It’s not every day I get to sit down with a customer doing the kind of things on Facebook Messenger that Sinéad and CoppaFeel! are doing. What started as an average day at our office in Manhattan turned into a great one when I had the chance to speak with Sinéad over video call in London. It got even better when I learned about the social impact of her Messenger Bot, the results from their Facebook brand lift study, and her experience with Spectrm.
Let’s start with some background on Sinéad and CoppaFeel!
Tell me about yourself, your role, and your experience. Who is Sinéad?
I’ve been at CoppaFeel! for 5 years. I’m actually just coming up on my 5 year work anniversary this month. I look after all of our creative and our brand campaigns. That includes things like our Spectrm Messenger Boob Bot, our website and digital comms as well as our big media campaigns in cinemas and out of home. It’s quite varied.
I’m essentially brand guardian for CoppaFeel!, which is very digital first and always looking for ways to innovate. As a charity, our focus is particularly on digital because we’re trying to reach a young audience who are accessing everything through their phones.
It’s been amazing to see that we’ve had such a phenomenal response to the bot. And that, aside from the fact that people think it’s cool, which is obviously nice, that people are using it in the way that it’s intended. Going to check their boobs, and feeling more confident about it. It’s pretty huge for us.
Yea that’s pretty incredible. And thanks for all that context. I’ve actually played around with the bot and it’s great to see a use case like this that I don’t think any of us at Spectrm or our founders would have expected.
What, you mean their intention wasn’t to build boob check bots? Weird. Really big opportunity there.
Ha. If only we’d known…
The challenges of changing behavior online
Before we get into the actual impact of CoppaFeel!’s Boob Bot on Facebook Messenger, can you outline some of the challenges you were trying to solve?
Absolutely. There were lots ha. To be completely honest, the opportunity from Facebook came up right after we had just completed a very substantial research project. Last year marked ten years since CoppaFeel! was founded and a big anniversary year, so we were looking at what was the progress we’ve made so far. And what were the problems. Really trying to hone in on some of the behaviours of our target audience.
Just to give you some context, traditionally what we’ve done is typical market research. Things like, “do you check your boobs” or “are you aware of CoppaFeel!?” But what we really wanted to do was dig in a little more into the everyday behaviour of 18-24 year olds, what the barriers might be, conscious or unconscious, and what are the things that are stopping them from checking their boobs or pecs, or seeing a doctor about it.
What came out of it was that one of the biggest barriers to regular checking, which is one of the main things we’re aiming for, was the lack in confidence of how to do it. I think the lack of having means for reassurance that you’re doing it correctly was the big problem. If you think about the habit loop, you kind of want this reward. But what’s the reward? People don’t want to find something. It’s the opposite of a reward. I guess the reward here is knowing that you’re doing it well, that you’re taking care of your body, that you’re getting it right.
That’s where we started thinking about having something more interactive. That there was an opportunity to talk through it but also give people the sense that, as they go through a check that, ‘Yes! This is how you do it.” or, if you notice something, actually be able to offer more information to say, “look don’t worry. This change can be hormonal and completely natural, but if it’s unusual for you, see your doctor.” And then Facebook got in touch with their Boob Bot idea and we were like “YES!” The boob gods were looking down on us.
I think for us the problem was really having a way of telling people ‘you’re getting this right” and kind of going along with them in real-time to provide reassurance. That’s really where the Boob Bot has allowed us to do that in a much more integrated way.
The Impact of the CoppaFeel! Boob Bot on Messenger
It’s just enough removed that you feel it’s ok talking about something quite personal. I spoke to someone this morning who had been in touch because she used the bot. She’s actually going through tests at the moment because she found something and saw her doctor about it. I was talking to her because I wanted to explore that more, and potentially share her story as a case study. What was really interesting is that she said the reason she would have delayed going to see a doctor were thoughts like “Oh I don’t want to waste their time because I’m not sure.” But she went through the process and found a lump and told the bot “Oh I found a lump”, and the bot went through all the things it could be, but ultimately said, “You’re not wasting your doctor’s time. You should go and see a doctor.” So that’s what she did. I think that final nudge was crucial.
Wow. I don’t get to hear a story like that from a customer every day. Is there anything else you’re doing at CoppaFeel! that gets results like this at this scale? How does it compare?
It feels unique for a digital tool.
