Salesforce Marketing Cloud Campaign Attribution: MarCloud ROI Framework

So just to get into this, Tom, I think a really good question to ask you would be why the ROI framework? Why is it important? Why do we think our customers need it?

Good question. The first reason why is we need a compass to sort of direct us. So the reason why we came up with this framework is to give marketing operations professionals a direction when they are on their market automation journey. So it gives them, no matter where they are in the journey, they can say, right, we are headed towards this particular destination which lends itself to the modular approach of the framework. So it doesn't matter if you're a beginner, intermediate or advanced, whether you've had Salesforce for 10 years or if you've just implemented it, the goal here is that you can apply the framework to wherever you are and you'll know what the next steps are are, whether it be short term or long term. And then importantly, it sets the destination so it gives you a target to aim for, it gives you the ability to track progression. So we work with a lot of clients who historically they've said we're using it as a glorified mailchimp. They might be sending one email every now and then and this framework allows them to zoom out, see the bigger picture and see where they could actually get to with regards to marketing automation rather than just using it as a one off email platform. And that applies to all of the clouds that we're talking about today. So this is the actual framework itself in a nutshell. So it's nice and simple. There's five phases. I'll run through it briefly now before we go into the details of it. But in a nutshell we have phase one, which is very much the implementation, the actual setup of whichever platform it is, data cloud, marketing cloud, etc. And here we want to be making sure that we're set up for success with GDPR compliant. All of the systems are talking to each other so that we have a well connected and up to date system. Then you can start to look at phase two, which is very much the data. So there's two parts to this. The first one is insights. Are we collecting and accruing information about our prospects in terms of engagement. So whether it be website engagement, trade shows, attendance, PDF downloads, PPC clicks, et cetera, we should be collecting that all and maintaining it in our systems that we can use it and fuel our campaigns later on. The data readiness, the second part to this is very much the data hygiene, data cleanliness. So we're avoiding duplication, we're making sure. That we're actually confident in our data so that we're able to fuel our marketing campaigns in phase three, which is very much the doing, that's the ability to send emails, automate campaigns, nurture people at the right time and making content very dynamic in some instances and very sort of personalized. Then phase four is the interesting part because this is where you are first of all surfacing all of the performance data from phase three, you are tracking revenue and roi, so you're able to track each marketing touch point and the impact that it had on the ultimate goal, which is sales and revenue. But then you're harnessing and you're using that data to go back to phase two and improve and optimize the data that you have within your instance. So you'd have that sort of feedback loop implemented to further excel your marketing performance. And then finally phase five, which is a relatively new concept. It's the idea of having that agentic layer on top. So whereby you are, you have a team of three, but you feel like a team of six because you're far more productive. You've got agents supporting you, doing the boring mundane jobs day to day so you can focus on the more sort of tangible, impactful efforts. And I believe we have a poll.

Yeah, let's. If you go down to the bottom of the screen, there is a section for the polls. The question here is which stages do you not have 100% confidence in just now?

