In 2015, we went from $23,989 in MRR to $63,941 (view our real-time numbers) and our team grew to 17, including me. A lot changed during this time:
- We introduced a premium plan
- We released an iOS app for Hubstaff
- We doubled down on content
- We started recording and publishing a podcast
But one thing that didn’t change was our email marketing platform, MailChimp.
We knew email needed to be a bigger part of our marketing strategy to help us reach a wider audience, get repeat visitors on our blog, and ultimately, convert more readers into customers. We had also been hearing more and more about marketing automation and how it could help us achieve our goals, and we wanted to experiment with it.
While Mailchimp is a great, user-friendly tool, I wasn’t sure it was up to the job and thus began the Sisyphean task of finding the best email platform for our needs.Find out how @Hubstaff picked a new platform for their email marketing channel Click To Tweet
How MailChimp Works
For those who aren’t familiar with it, MailChimp is centered around lists. When I joined the team, our account had 23 different lists which accounted for everything from our email course to our newsletter to lists for people who opted in on specific forms.
What made things worse was that in MailChimp users are on multiple lists, so planning out who gets what email and making sure you don’t double send it to anybody can get messy really fast.
While this system generally worked for sending out the occasional newsletter or delivering an ebook, it started to become a pain when we tried to do anything even a little more sophisticated.
For example, if we wanted to send an email to users who didn’t purchase after their free trial expired, we would need to build a list of all those customers from our database, upload that list to MailChimp and use it to create a new list or a segment of a bigger list. This “worked” on paper, but required me to constantly have to go back and forth with the engineering team to get what I needed. It also just got really confusing.
On top of this, I also knew we were leaving money on the table, but not leveraging the knowledge we had about users’ behavior between emails and automating what messages they received. MailChimp simply isn’t suited for this.
It isn’t that MailChimp cannot do any of these things, it’s that to do them we would need to dive into their robust API and adjust our app on the backend. Even having said that, after hours of research it isn’t clear exactly how we would do that. I’m a non-technical marketer, and while we have an engineering team that could help figure it out, we’re also a growing software startup, and they are already spread thin across feature requests, improvements, and new products.
While MailChimp may allow for our requirements to be met, it isn’t built to allow a non-technical marketer to do so. Simply put, we need a tool that a non-technical marketer can figure out without having to keep annoying the dev team. If you have the time, resources, and skills to adapt MailChimp to your needs, then it can be incredibly powerful. For us though, MailChimp wasn’t a good fit.
Planning our requirements for a new email provider
We knew we wanted to switch email marketing platforms, but with all of the different options out there, we also knew we needed to figure out upfront what we needed so that we could objectively weigh the options.
First, we had to decide specifically what we wanted to use the new email tool for:
Over the years, our email list has grown to nearly 35,000 names, but we haven’t done a great job of being consistent in our communication with them. We need an easy way to create and send a newsletter to our customers, and ideally want to be able to easily segment that list to send different versions of the newsletter to different types of customers.
We had experimented with these in the past but didn’t give them as much attention as we should have. We wanted to be able to not just set up an email course, but be able to easily test and track its performance so we could make sure it actually drove business results.
[Check out our email course on building a remote business.]
This is something we hadn’t experimented with, but we knew we wanted to be able to easily get our feet wet here. Being able to tag users or move them to different sequences based on their behavior within our app has the potential to make a huge impact. Even if we just set up campaigns to automatically follow-up with people who didn’t convert during the trial, we knew this was something that would be worth trying.
Our requirements weren’t anything crazy, but most importantly we wanted room to grow. If we were going to make the switch, we didn’t want to have to re-evaluate the decision the second we tried to do something even a little more complex than originally planned.
Doesn’t Require Ongoing Engineering Support
Nearly any email marketing platform is going to require some engineering support to integrate with your app, but we need a tool that doesn’t require ongoing support. I can’t go to engineering every time I want to run a test or reach a specific segment of our customer base, and it isn’t worth our engineer’s time to build a tool to let me handle this on my own when we can just find one that doesn’t need customization.
Lastly, we are a completely bootstrapped company, so while not necessarily the most important criteria in our decision, the switch needed to make financial sense. We didn’t want to pay a ton extra each month just for the possibility of being able to do cool stuff down the road. We don’t want to have to pay a big premium for this room for growth.
Our findings from testing 5 popular email marketing platforms
Based on our list of requirements, we identified 5 email marketing platforms that we wanted to thoroughly evaluate: ConvertKit, Intercom, Customer.io, Active Campaign, and Drip.
Now, before we started I knew that ConvertKit wasn’t going to be a good fit, but honestly, we were just curious to play around with a product that was basically doubling its revenue every other month.
The product seems great (I actually just switched to it on my personal blog), but if their tagline of “Email Marketing for Professional Bloggers” wasn’t enough to dissuade us, the founder Nathan Barry even said, “If you are running a SaaS application, make heavy use of the API, or need features like lead scoring, go with Drip.” ConvertKit makes it easy for bloggers to sell books and other products, but it simply isn’t geared towards SaaS companies.
This one may seem a bit strange to have on the list, but let me explain.
We currently use Intercom to manage all of our customer support, and we would love to be able to just use one platform for all of our customer communication. In the hopes that maybe it could work, I sent over our email plan and jumped on a call to discuss if and how we could make it work. The sales rep I spoke with was really helpful, but it quickly became clear that if this was going to work it would require a lot of hacky solutions and engineering resources on our end to work with their API.
