Experience Makers Spotlight: Streamlining Follow-Ups in Customer Lifecycle

Transcript

Hi everyone and welcome to my talk on the Skill Exchange. We’ll be talking about executable campaigns, which is something I’m really passionate about. You’ve probably heard about them quite a bit in terms of request campaigns, execute campaigns. I’d love to explain why I think executable campaigns are a really good thing to be using in your Marketo instance. So we’ll jump right in. My name is Minna. I work at BlueprintX. I’m based in Melbourne, Australia. I’ve got a few certifications, including the architect and expert exams. I’m currently on the committee for drafting the next Marketo architect exam. I’m a Melbourne Marketo engaged user group leader, so a mug leader in Melbourne. I really enjoy running those sessions.

My passion is really about mapping out lead life cycles and sharing the power of executable campaigns to really fix common issues internally and also for programs that you’re setting up within Marketo.

You may have come across the executable checkbox within Marketo and you may have heard about this but not actually used it. So just a few tips on how we can actually access the executable campaigns at Marketo. If you select the executable checkbox to use this function, it’s called execute campaign in a flow step. It’s essentially similar to a request campaign in that it requires another campaign to call on it in the flow step. It’s primarily known for their use in completing tasks for operational items. But today I want to cover how we can take this another step further.

So executable campaigns can also be used and in this case study for declined credit card follow-up prompts. A not-for-profit organisation actually had many people signed up to give monthly ongoing donations and they were finding that the credit cards were declining and it would take a while for the issue to be identified and for this to be followed up manually. This was actually costing the business a lot of time and effort to actually chase donors for updated credit cards. So what we wanted to do was to implement something more efficient within Marketo to do this for them. Once implemented, there was a 66% increase in successful credit card updates and the time taken was reduced from one to two days instead of two to three weeks.

So here’s a visual of the program architecture.

Of course it is zoomed out but the idea here is that the blue was emails to prompt people to update their credit cards which had declined.

The green one is for SMS messages to go out to people if they didn’t respond to the email prompts and there were a few more hard copy letters that were sent out afterwards if those prompts weren’t followed. But a zoomed out version will be able to see that the main area here I’d like to focus on is that we were checking if the credit card was updated using executable campaigns and this ensured that we knew if the credit card was updated or not before emailing or prompting the person for an updated credit card.

So our viewer Marketo shows that executable campaigns were used for this check. It was kind of used for the checker to basically check if someone had updated their credit card and if they had updated it we removed them from the flow.

If they hadn’t updated the credit card then they went on to receive the next prompt to update the credit card. There were wait steps as well and as you can see in the yellow boxes executable campaigns were used in order to ensure really accurate emails and SMS messages going out to ensure that this wasn’t going out to people who had updated their credit cards. So executable campaigns help internal processes run smoothly as well. So you may have heard about using executable campaigns for data standardization which is a really powerful way of ensuring that your data is updated ahead of, for example, assigning someone to a queue and ensuring the smooth process is there. Additionally, the lead lifecycle, ensuring that a score is assigned to a person ahead of moving them on to the next bucket, ensures that they are actually moved along to the next bucket without getting stuck without a score and therefore stuck in that limbo stage. So we’ve all seen those errors come through where someone’s not got a score, they’ve not got a value that we’re looking for and therefore they are not actually passed on to the relevant person. We can use executable campaigns to really ensure that these are not occurring and ensuring that our processes are running smoothly. Additionally, for attribution reporting this often has values due to a source not being finalized ahead of a sync.

So just a quick screenshot of standardizing using executable campaigns ahead of the next step. So for example, in this instance if we’ve got country and people are putting in US, USA or United States, we ensure we standardize ahead of syncing to Salesforce. But what I’d like to show as well is our product push prompts using request campaigns. So often we use executable campaigns in comparison to request campaigns and there still are really good use cases for continuing to use request campaigns and we do offer more flexibility in some areas. So a visual of our program flow here is that a global credit card company wanted to send out some prompt emails once a customer had already signed up for a credit card to basically move them on to the next product.

So in this view you’ll be able to see certain prompts which I’ll zoom into a little bit more. But essentially what we needed to do was check if someone had that product and then prompt them for a certain product that they may have not yet signed up for. Within Marketo’s smart list I was able to use request campaigns and lots of choice steps so that we could check each product and using the choice steps we were able to prompt the person for the relevant products. This is something that we wouldn’t be able to do in an executable campaign so therefore request campaigns was a better use in this scenario.