So of the five phases, you should be able to click more than one, if not pick the most sort of questionable phase and then we'll see what the results results are. So insights and data readiness, which is actually quite common, we find a lot of clients are not confident in their data. They don't know if it's clean. They might have multiple systems, you might have data silos, so that. That is relatively common. And then agent force is up there as well, which is to be expected because it's quite a new thing. Amazing. We'll still collect those results as we go through and I'll press on. So looking into sort of phase one, so implementation integration, what actually is this? So this is laying the technical foundation to support you moving forwards. And there's a lot of this that does play into how you can actually achieve sort of campaign attribution. So it's quite an important step and some users will naturally know what I'm talking about. They'll feel like something's off in their instance and they kind of get the concept of something not being set up. Correctly, the first question to people to ask themselves is, does your tech stack align with your overall marketing strategy? Now this is important because sometimes over a period of time you can kind of be hindered by the technology to a point where you are led by the technology. You kind of think, well, it can't do this, so you have to adapt and you have to change tack. And the technology ends up informing the strategy, which is not where we want to be. So we need to be asking, do we feel enabled by the tools that we're using? Have we correctly translated the wider marketing strategy into the systems language? That's quite an important step. And creative marketers tend not to be technical. So there can be sometimes a sort of a friction there between the plan itself and getting into say marketing cloud. So that then leans into is marketing cloud, sales or data cloud implemented correctly? Now what does correctly mean? We're talking about no sync errors. We're talking about an easy, seamless campaign creation process. So it should feel relatively easy or straightforward to create a campaign and in some cases this will be documented if there's anything specific going on. And campaign hierarchy is really, really important because this will subsequently allow you to track ROI on marketing touch points later on. So that's a fundamental piece that you need in place to get to the reporting phase four later on. Then we'd ask, do all your systems speak to each other? So this is again another really important step. We don't want data silos. So as a marketer you should be able to see the full customer journey. You should have the data at your fingertips and you should have a clear database of record. And what that is is a, it's a GDPR compliant sort of topic. It's where you have your single source of truth. In other words, you don't want people opting out over there unless it syncs over here because otherwise you'll have two different answers and then you could potentially become non compliant. And then lastly, we want real time prospect engagement data. So this is collecting all of the different engagements across all of your different marketing touch points that's going on.

So Tom, just a quick question on this. What would you say in your experience of implementing this technology, what are some of the challenges that customers see along this journey? Phase one, so to speak.

Yes, it's typically, it's connecting the dots. So it might be that, you know, there's lots of fragmented systems or SaaS companies, it's pretty common whereby you're trying to connect this new tool with an existing tool and there's some things that just happen in real time that you, that you need to bring it across and in some instances, some instances that can be quite confusing or can get confusing if you don't map it properly. So it's always important to have like an implementation partner that can sort of guide and hold your hand through that to ensure that all, all the dots are connected and everything's talking to each other.

So the devil's in the detail as with anything.

Correct? Yeah, yeah. So we use sort of data flow diagrams and sort of lucidchart to map everything and connect the dots which is a really important step and for future alignment so you can understand where you were at a point in time.

Awesome. Thank you.

So in this phase the success criteria very much is a single source of truth for all your data of missing carriers or discrepancies. You want to make sure that you're on the latest sku. So feature enablement plays a part in this. So we can always re review this step. And then lastly, compliance with GDPR and tracking best practices is super important and cookie consent comes up a lot. It's a detail and part of this phase. So here's just some typical actions, some jobs to be done as part of this phase. And as we look at implementation itself, business unit design becomes quite important for larger organizations, especially global the database of record IP warming quite a complicated topic and if you get it wrong it can be quite, quite damaging as part of the implementation phase. AI enablement, integrations, asset creation. So having the templates you need to create a good campaign, email templates, forms, landing pages, etc. It's all important as part of the implementation phase, which then leads nicely into phase two, which is very much about the data. So again, two parts to this. There's the insights, collecting behavioral data and then there's the data readiness which is very much the cleanliness. So again, the question we like to ask is do you see all key data about a person that you need in real time? So if something happens over there, whether it's a sign up, a renewal, a customer complaint, is that visible to everyone who needs to see it? From a marketing perspective, we would be very interested in sentiment and customer history, etc. So something we look for is no duplicates. We don't like duplicates because what that subsequently means is that your reporting will more than likely be off later on because you're doubling, quadrupling up on reports. We would like clearly defining segmented lists. So easy list creation based on Personas and based on your Audience. There's the engagement side of things, scoring and grading where appropriate. That's a really key one for generating MQLs and to be able to then automate lead generation for sales and then granular UTM tracking. This is a really good one. We've got solutions for this. We can be really, really granular with first touch, last touch and everything in between. And we can create reports based on the individual ads if we want, or the individual UTM and parameter. And you can get quite clever with that sort of thing, including lead source and then lastly the Personas. So Personas are quite interesting because this is your target audience and the more we know about a particular Persona, the smarter we can be with campaigns and the more we can do better marketing. But it's important about how you translate that into marketing cloud, if that makes sense. The success criteria for this phase is very much confidence in the data, confidence in automatic Insight collection, clear ICP, lead source being tracked with UTMs and having a scoring grading model that makes sense.

