EP24: How 7 & 8 Figure Brands Segment Their Messenger Subscribers


Episode Summary:

In “How 7 & 8 Figure Brands Segment Their Messenger Subscribers,” we cover how to successfully Tag and Segment your conversational messaging list.


We talk about providing customized Messenger experiences based on where people are in their customer journeys. We also break down best practices for Tagging, and how to use Tags to capture essential customer data.


Let us know what you think below 👇👇.


Episode Highlights:


  1. 1:20 What are tags in ManyChat, and how do you use them?
  2. 3:30 How to customize tags based on data you get from the audience
  3. 6:08 Using audience segmentation to target groups of people with specific kinds of messaging
  4. 7:42 How to avoid annoying your customers with bad messaging
  5. 9:35 More customer information leads to a better customer experience
  6. 10:34 NINJA TRICK 1: Using tags to target people based on trends
  7. 11:50 What you should actually be tagging
  8. 14:37 Rewarding your company’s best customers with The Early Access Bot
  9. 15:40 NINJA TRICK 2: Reducing the number of people who unsubscribe from Messenger


Resources Mentioned:


  1. Facebook Messenger Tool – Manychat
  2. Email Platform- Klaviyo
  3. Google Analytics


Episode Transcript:


Jeremy: Welcome, welcome to another week’s episode of The Messenger Mastermind Podcast. As always, I’m your host, Jeremy Horowitz, joined by my amazing co-hosts, Mark A. and Ben Vandal. How are y’all doing today?


Mark: Gentlemen, how are you?


Ben: Good, how’s it going?


Jeremy: Really excited. Coming back to you with a really interesting topic, something that we’ve found to be instrumental in the way that we run our Messenger efforts and really wanted to share it with you all. Tags and then how do you segment those tags. Ben, Benny-boy, the master of tags and segments. You want to kick us off by explaining what are tags in ManyChat and how do you use them?


Ben: Tags are a very important part of business and the way you communicate with your customers. You don’t want to talk to all customers the same. You want to tag them with a breadcrumb. Some websites call them cookies. Email and Messenger, we call them tags. You tag them based on actions that they’ve taken, interactions with your bot or your email and you can contact them later differently based on those actions. For example, you reach out to a person and you say, “Would you like to buy tanks or leggings?” If they click the ‘tanks’ button, you tag them with a tag that says, ‘tanks’ or ‘interested in tanks.’ You can talk to them about tank tops. If they click the ‘leggings’ button, you can tag them with ‘leggings.’ They can talk to them about leggings. Any time you’re creating a bot, tags are a vital part of our bot. They go into every button that we build. What you do is, when you click the Edit button option, at the bottom there’s a tag that says ‘additional actions’ and you click the action tag, and it says ‘add tag’ ‘remove tag’ ‘triggers app.’ We’re going to use the ‘add tag’ feature. You click ‘add tag’ and you can enter any tag you want. If you have the paid feature of ManyChat, you can create as many tags as you’d like. They do get a little out of hand if you don’t re-use them, but if you’re efficient and organized when you make your tags, they’ll be really, really pivotal in your action plan going forward.


Mark: When a user is going through a bot, every time they click a button, if you have assigned that button with a tag, that tag will then be put into each of those person’s profiles in ManyChat as a segment that they’ve previously interacted with. We’ve covered examples in the past briefly about how this can be done but just to give a few more. A pretty good standard practice would be, as Ben mentioned, if someone clicks on a button for leggings then we know they are interested in leggings so when we are going to send out a broadcast in the future, if we’re going to send out something specific to leggings, we’ll make sure that we talk to that audience differently than someone who didn’t click ‘leggings’ because we know that they are into this. We can go more direct with them than we might with someone who hasn’t previously clicked on a button that was tagged with leggings.


Ben: We covered this pretty extensively in Episode 12 on our Product Relaunch Episode. We send out an option and have the customer vote on do they like this pair of leggings, leggings A, or do they like leggings B? Based on what they vote, we actually follow up with them next time saying, “Hey, your product has been released” or “Hey, your product wasn’t the one that won but check out our new product” type of thing.


Mark: We do it exactly as Ben has said there. We’ll literally split our audience and we’ll say if they voted for this, they get a message saying, “Remember when you voted for this before?” Whereas, someone who maybe didn’t vote for that, so they haven’t been tagged and segmented based on their previous interactions, we will have to send a message that’s explaining more of what’s going on. Whereas someone that we know has seen previous messages about this, we can be a little bit more straightforward and we don’t have to beat around the bush.


