about as much that you could super-serve. So that’s the kind of it, I also find a lot of value in looking through the data. It does take time, but oftentimes like the kind of questions I was talking about, you can validate with data. You know, if you think that, Oh, yeah users are going to explore, you can look at how many users actually clicked around and looked at the different parts of the site before making a purchase decision.
And you might find it’s fewer than you think. This is a conception you have as a product owner about what people do, but it’s not reflected in the data at scale. So I find the data is very good at kind of invalidating my assumptions. One of my mottos for growth is that the team is always doing one of two things, they’re learning or succeeding. And there was a lot of learning.
AO: Man it’s humbling, isn’t it?
Darius Contractor: It is.
AO: I appreciate that. So what are some roadblocks that you think that someone’s going to run into as they’re applying this approach of looking at other segments and not just the obvious one to improve?
Darius Contractor: The biggest challenge often is getting enough data that you can meaningfully talk about segments or cut it up in ways that make sense. And it’s not just getting that first cut of data. It’s additionally getting the cuts of data related to actions they took or downstream results, and conversion or what have you. And it does take a lot of time. You really have to sift through the data and really organize it in a format where you can run queries on it for some data engineering first. So, I think that’s the biggest blocker really doing growth at a lot of companies. There is also a question of scale, you do need enough people that you have a growth scale data set.
So you at least thousands, or preferably tens of thousands of people before you even start to make any conclusions with date. Ideally you have a social scale dataset in the millions. But it depends on company stage. If you are a very early company, you’re better off simply talking to your customers directly and not over-indexing on a very small end of data.
AO: This has been amazing, thank you so much for your time. That’s all I got.
Darius Contractor: Yeah. It was a lot of fun. I am glad to share this moment with you and excited for any questions or feedback from your listeners.
AO: Boom. That’s it. Another great episode of The One Growth Show , the official podcast of growth marketing conference to learn more about upcoming events, visit www.growthmarketingconf.com and subscribe to the newsletter. If you enjoy this episode, let us know. We’d really appreciate it if you’d give us a five star rating, super easy, just click the last star on iTunes, and also share this episode on social media. After all you want your network to know you’re the person they can always turn to for the best growth and marketing content, don’t you?
Follow GMC everywhere: https://linktr.ee/GrowthMarketingConference