Webinar

How Drift Doubled its Conversion Rate by Personalizing its Onboarding Experience

Drift improved conversion rates by 265% on its homepage by personalizing its onboarding experience using lead scoring data. 

Drift is the leading conversational marketing platform that is changing the way companies interact and communicate with their prospects and customers. The growth team at Drift has more than doubled conversion rates on its homepage generating more meetings and account sign ups.

As part of its commitment to its customers’ experience, Drift uses Intellimize to personalize its onboarding experience for each prospective customer based on a lead score from MadKudu. Drift has created content and tools to enable prospects with lower lead scores to onboard themselves efficiently and quickly. This allows the company to focus its sales team’s time on prospects with the highest lead scores. In this webinar, Sara Pion, Growth Marketing Specialist for Drift, describes their approach with examples, results, and lessons learned.

Webinar Transcript:

Introduction

First things first. Hi, I’m Sara. My current title is Growth Marketing Specialist at Drift. And I was previously a Conversational Marketing Specialist, and before that I was a Customer Advocate. So I’ve been at Drift for about a year and nine months, and I’ve been working on the growth team since late August of 2018. So I’ve been on the team for a few months and we focus on touchless activation, which includes free user signups and activation, as well as demand gen, and we own our own lead segments. I work on both teams, but today I’m talking about how we approach onboarding for our free users.

Today I’ll show you how we use lead scoring to create different onboarding experiences for our free signups. So before someone even signs up for Drift, their onboarding starts when they come onto our website. We already want to create a custom experience for those people. We create website variations based on MadKudu score. We leverage MadKudu and Intellimize as two different tools to help us perform these fun experiments.

The Technology Drift Uses

MadKudu is a lead scoring tool that leverages AI and customer data to score your leads in a dynamic and ever-changing way based on their firmographic information, but also their behavior inside your application and on your website to really help you and your sales reps understand how “good” your leads are.

Intellimize is a platform that allows you to create website variations and target them based on a lot of different criteria. We use Clearbit to help us understand the visitor’s company and then we target based on MadKudu score. We create a lot of different website variations for the different kinds of traffic that we have coming onto our website. So, we have created a bunch of Intellimize variations on three super high converting pages on our website, to kind of help leverage the pages where we want people to sign up versus chat with us.

Personalization Examples

On the right, you’ll see our Simple Ass Homepage variation versus the control which is on the left. On our homepage we really want our bad leads to create a free account first. We want to create a more touchless experience and have them go right on forward with creating a free account. With our good leads, we want to leverage a conversation before we push to a free account. And the reason why we do that is because for us, meetings booked generally helps with higher MRR than a free account converting to a paid account. So, for good leads, we try and leverage meetings, for bad leads we try and leverage free sign ups almost immediately. And the Simple Ass Homepage has a 1.79 conversion rate, versus the control which converts at a 0.49 conversion rate. So we’re actually doing better at getting the people that we want to sign up to sign up with this variation.

We also leverage headline changes to give people a little bit more information on what they’re actually signing up for and what Drift can actually do for them, so we want to give them a little bit more information on why they should sign up for Drift and add a little bit of social proof in there. So 150,000-plus businesses are using Drift to connect with their customers now. See how Drift tryout works on your site right now. And we also like to play around with the hero image and the rolling logos that we show, to see if we can try and drive some more signups and some more signups from that social proof.

So, we then have our powered by page, and the way that you get to drift.com/powered-by is through our widget referral. We have done a lot of work on creating viral loops through the Drift branding that shows on our chat widget, and when you click on that chat, powered by Drift hyperlink, you actually come to our powered by page. And about 18% of our signups happen on our powered by page, so we’re driving a lot of freemium traffic to the power by page that then drives freemium signups. It’s a fun little viral loop that we’ve worked really hard on in 2018 for sure. These powered by variations convert way higher than the control, and we think the reason why is the control is a super long form page, really text heavy, and it talks a lot about what conversational marketing is. Versus these power by variations, which are a single split screen, there’s nothing below the fold. We do a product highlight, so we’ll highlight our bot or our meetings tool or our Drift profiles image or a tool, or we’ll do a social proof.