Jeremy: This is so important because when you think about it, this is a communication tool that they customer is using to talk with their friends. It allows you to simply send a message and cater. They’re saying the same thing essentially, right? In the previous example we were talking about it’s a product launch. You’re allowed to cater that message to make it more specific. It makes it seem like, you are, you’re listening to what they’re telling you and based on that information, you can get more granular and if you tag everything properly, you can really get down to the minutiae. I’m sure you guys also tag all leggings stuff. In ManyChat I’m sure there’s mounds of data around all the different voting and leggings. You can get as specific as “Hey, we noticed that you like these three pairs of leggings. We’re launching these new ones that are a very similar style that customers who have typically like those in the past like this” and really cater that specific messaging so that when people read that, they feel like you’re really talking to them.


Mark: Exactly. If you’re listening to this and you’re familiar with email and segmenting an email, use Klaviyo, this probably sounds very familiar to you. But if you don’t, that’s okay. I think the message that we want you to really understand about this is that segmenting is the best way to get the best value out of your message.


Ben: We always talk about how precise of a platform Messenger is, so this really helps us hone in on who we’re talking to and get really conversational with them and get down to the details of what they actually want to talk about.


Jeremy: You’ll find this kind of crazy counter-intuitive thing where you’ll send messages to fewer people and see better results. That was always something, especially with the team, it always blew us away is we could cut the lists into half, a third, a quarter, an eighth but those better, more targeted messages would yield more revenue than emails that went out to fifty, a hundred, two hundred thousand people. As you really start to scale it down and really make it more personal, especially in this intimate channel where so many people are opening the message and so many people are reading it, your bang for buck just skyrockets the more granular and the more targeted you get. The beauty of a really good tagging system is eventually your segments should allow you to manage this without too much overhead and without sending nine million versions of the same message.


Mark: Perfect, Jeremy, that’s exactly right. That’s exactly what we want to be thinking about this because we want to say that if we’re going to send out a particular message- let’s say it’s a content message versus a sales or vice versa- we need to be actually looking at it and saying, “This message might not be right for everyone and if we send this to everyone, we’re probably going to have a lot of unsubscribes so let’s go ahead and let’s remove some of those people that we know, based on previous interactions because we’ve tagged accordingly, that this message will not resonate with them so why would we send it?” All we’re doing is increasing the chances of having an unsubscribe and we’re not going to gain any benefits from it by sending it to them.


Jeremy: Right, and there’s also that exposure effect of if a company keeps hitting me with stuff that I’m not interested in, I’m probably not going to end up buying from the company even if there are products I’m interested in because it doesn’t feel like they understand me as a customer and understand what I want. On the flip side of that, if a company feels like they really get me and they’re almost like a friend or a doctor- like a doctor diagnosing a problem and making a recommendation- I’m actually more willing to buy from them even if I’m not super interested in the product. I think that there’s a lot of people and a lot of good marketers out there who know how to play those heartstrings. If you can first off, just tag everything. Just having the data in the first place is super helpful because you may not think that you need something today. Where someone came from, whether there was an ad or a ref URL or from an email, but once you have all that data and you can start to filter and layer things on top of each other, down the road it just provides you with some fascinating opportunities to really talk to the people that you need to be talking to and not bothering the people that the message isn’t necessarily relevant to.


Mark: Yeah, this information doesn’t go bad over time. Everyone’s profile in ManyChat are tabbed the segments so the more information that you can get about them, it honestly is better for them. People often are concerned about privacy, especially these days, but the fact is the information that we’re gathering is not something that’s compromising to them. If anything, it’s allowing us to give them a better experience.


Ben: If you’re listening to this and your thinking, “Wow, I’m kind of overwhelmed. I don’t really have a plan going into this.” Just sit down and take a couple notes on how you’d like to talk to your customers. You don’t want to talk to your actual people who purchase from you the same as you do to the people that didn’t purchase from you. You don’t want to talk to somebody who purchased from you ten times the same way you talk to somebody who only bought a sale item from you once.


Jeremy: Yes, yes, yes.


Ben: Make a plan and go on- you can actually backlog some of your tags, too, which is really cool. Kind of a ninja trick. If you go into your broadcast message, it shows you the statistics of the buttons people clicked. It will show you total clicks on this button and the users that clicked it. That’s actually highlighted in blue and you can click it. It shows you a list of all the users that clicked that button. In the top right, there’s a button that says ‘tag these users’ so you can go back if you notice a trend in a bot, something that you really didn’t expect to happen, you can go back and tag these users and message them separately with whatever tag you’d like. That’s something we do when we notice a trend in a broadcast, we’ll send out something with an action, a CTA, and then maybe we’ll include a button where if they want more information.