Can you just, can you give me an example of an organization where we've helped them to do this really well?

Yes. So we've, we've done lots of work around Personas and scoring and grading, et cetera. But I think one of the interesting ones was an organization in the US where they, they needed help structuring their, their data in, in line with how they were growing as a business. So, so they had a, they were very mature in a particular state in the us. The rest of the US was slightly less mature and then everywhere outside the US was kind of new business. So we structured their campaigns and structured their data in a way where we were able to report on quite accurately the performance of that particular state, which was sort of quite, quite mature. So we're looking at, you know, retaining clients and we were able to marry that up separately from the PPC marketing trade show performance out in Europe, for example, because they're completely different apples and pears comparisons type thing. So we helped them set that up with a strong campaign hierarchy and customizable dashboards.

Awesome. Thank you.

So, tasks and jobs to be done within phase two. It's very much scoring and grading. It's what a lot of marketers like to see. Data duplication. What we don't like to see UTM implementation for tracking dynamic content. We can start to set that up during this phase because we understand if the data is, if we're confident in the data and we can see what people are interested in, we can then be clever with how we use dynamic content and we actually do this on our own website. So if anyone ever has ever been to the Markloud blog, we have a sliding form on the side of the blog that actually changes based on whether you work at Salesforce because they're our partner. So the messaging changes for them whether we know you. So we've got a little customized message for if we know you and we have a different message for those who don't know. So we actually personalize the website ourselves. Real time notifications. If you're in a competitive industry or you need to react quickly to sales, then we can implement automation around that. But we need to be able to be confident in collecting the insights to be able to make that work. And then lastly there's the unification of data across different systems. So again, having that single source of truth. So then phase three, that's the actual doing. So this is very much sending the right message to the right people at the right time. Basically the fun stuff. So the question here is, is data powering your marketing journeys? So are you taking a data driven approach to your marketing automation? And you'd use the insights gathered from phase two to inject prospects into particular journeys. We'll get onto the idea of content stages in the next slide. Something's a really nice concept, but what you can do is you can gauge where someone is in the funnel based on the type of content that they're engaging with and then you can put them into relevant journeys off the back of it. So we can also do a B testing that's a really, really powerful one for optimizing journeys. And we've got more about that in the agent section because there's some clever things we can do there. The omnichannel approach. It isn't just email that we're working with, it's SMS, it's WhatsApp, it's push notifications, events, et cetera. And we want to be able to track all of it and report on the ROI of each of those activities. Automated lead assignment plays a big part in this. Again, we've already looked at phase two, we're confident in the data. So now we can create automation to then say when someone meets a threshold, we can then pass them to sales as an MQL Smart campaigns. So leaning on the Einstein AI features like send time optimization, that's sending an email when they're most likely to open. There's the SDR agent. We use this ourself ourselves. We've actually started to get results off the back of it. Which is really cool personalization on the website I've mentioned. And then content stages, which I'll show you in just a moment. So the success criteria for this particular phase is very much automated journeys aligned to content stages. So we actually have automation set up, automated journey set up that are producing results. We're increasing conversion rates and we're leaning on the features that Salesforce provide, like Einstein, etc. So jump into this nice image here. This is the idea of content stages. So in a real, real simple way, this is very much grouping and creating content to align with the marketing funnel. So stage one content is very much brand awareness, it's very, very light content. Whereas stage three is a little bit heavier. It might include pricing, it might include stronger call to actions, you might even call people off the back of downloading a piece of stage free content. So it's just differentiating the content that you have and understanding that if someone is downloading and consuming this content, this is what it infers. So you're inferring whereabouts they are in the marketing funnel off the back of their activity. Which is why phase two is so important because this is the alignment to that understanding, if that makes sense.

I think we may have another poll going live now which is have you structured your content to align with the customer journey? The answers are yes, we have a clear view. No, this is a brand new concept for us. We have some parts of this covered with some gaps.