We do a lot of work that’s grass-roots. We have an amazing network of ambassadors at universities. We also have a group of ambassadors called the Boobettes, who are young women who’ve experienced breast cancer in their 20s or 30s, whether it’s a diagnosis, scare or preventative surgery, or through someone close to them. They’re amazing. They do a lot of talks in schools and elsewhere.
What we’ve seen is that when people have met Boobettes and had real conversations, the impact on behaviour change is really strong because people have a chance to ask questions and go through a process that feels more personalised and tailored. Someone might say during a talk, “Oh but I have breast implants, how does this guidance affect me?” Ultimately the guidance is the same, but the Boobette is there to say, “actually this still applies to you, here are some techniques that might help make this easier for you”. And that’s what the bot can do. It helps a lot in those cases.
As a digital tool there’s nothing that compares. Maybe replying manually to a bunch of individual Instagram messages… But nothing that we’ve ever been able to do at that scale.
How Spectrm helped CoppaFeel! engage their audience
Yea that makes a lot of sense. And thanks for the detailed info. Before we get into the results, I’d love to know more about your experience with Spectrm’s platform. Are there things you used more than others in Spectrm?
Having the option to tailor it down different routes. Having the ability to give that extra guidance in certain cases. Normally, when you’re communicating messages at scale, you have to simplify things down to the bare minimum. Having a conversation people could opt into for more info was really helpful.
The automated messaging through the platform was pivotal. That meant being able to offer a regular reminder through the bot itself. Initially we were reluctant because we also have a text reminder service. That’s been going since 2011 and it’s really successful and popular. People get a text every month. Something funny reminding them to check.
Initially we wanted to get people to subscribe to a text reminder, but ultimately we decided it would be more effective to offer reminders through Messenger, because they’re already using it. And we’ve seen so many people subscribe directly through the platform. It keeps the experience in one place. And it means we can also see how many people are getting those messages each month and whether people are then asking for recaps on how to check. That’s really, really useful for us.
That’s great to hear. So you had a lot of success with the re-engagement messages you sent on Facebook Messenger using Spectrm?
Yeah. 22% subscribed for a Messenger reminder. On top of that, another 9% subscribed for a different kind of reminder like a text, or ordering shower stickers we offer to have a reminder in the shower.
Engagement was much higher than we expected. We were driving traffic through ads. So this was a lot of people’s first interaction with CoppaFeel!. Because of that we expected the likelihood that someone would subscribe to be much lower since we were unknown to them.
Thanks for sharing those results. Since we’re talking results, this feels like a natural point to ask how your Facebook campaign performed in general. And before sharing that, can you tell me a bit about who you were targeting and what your vision of success was? And what happened? Walk me through some of the context of the campaign. I’d love to learn more.
Of course. Although, actually, before we get into that I just have one thing I want to mention I love. THE GIFS! That we can have GIFs. Because we’re really visual and playful, and you can’t do that through text. That has honestly been big. I’m really excited about the reminder we have coming up for May the 4th.
Yea it is a very visual medium. Having gone through your bot, I noticed you have a lot of great imagery. Was all that created for the bot? Or did you re-use assets you had?
We had a lot of the content already. I edited videos we had that lent themselves well to GIFs. There were a few things I made that were bespoke but actually it was very little. It was such a great opportunity for us because our brand is very playful, colourful and very visual.
If you removed all of that media from the experience, it would feel much more plain and much more clinical. This meant we were able to communicate something serious in a way that felt more friendly and approachable. Lots of people have commented on that. It was the perfect fit for us in terms of having something that allowed us to be very playful. That’s how we communicate on any medium really but it worked so well with the bot. It was a good match.
That underpins our whole approach to the charity. People are scared. Cancer is a big scary word, which is something I sometime’s struggle with. Obviously cancer is not fun, and it can have a huge impact on people’s lives. But there’s a lot of dialogue around cancer that’s about “fighting it, battling it” and that maybe does a disservice to the fact that you can live with cancer and survive it, and that the chances of doing so are significantly higher if you find early enough and get the right treatment. So we’re always looking for ways that we can talk to our audience to tell them all the practical things they can do and be quite positive – this isn’t about looking for cancer but about being proactive about your health. The bot was great for that because you’re able to reassure people and also take some of the sting out of it.
On CoppaFeel! outperforming their industry and the EMEA region on Facebook brand lift metrics
Tell me a little more about the campaign.
We finished building the bot in end of October. Soft launched through their own channels for first couple of months to iron out any problems. That was useful. Got rid of things that didn’t work and refined it a bit. There was more to do but hard to get through everything.
Was it quite fast to be able to build everything out?