Mark: A good example of that is if we send out a broadcast for a new product launch, we will often times go back and create a list of people that actually clicked through to the site, and then re-broadcast to those people maybe a day or two later asking them if they made a purchase. If they did, then we tag them that they purchased that item. If they didn’t, then we tag them that they didn’t purchase that item and we send them down a path that’s appropriate based on how they answer that question.


Jeremy: Something we should really focus on is what you should be tagging. To me, the quick-hitting list is every interaction. So, every prompt you give the user of any yes/no, any multiple option interaction, where they came from, if they click through, if they leave the site. Is there anything that I’m missing? Is there anything that you guys really like to tag that you think is super important?


Mark: Let’s go with one of the most basic. If you’re not a business that has a lot of product launches or maybe don’t even run a lot of sales or promotions, I would guess that around Black Friday you are probably doing something. Let’s look at it on the simplistic form of just Black Friday. Say last Black Friday you sent out a broadcast to your list. Say you didn’t even tag or segment any buttons at that time. As Ben mentioned, you can go back to that broadcast and create a list based off of people that clicked a certain button, say a button that brought them to your website. You know that those people are much higher intent than others. Come the next Black Friday, you can segment those people out and treat those people in a certain way because you know they’ve clicked through to your site on Black Friday in the past. I would say that’s probably the most simplistic way to handle something without getting too crazy. That’s a good place to start.


Jeremy: I agree. This is actually the very first message that I ever sent out through ManyChat was going back to Ben’s point earlier about how you wouldn’t talk to a prospect the same way as someone who has bought once or twice from your store versus ten times. We literally just asked the question, “How much do you love our brand?” The answers were, “I haven’t bought yet.” “I’ve bought a couple times.” “I’ve bought 3-5+.” We immediately tagged everybody. In stacking that on top of the Black Friday example, now you know where people are in their relationship with you, how many times they’ve bought, and then do they like Black Friday promos? Do they like promos in general? If you have a new sale coming up, you can cater the “Hey, new sales launching tonight.” You say “Try this out” to the people who have never made a purchase. You say, “Hey, add this to your collection” for the people who have made one to two. Be like, “We know you’re going to love this, just come back and buy it” for that super purchaser group. It’s just two ways to cut those really simple data points and as you start collecting, you can start stacking on more and more. Those questions get you in for the 24 plus 1 rule. That’s exactly why the C.V.G. product launch strategy is so brilliant.


Ben: Another good one we use is early access. Like Mark was talking about for Black Friday, Cyber Monday, if we have any releases too, or any training campaigns or anything, we do an early access bot so people will find out first through that bot. There’s super high intent on those people, as Mark said. If they were interested in one early access for a launch, they’ll probably be interested in the next one so you can talk to them a little differently. It’s really helpful. I mean, it’s everything when you’re talking to your customers. If you just think about it as being a regular person, you’re not going to walk up to somebody on the street and talk to them the same way you would a friend or a family member or even a co-worker. You’re going treat all those people differently. Having that information, is really, really helpful for a conversational messaging tool like ManyChat.


Mark: I can give another pretty good example, too, and I’ll compare it to email. From the email side of this, it’s definitely more of a ninja trick, whereas with Messenger it’s not quite as complicated. If you’ve ever seen or you do it yourself for your email, where you click an unsubscribe button and rather than just unsubscribing the person from your list, you give them the option to say what type of messages they would like to unsubscribe from. Maybe they don’t want to receive certain content, but they do want to receive certain sales. On email, you would just go and say, “I want to receive these things, but I don’t want to receive these things.” We can do that exact same thing with Messenger except it’s a lot more seamless and a lot easier. A good example would be if you send out a broadcast for a piece of content and you just have that first message asking them if they want to receive that content. If they put yes, then you tag that and say that this is a person that in the future wants to continue to receive content. But if they click no, then you tag them and say that you no longer send that person content because you know they weren’t previously interested in it. Now you have a way to identify what people are more likely to want that content versus not and you don’t have to go and bother the people that don’t want the content.


Jeremy: We covered a lot of really interesting stuff here today. If you’re looking for the one key takeaway as something to get done this week, start tagging where people are coming from into your bot. If it’s an ad, just make sure you tag them with the ad the same way you would tag a UTM parameter in Google analytics. If it’s from a ref URL, just name the ref URL. Whether it’s a promo, a product launch piece of content. Start knowing where your Messenger list is coming from. It will help with understanding the growth of your list as well as where you need to be investing more and how you should be talking to each of those customers. It should help you get going and then obviously, play around with it. This is very much like pulling a thread. Once you start pulling, the next thing will pop up and the next thing will pop up. Next thing you know, it’s a year and a half later and you’ve got 400 tags or whatever the C.V.G. team has in their crazy account. It allows you to have a nice, powerful dataset to start filtering great messages to your customers.