We’ll highlight a customer that’s having a lot of success using Drift. And we also leverage the website that they came from. Do you want the experience you just saw on Drift.com on your own site? Get started with drift. It’s totally free. Create a free account now. So we kind of give them enough information and give them enough leverage to then feel the need to create a free account. And these perform very well.

And then we have our pricing page, which obviously has a higher converting page as well. This is our top variation right now. It’s actually being shown to about 80% of our traffic versus the control, which is sent to about 20% of the traffic. And the reason why that’s happening versus a 50/50 is Intellimize actually does a Bayesian split. So instead of just continuing to send traffic to 50% of your variations versus the other, once your variation starts outperforming the control, that will be shown to more and more traffic to try and drive more conversions. So if you go to drift.com/pricing chances are you will see this variation and not the control, which is what we want because it converts at a 7% for signups against 3.46% of traffic that signs up on the control. I didn’t even put the control image on this page. It’s honestly not very pretty. This one is definitely better and we’re really excited to see that it does also aid in conversions.

Making Intent Part of Onboarding

So awesome. High five. Someone signed up for a free account, now what do we do? So we make intent part of onboarding. This is so that we can ask them what they’re interested in using Drift for so that we can create a more custom experience along the way. So this is the final step of the onboarding process before you install Drift. And we really wanted to know if people were looking at these, checking all that apply, or if they were just checking random ones.

We tested it and we switched around the order of things and we saw that people were actually looking at the options and checking all that did apply. We found that the top two reasons people were signing up for our free plan was talk to leads and live chat for support, which is perfect. That’s what we’re made for. We’re super excited. We were very interested to see that chat bots was lower down in the … Like number three as part of the intent rather than number one or number two, which makes sense because our free plan is just live chat, it doesn’t include bots. So it kind of helped with some of the inferences we were making and it helped solidify those.

Another good example of a company that asks for intent when they’re onboarding is Amplitude. So I was invited to our Amplitude account this week, and right after I was done filling out my personal information, it asked me what team I worked on. So I said that I worked on our product team because technically our growth team falls under product, and the next page was specifically tailored to what I put in that previous step. So product managers love Amplitude for these three reasons, time for you to get started, just like them. So I wanted to highlight another good example of just how people are really easily asking for intent, and immediately customizing the onboarding experience. So shout out Amplitude. Big fan of your product.

The Onboarding Process

So someone signed up, they’ve installed Drift, or they may not have installed Drift, but now the onboarding starts. So they are dropped into the product and this is when we start sending them nurture emails. The first section here is what we will send to people if they have not installed Drift yet. So, from days one until 30, if they haven’t installed Drift, we send them a nurturing email every couple of days to kind of bump, let them know what they’re missing out on when they don’t have Drift installed on their site. But the fact that they can get those things, all they have to do is install.

If someone does install Drift, they’re then enrolled in our activation sequence, which again goes from day one until day 30. And our goal here is really to activate Drift. So get fully onboarded, fully understand the product, to get your calendar connected, integrate your help center, all of that fun stuff that really helps show the value of Drift. And we also want them to start a conversation with our support team, because we know if they’re asking questions about how to use the product, they’re going to be a little bit more into using Drift versus someone who doesn’t have any questions about using Drift. We’ll be a little bit more worried. We want them to activate, we want them to have questions so we can get them deeper into the product.

And that’s what we send bad leads. We are really trying to help them help themselves. So we want them to onboard themselves, we want them to install themselves, we want them to understand and find parts of Drift that they wouldn’t have found otherwise, and then ask questions to us later. With good leads, we send them an email sequence from day one until day 30, where we want them to actually talk to our team. So instead of showing them how to activate Drift and how to use Drift, we leverage the intent data that they’ve given us and tell them how they can level that up if they’re on our higher paid plans.