So one key difference to understand is how these campaigns complete within Marketo. With executable campaigns the campaign will fully complete before the parent campaign moves on to the next step. And this means that all the data updates and processes within executable campaigns are guaranteed to be finished before anything else happens. By contrast request campaigns often operate in parallel. The parent campaign fires off the request but immediately moves on to the next step without necessarily waiting for the requested campaign to finish. This can sometimes cause issues. For example if the lead data hasn’t been fully updated yet but the parent campaign continues and passes the lead along, that can result in incomplete or insufficient data being used downstream leading to errors or inconsistent behaviour. So understanding this key difference helps you to design a more reliable and predictable workflow by choosing the right type of campaign for each task. And the comparison I like to use here is a cake analogy where if you’re baking a cake it’s really important to understand if the cake has been iced before moving on to the next stage, putting the cherry or the decorations on top. Whereas with a request campaign we don’t know if the cake has been iced yet or what part we are up to. So depending on the processes that we’re looking to use executable campaigns for, sometimes it is really crucial to know what stage we are up to and if the first step or two steps has occurred before moving on.

There are some similarities though between executable campaigns or request campaigns and that’s why they are compared so often. Both of them require another campaign to call on them. But each have their best use cases and it’s important to understand them and to apply this when considering using executable campaigns or request campaigns. Executable campaigns are great for sharing flows like stamping, data cleanup, lead routing and follow-up communications. Request campaigns on the other hand are better for lightweight processes, alerts, custom triggers, API callbacks and a little bit more flexibility.

The key takeaways I have today is for you to consider which campaigns are the most crucial to your instance and what does it take to actually ensure that those run without errors. Is there a need for data standardization ahead of some lead queues to ensure a smoother path? I’d also like you to consider processes where you need to follow-up with end users for real-time events. So think the credit card decline program. Is there something where you’d like to follow-up people in real-time with the correct information and ensuring that you’ve got the most up-to-date information about their account? Request campaigns are lightweight and great for fire and forget tasks but executable campaigns are perfect for following up in real-time and ensuring that we’ve got the latest information on the person. Also consider effort versus reward. So going back and updating all your request campaigns to executable campaigns may not be doable but think about the most crucial ones that need to be updated. And also moving forward, how you can consider using executable campaigns in the future without redoing existing ones. Happy to open up to some Q&A a little bit later on in the session. Really keen to hear your questions and understand how you can use executable campaigns in your programs. Thank you. Thank you for all those great tips, Meena. Our final Experience Maker Spotlight will be with Maliniya Chan. Welcome, Maliniya. Hello and a warm welcome to this session. I’m happy to share with everyone how you can master multi-channel lifecycle engagement.

Today’s session is split into two parts. First, we understand the intricacies of how we develop the strategy across different stages. And the second part, I will zoom into providing some practical applications using WhatsApp, Viber, or push notifications. But before that, let me give you a quick introduction about myself. So I am Maliniya Chan, working in IDP Education with extensive experience in the Southeast Asia region, covering seven countries with multiple languages and cultures. As IDP Education is a global leader in education service, we help people with ambitions to study abroad. And as a co-owner of IELTS at English Proficiency Test, we want to actually help test takers to move forward if they are next steps in their life journey. As a Marketo Champion, I’m grateful for the opportunities provided by Adobe, including being an Adobe Summit Speaker this year and writing articles in perspectives. As an Adobe Certified Master at Architect level and obtaining my subject matter expert, I’m delighted to share my experiences in this simple Marketo User Group as a Marketo User Group leader. Let’s connect to create a vibrant sharing community for Marketo language. Let’s start with key stages in lead nurturing. Depending on your industry, there could be different stages used in your organization. So I have created generic key stages that could be easily applied to your organization, be it e-commerce, B2B, or B2C.

The four key stages commonly used in lead nurturing are the first one starting from discovery and then moving down to engagement, conversion, and finally, retaining and growing them. Retaining them is like a renewal, for example, a subscription service or an exam that needs to be taken every two years, while growing could be an upsell or cross-sell for other products and services.

At IDP, I create the campaigns for cross-sell and upsell across these two instances. This template can help you map all your current campaigns in Marketo and to plan for new campaigns. Using the four stages, I’ve included some campaign examples that could work for your organization.

If you have filled in your template, you can look at the performance of these campaigns to understand the conversion or engagement.

I use this for planning multi-channel strategies. For campaigns that have key touchpoints, it is possible to reach to more customers through their preferred channels. It also serves to reinforce your message from multiple touchpoints, thus improving engagement and conversion rates. If your organization has multi-channel attribution, this is what it will look like.