For example, you can’t create an actual, cohesive email course. You have to create a separate “auto-message” for each email and hope that you defined the triggers for each message with enough foresight that they don’t cause any conflicts with the other auto-messages. You can create a folder to organize each of the auto-messages so it is easier to visualize, but they still don’t act as a unified course.
Like with ConvertKit, this wasn’t a good fit but we can’t really blame Intercom. Right now this just isn’t what they are built for, and that’s fine. BUT if anybody from Intercom is reading this, if you had better support for email marketing and basic marketing automation, then we would probably just use Intercom for everything (hint, hint).
Customer.io has the capability of doing everything we need and more, but they take a slightly different approach to how they accomplish this. They connect directly to your app via their API or via Segment and pull in all your data, but unlike an analytics platform, they don’t use it for reporting, and instead use it to allow you to create highly targeted segments based on your users behavior. We use Segment, so, at first, this seemed like a perfect fit.
It isn’t really built with content marketing in mind, but you are able to create grouped email courses and it has the features that we would need. The trouble is with how customer.io is built conceptually.
Customer.io simply pulls in all the data from our app and everything is based exclusively on that. So if somebody subscribes to our mailing list from our blog, that data isn’t sent directly to Customer.io, instead they need to be brought into our system and labeled with an attribute like “newsletter = true.” Then in customer.io, we would need to create a segment of people with the attribute “newsletter = true,” and then we would set up a campaign so that all new users to that segment receive the email course, or whatever we wanted to send them.
Additionally, Customer.io doesn’t change the attributes of users, it just reads them from our data, making marketing automation a bit of a challenge. If a user completes the email course, we can’t have customer.io “move” them to another list automatically. Instead, we would create a new segment that identifies people who have met the criteria for completing the email course, but we wouldn’t be changing their actual attributes or tags. This could get messy really quickly.
I see Customer.io as a tool for the uber technical marketer who wants complete control over every bit of data without as much concern for ease-of-use. Any type of segmentation you could want is possible with their platform, but it requires a series of filters to achieve it that is honestly a bit intimidating for somebody whose technical skills aren’t that sharp.
To be honest, I am sure I could figure a lot of this out, but it seems to be designed for somebody more technical than I am, so I think I would need a good amount of support from our development team and also would require changes to how we handle email opt-ins on our blog, making it a bit impractical.
Simply put, Active Campaign was a bit of a boost to my ego as the interface was straightforward and made me confident I wouldn’t have to annoy our engineers with 1,000 questions about how things work. To top it all off, I was also confident that they had all the features we needed. They allow custom forms to directly import opt-in information, but do not have a native “toaster” widget, but that isn’t a huge deal because there are third party plugins that work (although they are annoying, but that’s a story for another day).
The issue with Active Campaign is that while their “Basic” plan is one of the cheaper options we found, their “Plus” plan is the most expensive. While it does include a Sales CRM, the added cost doesn’t make up for it (We could use a Sales CRM like Tout for only $30/mo while the Plus plan is ~$200/mo extra). Furthermore, while the basic plan has most of the features we want, it doesn’t have lead scoring which would help us run some direct sales experiments as we grow.
It puts us in a weird spot, because the Plus plan is more than we need (and much more expensive), and while the basic plan is close we could outgrow it.
Drip is similar to Active Campaign in that it has what we need and is clearly built for our use case. It is a younger product than Active Campaign with a smaller team which means that bug fixes may take longer, but with all of my questions along they way they were very responsive and our input probably means more to the feature development.
Drip does not have nearly as many integrations as Active Campaign, but they do integrate with LeadPages, Segment, and Stripe which are the main ones we need.
They’re more expensive than Active Campaign’s Basic plan, but also more powerful. The other benefit is that they have the toaster widget as part of their package so we won’t have to mess around with third party plugins.
It was clear that for where we were at in terms of marketing sophistication and the type of business that we ran the choice came down to Active Campaign or Drip. This wasn’t a simple choice.
What about Infusionsoft?
To be honest, we didn’t really consider Infusionsoft. For starters, the pricing is more than we are looking to spend (their basic plan is $199/mo for only 2,500 contacts, we had over 25,000 at the time) and the prospect of paying a significant setup fee just isn’t appealing. While I know it can handle our fairly straightforward needs, from all the stories I have heard about it, achieving it won’t actually be straightforward. We want something that lets us get our feet wet with marketing automation and gives us some room to grow, without a steep learning curve.
Which leads us to our decision…
We ultimately made the decision to switch to Drip. While Active Campaign has a very similar offering, what swayed us in the end were the intangibles. Drip is a bootstrapped company like us, and it seems like they are the default choice of bootstrapped SaaS startups. We aren’t worried about being in the cool club, but there is a lot to be said about the benefits of using the same platform as your peers. Drip tags themselves as “lightweight marketing automation that doesn’t suck” and I couldn’t find a better phrase to identify what we were looking for.
On top of that, with Drip, we are basically getting lead scoring for free when compared to Active Campaign. While we don’t have immediate plans to put it to use, I know there is a ton of potential for growth once we take advantage of that, so Drip lets us keep our options open without needing to commit to a more expensive plan than we currently need.
In my next article of this series, I’ll talk about what the process was like for getting up and running with Drip. The migration wasn’t difficult, but there were more than a few hiccups along the way.
If you liked this post, you might want to check out the other articles in our Growth Series – we are a fully transparent startup and we share in them every step along the way of building Hubstaff.