Yeah! The only thing I would say I found challenging is the minimap… I’d look at the map and think, “this is the inside of my brain right now.” But, it was very satisfying learning how to navigate it. It was pretty easy to use once you got used to it. Minimap aside. It was really quick when we turned it around.
So we did that. We did some testing. There was more we could do but we got to a good place. Then in January we launched the paid ad campaign.
Our objective was really simple: to reach as many 18-24 year olds as possible. Most of the spend was targeted at women but some to guys as well.
As an organisation we speak to people of all genders because breast cancer can affect you whatever gender you are. Breast cancer affects about 400 men each year in the UK, but it does predominantly affect women, or people who were assigned female at birth.
Let me pull up the brand lift study Facebook did to show you some of the results. This was really focused on the paid side but the results were amazing.
Yea that would be great if you could walk me through the results of the Facebook brand lift study.
For me the point of interest is actually the results we get through Spectrm in the backend of the bot because what I’m interested in more than anything is the behaviour change. To see if people are actually using it to check their boobs. That said, people responded really well.
Since you touched on that, before you go into results, what are the things that you’re looking for in Spectrm that help you?
Mainly to understand if the bot was helping young people to feel more confident about checking their boobs or pecs.
The conversation analytics were really helpful, as we were able to ask people if they felt more confident at the end of the experience, and track the impact of the chat. We asked, “Has this chat increased your confidence when it comes to checking your boobs/pecs?” 98% of people said yes!
It was also really useful to be able to see the gender split and to see how many people are subscribing to a Messenger reminder, and therefore committing to making checking part of their regular routine.
Ok great, so what about the Facebook brand lift study?
We did a series of feed stories and poll-type questions, on Instagram and on Facebook. I built all the ads with the idea that each ad would simulate going through the experience in a sense. The ad utilised taps, touches and swipes to let you ‘complete’ one part of the boob check, which you could swipe up to complete in the Boob Bot itself.
Facebook were tracking the brand uplift. We were more interested in whether people did the check. But it was great to see that people responded so well to it.
Ad recall was really high. Awareness of the bot and CoppaFeel! was really high. Facebook basically told us it was way above the benchmark and had incredible campaign awareness. Of the bot, of CoppaFeel! and of the action intent to actually use it.
Basically, a lot of people used it and a lot of people remembered the ads so I think they either like the idea, or they really liked the boobs.
How the biggest impact goes beyond the brand lift
Ha. Good point… Hearing you talk about your goal makes me think how beyond the brand lift study, for you it seems your measurement of success wasn’t directly brand lift but whether people are going to doctors and getting the checks. It sounds like that was also quite successful. Did you have people talking a lot on social media about the bot?
Yeah we got an amazing response to it. For us, the key metric of success was people checking their boobs monthly, which is why the subscription part of this was really key. It’s people committing to keeping this up as a regular part of their routine. Also the stat about whether people feel more confident checking. Those things go hand in hand.
In terms of social, we ran these ads throughout January in the lead up to World Cancer Day on the 4th of February. We have big network of ambassadors like bloggers and Instagram influencers, and so we asked them to promote the bot to their audiences to mark World Cancer Day. They shared the bot and we saw a really amazing response. One of our ambassadors even did a live boob-check using the bot to guide her through it (complete with tape over her nipples, to avoid censorship!) I don’t know if you know the band Little Mix – they’re massive in the UK among 17 year old girls.
Haha, I lived in the UK for 10 years but that description isn’t quite me…
Lots of people shared it then. So Perrie Edwards, one of the singers from the band, with about 10 million followers on Instagram, shared the bot on February 4th. So did Ferne Cotton, who is a TV presenter and one of our patrons. She also has an incredible reach.
We saw an amazing response. Lots of comments. Huge uplift on that day. What was amazing is that people weren’t just sharing it, they were actually using it. There was follow through, and in the course of one day, we gained over 3000 new users.
Ultimately, it’s about people like who I spoke to this morning. A woman who had been empowered to go to her doctor and found something. Hopefully her symptoms will turn out to be benign but, either way, she found the lumps early and, ultimately, that’s what we want. To make sure that people are in the best possible position, if it is cancer.
The response has been really phenomenal and better than we expected. Now we’re regrouping to try and figure out going forward. We don’t really have a budget for social. We’re not one of those huge charities. We’re trying to work out how we can continue the traffic to the bot and keep people using and talking about it.
Well that’s amazing to hear. About the response. Not the budget obviously.