D3 strategy. At this stage, I’m sure you’re all interested to find out about how to determine the usage of other channels. I call this the D3 strategy. First, to decode the data. It’s highly important to know how your audiences interact with your brand from websites, mobile app data, social media messaging, dynamic chats, and also understanding these lead sources driving traffic. So, for example, here we look at Gen Zs or even Gen Alphas with mobile app data. This is your foundation to gather insights to fuel smarter decisions. Next, define your audience. If you have personas that you’re targeting, etc., understand what is your preferred platforms for the audiences. Example, IDP has our Emerging Futures report that represents the voices of international students, where we understand their motivations for international study, for example, career development and educational quality. This step helps you speak the language of your audience, know their needs, habits, and motivations. In the final step, you look into determining the channel from decoding the data and audience. I would suggest to pilot with key campaigns instead of trying for all campaigns, as most of these new channels do involve calls. For example, for each message sent, monthly fees, integration with API, etc. So, this step ensures your message lands effectively at the right place, right time, and right format. So, example here, like for IDP Fasting, this product is actually utilizing both web and app channels for study abroad applications.

Using the campaigns I’ve marked up, I will focus on two high-performing campaigns. It is not a myth, but you can manage all multi-channel strategies and campaigns in one centralized platform, Market to Engage.

Let’s dive into discovery stage, where we look at prospecting leads. So, prospecting leads campaign, this here is for users who have signed up via newsletter or events, but are stalling at the discovery stage, showing little or no engagement to move forward. So, we want to move them from discovery to engagement using the decode data of website behavior, chat interactions, or app usage. So, we’ll define audiences who have not attempted a booking or those who might consider an additional task booking a game. So, we determine our channels where email remains the primary outreach method and alternative channels like push notifications are good with users with high activity using the app, but no phone number, and in-app messaging for real-time nudges during app sessions.

By combining behavioral signals with a diversified channel mix, we can personalize outreach, increase relevance, and accelerate user progression into engagement fun.

So, here this is an example of how we set up in Market to Engage. Please do note that you need an add-on for Market to Mobile engagement. So, you could do using the filters or triggers for visiting the web page. You also could look at the app activity, also looking at dynamic chat where you have a dialogue or you have engaged an agent in a conversational flow. And then you set up this flow as send email, you wait for 30 minutes, and then you send a push notification. So, if you are wondering why there’s a duration of 30 minutes, so once the user is receiving this on their mobile, there might be too many trivial messages that is happening from email and the app, so this would space it out.

Here’s an example how I set this up. For the push notification, you just need to use a token and then you can put in your message. For example, the first name is used here. So, once you have inserted your token, you can include a message, you could copy and paste in your message, you could also use emojis for this purpose.

And then you have the option to do three call to actions, which is launch app, a landing page, or external URL. So, for this example, I’ve used external URL and I pasted in a URL with new TM codes.

So, you can also then preview and approve your notification. So, prospecting leads, this is a case study where we actually still have other assets, example email, and also the other one is the in-app. So, in-app example is pretty straightforward. You can have an image with a text and then a call to action. So, in this clip here, it shows when you open the app, this is the in-app message you will see. So, let’s dive into the engage stage for abandoned cards. With all this activity, for example, prospecting leads, we want to close the loop for those who have not completed the application process.

So, here we are looking at how to optimize an existing campaign that is doing well on email but could do better with other channels. So, we want to engage them through alternative high impact channels. So, looking at the decode data, we look at the SMS versus WhatsApp, high mobile penetration, strong consumer engagement with messaging apps, and mobile app usage. So, how do we determine the channels here? We have a stakeholder collaboration and from the discussions we recommend from the stakeholders to use WhatsApp for selected countries like Singapore, Malaysia, and Indonesia. How do we set this up? So, here are some of the screenshots to set this up. So, you will need to map the tokens with the templates for the WhatsApp. These are examples of the token, be it WhatsApp or be it Viber, the tokens will be similar and use it in the description. So, you could then set a campaign for request to do a trigger and the trigger will request a campaign and you will call the webhook. Here is an example.

So, how would this look like with your service provider? So, you need to actually get your templates approved by Matter before you can actually use them on Marketo. So, this is a very simple example whereby there is an image and then the text message could use the tokens that you have also set up. So, your next question will be what would be the results? Overall, all campaigns using non-email channels increase conversions by 123% and GA sessions increased by 420%. This is estimated over 8 months period.