No one said yes yet, which is great because if you, if you start to just think about content in this way, I think it starts to clarify how you think of nurture journeys. It isn't just one route for all. People can enter the system at different points in the journey. And you just need to very quickly identify what is going to resonate most with this particular person. And I bet you will see more results off the back of it. Even just this concept alone.

Just over half the people have some parts of this covered with some gaps. No yeses.

It's good to see. It's good to see. So this phase, actions, jobs to be done, triggering journeys based on buying intent. You're looking out for certain behaviors and then you're injecting people into journeys off the back of it. You're implementing a B testing because you want to learn more about your, your target audience and then feed that back into support and fuel your campaigns moving forwards. There's the automated lead assignment. So generating MQLs for sales, not just using email, using WhatsApp, using SMS, etc based on where your audience Live and what they use, Einstein features and personalization. So everything about sending the right message to the right people at the right time and then onto the interesting phase. So for me this is probably the most important point because this, this goes in two directions. So this is all about making data driven decisions. So the question to ask yourself is, can you confidently tie revenue back to each marketing touch point? If you can, great. If you can't, that's where most people sit. And this is kind of the goal for most marketers. So what we mean by this is having a multi touch attribution model set up. So this could be first touch, last touch, even touch. You might use AI Einstein to figure out the model for you. But this is how we're able to apply revenue to a particular thing. It might be a paid ads campaign, could be a trade show, could be an ebook. It could, as I'm going to show in the next slide, could be an agent. We could work out the ROI of each thing. Now this is really important. The implementation phase is where we set this up. So I referred to it earlier in the campaign hierarchy, this is where the framework becomes quite useful. If we can see parts of this or we can see some reporting, but we can't quite confidently tie revenue back to a particular marketing touch point, we go back to phase one, do the sense check. Okay, campaign hierarchy hasn't been set up, let's work there and move forwards. So this is where the framework becomes quite useful as a compass. So we want to see dashboards set up to be able to prove revenue. So marketers don't want to be sharing opens and clicks, et cetera. These are the old, old metrics. And now, now we're in a position where marketers are really keen on proving roi because then you can get more budget to do more exciting things and place it in the most effective areas. Basically what's working, what's not working in simple terms. And another thing that marketers really, really love or would love is the ability to see all of the MQLs that they generated and to then see what happened after. And this is really important because for a lot of marketing, marketing teams they create, they put all this effort into, you know, generating MQLs, they go over to sales and it's almost like it goes into like a black box. They never kind of lose sight of them at that point and they don't know what happens next. Did it go on to an opportunity? Did it, did it lead to revenue? We don't know. It's like an invisible wall. So it's really important that we fix that. Marketers are able to actually track their efforts once they go over to sales and then visibility on lead source. So then going back to the beginning and saying where did this lead actually come from? An example I used to like giving was if you had a hundred leads from LinkedIn and 1000 leads from Facebook, which one's better? Well, it's impossible to say if you generated 500,000 from those hundred leads on LinkedIn, but then 10 pounds from the Facebook thousand, you should put more budget into LinkedIn. It's not always the first number that you come across, the amount of leads generated that that's the most impactful. And these reports that we set up can highlight and surface this information.

And.

Then it moves on to like the feedback and optimization process. This is that arrow when it goes back to phase two. So how are we actually learning from what we, what we're doing? So we report on things, we learn something, is it being applied and then implemented back into phase two and three? So this is using campaign performance generally on each of the campaigns to then further optimize everything else. So we're constantly learning as an organization and it could well be that we learn more about a particular Persona. It could be as simple as, I don't know, finance professionals prefer more text heavy emails as opposed to images. Little details like this that we can then go back to our campaigns and iterate and improve each time. And then we can improve on segments and targeted communications based on prior learns and a B testing. So we've got an idea, got a hypothesis, we're going to prove or disprove that with a B testing.

As a marketer, what would you say is the single most important view to prove back to the business or yourself?