And the CTA for all of that is to talk to our team. Let’s talk about this. Let’s talk about your strategy. What CRM are you using? Let’s chat. So that goes from day one until day 30. If they go all the way through and they don’t talk to us, we’ll enroll them in another email sequence that’s similar to the activation sequence, but the final CTA at the end of these is still to chat with our team to learn more about how they can leverage Drift. So the clear CTA for good leads is to talk to us, and the clear CTA for bad leads is to help us help you, and help you become more successful by activating yourself.

So people have gotten our emails, they’re onboarding themselves onto Drift. I wanted to highlight some successes that we’ve seen with these sort of experiences. We have this onboarding path for non-ICP customers. ICP is ideal customer profiles, so these are people who aren’t necessarily the perfect fit for Drift. We don’t want them to not sign up, we just want them to be more successful without us having to handhold them the whole time. This flow chart is a little bit scary to look at, so I broke down basically what we do.

We have automated onboarding sequences for these non-ideal customers, and we look at how long they’ve been using Drift, are they currently live? Do they currently have a playbook running? Or do they need some pushes? Do they need a little bit of extra oomph to really help them go live? We have pre-recorded on demand webinars for these leads to tune into. We have office hours held by dedicated Customer Success Managers that focus solely on this touchless segment. And we have Async reviews of instances before customers go live, so that they can get an inside view into how they’re doing without having to schedule a call with our Customer Success Managers, they just get it directly sent to their inbox and they can watch it whenever they’re free. It helps our Customer Success Managers dedicated to those people to kind of do more Async reviews, and it helps enable our customers so that they can help themselves on their own time.

Results

So what are the results from this specific onboarding experience? In 2018 we had 62 total office hour sessions. We had 531 accounts fully onboarded automatically. We had 700 total webinar registrants for those on demand webinars. And the biggest point here is that we reduced the average time to go live from 54 days to 14 days. So people were getting the answers that they needed faster and on their own time, and then they were able to understand the product better for themselves, without anyone hopping on a call and walking them through it, which was a huge, huge win.

Lessons Learned

But nobody’s onboarding experiences are perfect. And we definitely learned a lot of lessons while experimenting with our onboarding experience. So I wanted to highlight two failures that we had that did not perform well, even though we really wanted them to. So the first was our ICP Fork. So if a good lead signed up for Drift, they were an ideal customer fit. We wanted them immediately to book a meeting rather than go through the rest of the onboarding flow. So we had this split screen go for people before onboarding even started. So we just wanted them to book a meeting immediately. But unfortunately it just didn’t work. And the reason why we think it didn’t work is because they didn’t understand that they could interact with the bot on the right.

The bottom on the right is actually a conversational landing page, which is a Drift product that you can embed as an iframe into a web page. But it’s a relatively new product, people don’t really know about it when they’re signing up. And it wasn’t obvious that we wanted them to interact with it, so people would just skip this step. So it booked almost no meetings. It was such a failure. And our sales team was so excited about it, and unfortunately it just didn’t work.

Another thing that we did was we leveraged that MadKudu lead scoring and we only sent this bot free trial to our bad leads, so Madkudu lead score is bad. And it was positioned as a surprise, like congrats, you’ve been selected to have a bot free trial. You can have it for this amount of time and I’ll flip the switch for you and you can do it all by herself. People were definitely interested, but people who are a bad fit for us are a bad fit for reason, and unfortunately we didn’t give them the bot free trial for long enough for them to actually see results and want to upgrade.

This also was an experiment that we didn’t tell enough of our sales reps about and our sales directors about, where people were coming into chat and our sales reps had no idea why this free account had a bot. So it led to an internal friction, so we definitely learned about how to communicate with our team better about experiments that we were running and also understand what parts of the product will show our customers success no matter the website traffic that they have. So this was definitely another good learning experience of what does not work for our free users who are not a great fit.

As I wrap up, I just want to leave you with this message. No product qualified lead is exactly the same, so their onboarding experience should reflect that. We really want to tailor our customer experiences to how they’ll work with us, and so we make sure that they’re onboarding experiences are almost entirely custom made for them.

Want more ideas on how to personalize your website? Talk with us.