I hope you have gotten some ideas with your multi-channel strategy. I have summarized the following key takeaways into five points. First, always start with lead nurturing stages. Understand where your leads are, from discovery to grow. Map campaigns in each phase. Align your messaging and tactics to match the lead’s intent and readiness. Apply the D3 strategy. Decode data, define audience and determine channels. Think of your channel mix and use a mixture of push, WhatsApp or other chat applications with email to maximize engagement. Most importantly, test your campaigns, pilot and learn, analyze the results and scale what works.

After this session, there are next steps you can take to create your own multi-channel strategy, craft your own strategy using the 3Ds strategy, test and learn and of course audit and optimize. Thank you for your time today and I hope this session has been useful for you. I am working on a Perspectives article for a more detailed version so look out for it. I look forward to answer your questions in the Q&A section. Fantastic! Thank you so much, Malina and Mina for sharing all of your knowledge with us. Now we’re going to answer a few questions that we’ve gotten in the chat. So if you haven’t posted your questions yet, this is your time. Get those questions posted in there. So for our first question, Mina, this is for you. Are we correct to treat customizations of flow step choices the same as standard smart campaigns or for executable ones? So actually for executable campaigns, there is no flow step choice option and that is one of the limitations there that I was pointing out. So with request campaigns, you can get to add your choice steps within there. Executable campaigns, unfortunately, is restricted to just the one step. So that’s something to keep in mind when you’re sort of weighing up between using one or the other. Always choose request when you do need those choice steps not executable.

Got it. That makes sense for me. This one is for Malina. So what is the typical timeline for seeing results from a multi-channel engagement strategy? So really it depends, but I have seen as quick as like, let’s say if you send it out the next day, you can see some results from the time or the longest it will be for me about a month. But it depends on the type of campaigns that you are executing and the conversion results that you are looking for. When would you not want to use an executable campaign and what is the actual advantage of not setting up everything is executable? Yeah, it’s a really good question. So executable campaigns are really good at the things that they can do. So very much one step actions. So for anything operational, when we think lead scoring, think data standardization, anything you need to have done in a sequential kind of way. I always think executable, but if you have something that’s a little bit more left of field requires more flexibility, that’s where you would look at something like a request campaign. As we were just saying before, it provides that extra flexibility with choice steps. And so you really have to weigh up what you’re wanting to get done there. The way I sort of categorize things in my mind is executable campaigns is always when you need one thing done, the next thing done and the next thing done. So very much step by step. Mostly it is operational items, but it’s really good to start thinking about using it with follow ups. So if you’ve got people subscribed to certain things and they’re paying an ongoing subscription, if their credit card declines similar to how it’s to my original example, using executable campaigns to follow that up and ensure that you first update the data value or sort of check that they’ve updated the data value before that is really important. So executable campaigns is really for those nitty gritty checks ahead of your email sends, whereas request campaigns are more for your flexible, a little bit more creative campaigns.