It's all, it's always revenue. A marketing team that can say we did this and it led to X amount of revenue is always going to be the highest performing team because it's the firstly the confidence in the data and then secondly it's just our activities are working. I've been to events before where marketing teams have been referred to as the coloring in club. Highly insulting, but it happened. And that's not what marketing teams are. It's we are the essentially building the pipeline and we are generating this demand for the business. So it's super, super important that we can actually prove it with dashboards and with actual revenue.

It's certainly something we're seeing out there in the field is people want to prove what's successful for Them.

Yeah. And with Salesforce, with this framework, we can get to that point. And that's the most exciting thing for me to share is there is a framework that exists that can get people to do that point. So the success criteria for this particular phase is very much eventually you have full confidence and visibility into marketing ROI in terms of revenue generated and MQL progression. So you can see where the MQLs have actually gone. You can attribute revenue back to the campaigns and every touch point, however you structure it. So you can actually say this PPC campaign was successful because. And you can actually back it up with data.

We have a question here. Oh, finish that point and then I'll jump into the question.

Yeah, no worries. The last point is the ability to answer granular questions. So you've got the dashboard or interactive dashboards to be able to then answer curveball questions. So if you're in a leadership meeting, someone asks something unexpected, you're going to go to your report, play around with it and you've got the answer there. And then if you don't have that, then we need to build it because you need to have the information at your fingertips.

So we've got a question from Thomas which is, if revenue would take one year or more to come, long sales cycle, how would you do that? I assume the question is, how would you track this? Thomas? I'll let Tom jump in.

It's a good question and actually it's quite similar story to us. I mean, sometimes we've spoken to clients before, even at Marklab where they've known us for a year or even two years, it's a much longer cycle. So there's a couple of parts to this. Firstly, content stages need to align to this. So all of your nurture journeys should really be over the course of a year. Because we see some clients, they've run campaigns over two, three months and it's a much shorter cycle which doesn't align with the bigger picture. So it's a, it's, it's, it still leans on the framework. And so what you would do is you'd say, right, I know that it takes roughly a year for someone to convert and that informs some of the decision making back to each of the phases. But the approach is the same. You'd still assign people to different campaigns and different touch points. You still track utms. Your campaign attribution model would need to align with the year long approach, so you would have a campaign influence window of a year or a touch more so that you can make sure that within the last year we know they went to a trade show, downloaded this, attended a webinar and clicked this paid ad. And then we'd attribute over those particular campaign based on the model that you're using. And so it's actually, it's quite common that buying cycles are a year or more at least for some of our clients.

Awesome. Thank you, Tom. Hopefully that answers your question. Thomas, if there's anything else, feel free to pop it in the chat.

Amazing jobs to be done. So within this phase, attribution modeling is a big one. That's where all marketers essentially want to get to having some somewhat of a real time MQL progression report. Imagine you can just jump in, you turn up to work one day and you can see exactly what happened to all of the MQLs that you generated last week. Persona Dashboards is an interesting one, is to have an actual dashboard dedicated to your Personas. So what are we learning about them? How are our campaigns performing against that particular audience and that segment so that we can iterate and improve moving forwards. Maybe do some A B tests, there's some interesting things there. And making your dashboards interactive, that's quite key. Last point, understanding not just what is working, but what's working really, really well. Because if you can then go back to leadership and say we need more budget, why? Well, because this is doing really, really well. It's generating this. It then helps the, helps the conversation cool onto the exciting side of things, which is agent force. So this is like I said earlier, a team of three feeling like a team of six. And it's all about productivity. It's all about what could an agent do to save you hours in the week. And we look, we need to think about the really boring, mundane, unsexy tasks that are really arduous. And the reason for that is because if the agent is busy doing those tasks, you could be doing something else. So just some basic ideas we've got. Segment creation. Could it help you create campaigns quicker? Could it help you outbound and help you target the best lead? When I say the best leads, it does all the calculations based on historic performance and then chooses the best leads for targeting. It could be Sentiment analysis is an interesting one. Understanding if someone's going to renew, going to attrit, you can kind of figure that out based on the sort of tonality of how someone's communicating. And then some interesting areas are AB tests. Again, people are okay at understanding sort of data with sort of lots of different variables. But an agent will be Way better. An agent can dig into the A B test and understand the Persona, the country, if there's any sort of cultural differences in terms of the communication as a business practice, in terms of regional expertise, seniority, etc. And it can calculate this is the outcome of that A B test. So realistically, what we're looking for in terms of success within this area. Sorry, James. Is marketers leaning on agents day to day in their workflow, Saving time, that's the main thing, being more productive and then ultimately sort of generating more revenue with fewer manual tasks.