Out of curiosity, you mentioned your key metric for success was that people go get checked but you also said the subscription side for the follow up checks was a big part of it. Is there any other medium you’ve used that achieved that kind of success in terms of subscriptions?
Not really. No. It’s hard to say that definitively because the other subscriptions we run are a. little harder to track, like our text reminder service. That’s something we’ve done for a long time. You might have seen it on social, or you might have heard about it at a talk. So there are multiple routes to subscription, ande it’s harder to track and attribute growth.
In terms of the scale and the speed of people subscribing and committing to action, we’ve never seen anything like it. It’s been a big thing for us.
What was, in a bigger sense, the impact on CoppaFeel!? Not just necessarily the bot but the broader impact.
One of the things that’s most exciting about it is that it feels so innovative! When you’re trying to affect behaviour change, you have to be clever and creative to get your message heard, and for it to have impact. As a small charity (we’re a team of 15!) one of our core values is that we’ve always tried to be very innovative. I think one of the nice things about being smaller and not having much money is that you have to be creative. I think we’re very good at being creative. We do a lot with a little.
This opportunity spoke to that. Gave us a chance to actually change the way people act. This felt like the next natural step for us as an organisation and something we could not have done before the opportunity with Spectrm. So to have that opportunity has been so exciting for our whole team. We’re now ready to go on Robot Wars. Probably another reference you don’t get it.
I do actually get that one surprisingly. And it’s great to see how much excitement this generates in general. Thanks for that high level vision.
What building a chatbot with Spectrm was actually like
The marketer in me also feels the need to ask if this means you would recommend Spectrm?
Oh my god 100%. Yes. Definitely. Apart from the minimap, I loved it. Also, Nicholas from Customer Success has been so great. Really really supportive. So much support and advice. We’re all big fans so thank you.
Happy to hear that. That was actually my next question as well. What was your experience like with customer success, which in this case was Nicholas?
Customer success couldn’t be better. So nice. So helpful. I felt like he bent over backwards to help us do this. That’s not always the experience. It’s always nice to hear that people are excited because it’s a good cause, but I think we all felt really supported in all the (less exciting and inspirational) day-to-day stuff too. Even since launching if we ever had questions about stuff we’ve been able to get answers really fast.
The idea of building a bot feels scary to a lot of people. Did this make that whole process less intimidating?
Yeah massively. At first, I thought, this is going to be a huge amount of work. At the outset of the project we thought the build itself would be outsourced, so we were like “great yeah, let’s do it.” Then we realised we’d be creating the bot in-house. The team went, “well I guess that means you’re doing it.”
I was intimidated by the idea of building it, but Spectrm was really quite easy to use. Once you get into it, it’s quite intuitive in terms of the flow of you how you build something. And whenever there were things I was stuck on, I would have a call with Nicholas and talk through the issues I was having. He would explain things in a way that even I could understand. So it was really much easier than expected. The bot building was really smooth and easy.
Ultimately I feel that part of why the bot is so great is that you built it. It speaks to your brand so much. I don’t know that Facebook would have been able to do that. Do you feel that you building it made it better?
You know what. That’s a funny thing. When Facebook first approached us we thought the build would be outsourced, which we thought would have been great from a technical perspective but from a brand and experience perspective might have involved a long process of back and forth – the way we talk about this topic is quite unique in the breast cancer space. Then they said, “No no you’re going to build this.”
When we realized that we were going to build the bot, I spoke to the team and said “Actually, I think it’s going to be all the better for that reason and a much easier process. Because we can shape the way that we speak and deliver it in a way that mirrors everything that CoppaFeel! does.”
I think it was better because we built it. And could tweak it and change it and play around with it. And have our team test it. And share it with our community.
Rather than outsourcing it and feeling like this is the bot we have and that’s final and we don’t understand how to use it and it just sits as it is. We know that if we want to update it, it really is quite straightforward. We know our audience better than most, and we can listen to them and use what we’ve learned to alter things and make it better.
Yea, you can really feel the brand in the experience because of it.
Well thanks so much Sinéad. I really appreciate your time and walking through this all with me. I’m going to look through this brand lift study in more detail and prepare a case study. I think this is an incredible story. Not just from the brand lift results perspective for Facebook Messenger but the broader impact for you as a charity and for the domain you’re working in. So thanks again for your time.
Not at all. If there’s anything we can do to help, then I’m very happy to do it. We feel very grateful. It is genuinely one of the biggest things that has ever happened to us. So thank you.