Got it. That makes sense for me. What metrics should I have? Oh, sorry. This one is for millennia. What metrics should I track to evaluate a pilot campaign success? So depending on your campaign objectives, so let’s say if this is let’s say you need to have a conversion, that means something is paid. Then we look at the number of conversions within that period of time. Or if you just want to know, like if there was like any activity on the website, it could be number of GA sessions. So these are a few things that you can actually consider. But if your call to action is obviously going to be, say, download the app, then it will be how many people have downloaded the app or how many people who have submitted, say for a demo or a request for trial. Got it. All right. And this is another one for you. Can you clarify your expertise when you would want to use a or sorry, what is the main differences between a request campaign and executable campaign? That’s what we’re looking to hear on this one. For sure. So this is something that’s often confused. And hopefully, this will help get rid of that blurry lines between the two because they are kind of similar. They both rely on another program or another campaign to call on them. But the main difference is that request campaigns work in parallel with other request campaigns and things in your market or instance, whereas executable campaigns work sequentially. So the main thing that we need to remember is with executable campaigns, we can rely on first steps actually occurring and completing before the next step occurs. So you might sort of see in a lot of market instances, a few sort of glitches or basically errors come through where we haven’t actually updated our data in time for the next step. And that’s because lots is happening in that one market or instance. And sometimes they happen in the in the slightly incorrect order. So we need to start thinking, how can we stop that and jump in with executable campaigns to ensure we have things basically normalized, data normalized and ready for the next step. So essentially, executable campaigns give you that confidence that the first step has been done, that we’ve normalized your data from a US to Australia, from US to USA or whatever it might need to be for the next step to actually pick it up and for example, pass it on into a lead queue or something like that. So the main difference is knowing that one works sequentially and one is working in parallel with other things. They actually do work in a similar way beyond that. So they’re both being called on by another campaign and they both require that same sort of thought process of one campaign calling on the other one. But just to think about it as executable campaigns, we know that that part of your program is complete. So we can then confidently move on to the next part and know that nothing will break in the process. So when we reach out to people about credit cards that have declined, we really don’t want to reach out to someone who’s already updated that. So that’s that sort of getting it really accurate as opposed to you may have already updated your credit card. So disregard this. We don’t need to have that kind of wording anymore because we are confident that they have done that. So executable campaigns are just that extra confidence that the steps in your flow step have actually occurred because they just act in that way to ensure that that that activity has actually completed with request campaigns. There’s a bit more fluidity there. Got it. Thank you. Thank you. Thank you. It’s a good explanation. So I appreciate you diving deep into that. All right. Molina, I have a question for you. How would I integrate WhatsApp or another application such as like Viber or Line or Telegram with Marketo engage? So you first need to have a service provider. So I do look for a service provider that is able to help you do these integrations with various applications that you’re after. So you don’t need to sort of integrate with multiple service providers if they only provide one or two of the other chats. So once you find that you would have to integrate and you would need to then set up it as a API so that you can do a web call. So once that is being set up, do actually know that you should also get in your communication logs from the service provider so that you will be able to get information into Marketo. When you talk about the sort of statuses, like is it delivered, is it read or even reply? So even though if it’s a one way that you’re using, meaning you’re sending out the communications, just one way to, let’s say, WhatsApp or Line or KakaoTalk, if you want to have a read notice or you want to reply notice, you can actually use this as a trigger and then send it as an e-alert to someone in the operations team. And operations team would know that particular person and they could follow up as an option. I’ve seen the first response is probably there and you can actually see it in the activity log if you actually click through. But do note that in today’s world of texting, the reply could be an emoji or an image to say, yay, I like the message. All right. Mina, another one for you. Any tips around troubleshooting errors in executable campaigns? Oh, that’s a good question. Hopefully there won’t be any errors. So actually executable campaigns are helping remove a lot of the errors that are happening in your Marketo instance. So your error log is hopefully going to look less populated there. But in terms of an executable campaign, I guess not working.

It’d be similar to a normal Marketo campaign where you want to check that it’s actually functioning. You’d probably check on a few contacts and make sure that someone has gone through that program and it’s affected them. Good question. Just trying to think of actual errors. I guess the main things to note with executable in terms of that functionality is ensuring that once you have selected the smart campaign as executable, you actually can’t change that back. So that’s sort of something to consider when we’re thinking about is this working? Is it not working? Also, something to look at is the icon for your smart campaign sort of changes from a regular globe to one with a cog and it’s yellow. So you’ll know it’s functioning and firing. But yeah, just your usual activity log will be telling you if people are moving through and it is working.

You’ll also be able to see if that executable campaign is, let’s say, normalizing your data. You want to actually sort of run a few checks. So you might sort of send through a lead with something that is not a normal value and see if it does change that and normalize it. So you’ll be able to sort of cross check like you would in any other campaign in Marketo. But just ensuring those little signs are on. So yellow globe with a cog. We know that it’s firing. It’s probably your best bet initially. Yeah. Great. Before we wrap, I want to give you both a chance to say any final thoughts, any golden nuggets that you want people to remember from your sessions. So, Malina, do you want to go first? Anything that you want to share as a last minute reminder? I think it’s test, test, test. Yeah. And always do a pilot. Look at results and scale up. That would be what I would advise.

Awesome. And Mina, for you? I would say look at your instance, check your error log, see what’s going on. If you’ve got a lot of errors and things going on, definitely consider changing them over to executable campaigns, popping in some checks for data to ensure that there aren’t any clogs in your database that you could be fixing in a timely manner ahead of your next sort of stages. And I think don’t hold back on trying the executable campaign. There’s a lot of smoke and mirrors around them, but I really think they’re quite easy to use. And once you do a very simple use case, you’ll be able to see that they’re quite powerful, that they work similar to every other part of Marketo. So don’t be afraid to sort of give it a shot, test it out and also to sort of go beyond operational campaigns and think about actual campaigns that are going out to clients, to donors, and to possibly follow up people in a way that is more accurate. I think at the moment, executable campaigns are used really well in the operational space, but they can also kind of expand on to things like follow ups. So use it as a bit of a checker. So is this data value updated? And then move on to that email prompt or that SMS. So just using that for a bit more accuracy in your campaigns. Give it a bit of a go. And yeah, that’s about it.

All right. Well, thank you both. Appreciate you joining us on this year’s Skill Exchange. We will see you all again soon. See you later.