That's really good. And, and this is now. And this is a topic that's at every board level in the world right now. Where do you see the nirvana for agent led work and AI?

Well, actually the, the, the future goal would be agents talking to agents. I spoke to Anton about this where there's going to be a plate, a point in time where agents say a salesforce agent might be working with a Google agent and they're just handling a task outside of any human interaction, which is going to be quite interesting. But for now, for today, the goal very much is to just do the arduous tasks. If it's boring, if it takes you a lot of time, if it's mundane, it should go to an agent and we'll be able to figure out the ROI on that. If an agent can save you half an hour a day, maybe it's not worth investing in, but if it saves you two to three hours a day, that's a little bit more interesting. So we need to look at the use cases and move towards a place where we can actually just make teams so much more efficient and way more productive.

Awesome. You'll have to bear with me. My, my technology is letting me down, so forgive me.

No worries. So we did write a blog on this. You can scan the QR code or just go to the website and look for it. But just a couple of examples of how you can use agent force in the real world and how you can customize agents to basically support your sort of day to day. So feel free to check that out. I'll leave that up there just another second more and you can read that in your own time. But jobs to be done here. Creating campaigns, making it easier to create segments, doing sentiment analysis, coaching your sales team lead assignment and escalation is a good one. And just analyzing lots of data at one time with an agent. Cool. So the outcome or the call to action, I guess for our webinar is to request a free account Audit and we'll let you know what phases that you can focus on. It's completely free. We'll do an assessment of your account. It takes up to two weeks and we can review any of the platforms on the screen. Sales cloud, marketing cloud, data cloud. And we'll basically work with you to help you navigate the marketing automations landscape. We have put a poll out. When was the last time you audited your marketing and sales cloud instance? It should be every three months is the answer. But it'd be interesting to see when was the last time you audited your account and you can answer that in the poll.

We've got a few people who weren't aware that they were supposed to order their account. Absolutely fine. We see it a lot. 612 months 1.

So the important thing here, so if it's anything over six months, what may start to happen or what you might notice is that the technology can sometimes then start to lead the strategy. So you're no longer, you're starting to leave that timeframe where your strategy is really clear. Everyone knows it in the marketing team, everyone's aligned. But now it's sort of drifting away from what's actually happening within the technology. So super important to get an audit booked in just to sort of realign to make sure that you are heading towards where you need to go. So yeah, super interesting. We do see it a lot where people haven't audited it in a. In the last year. Amazing. So with that in mind then we can answer any questions, if anyone has any, and you can just drop them in the, in the chat. And if there are none, we can.

A question from Jimmy. Hey Jimmy, how you doing? Would you differentiate between marketing generated revenue and marketing influence revenue? If so, would you use campaigns to measure the generated revenue?

Amazing too. I would, I would measure both. It would kind of be both because you'd still use campaigns to track them, if that makes sense. You'd still lean on the campaign structure. Marketing generated revenue. It'd be interesting to know your context because the influence model is very much marketing influence. Anyway, what I'm reading from this is potentially there could be an instance with this particular business where the marketing team can go direct to a sale. Sometimes this is common in SaaS for like subscriptions, et cetera. So it might be that marketers will go to events and go and send out emails, et cetera, and say click here to sign up and create a free account. And then they do where there's actually no sales involvement. So in that case that would be marketing generated revenue, whereas marketing influence.