The ‘AMP Up Your Digital Marketing’ Podcast Presents:

Marketing Budgets, Delegation, and Learning the Attribution Model

Shane Murphy-ReuterWhere do I spend the money? Is this campaign bringing prospects to the finish line? Is my talent pool established and who do I need to hire to hit this big, hairy, audacious goal in three years? All valid questions, and all questions your marketing leaders are likely asking themselves. If there was a SaaS tool out there that could perfectly predict a multi-touch attribution model, and tell you when to hire and fire, and where to invest your budget, it would be a unicorn overnight. Guaranteed. 

Of course, this doesn’t exist because figuring out how and why people buy is hard. These days there is so much direct and indirect noise in the purchasing decision that it is virtually impossible to account for and determine every little variable in the attribution model. Sure, we can make reasonable assumptions -- traffic jumped after this podcast aired, so it must be the podcast -- but they are nothing more than educated guesses. But educated guesses doesn’t get the budget - a solid foundation does. 

In this episode of AMP UP Your Digital Marketing we meet Shane Murphy-Reuter, the Senior Vice President of Marketing at Intercom, a customer messaging platform. Through depth and breadth of experience, and making some mistakes along the way, Murphy-Reuter gives us insight into the ‘three buckets’ he follows for marketing budgets, delegation, and learning attribution models. His actionable tips will help you answer questions such as:

  • How Do Marketers Spend Their Budget Wisely?

  • When Do Marketers Validate Their Hiring Decisions?

  • How Do You Delegate the Marketing Tasks?

  • How Does a Marketer Learn Their Attribution Model with Imperfect or Incomplete Data?

  • How Do Analysts Affect the Non-Directly Attributable and Not Direct Response Efforts?


Glenn: Welcome back to the show. Today we're speaking with Shane Murphy-Reuter. Shane, welcome to the show. 

Shane: Thank you, Glenn. Really happy and excited to be here.

Glenn: Shane, tell us a little bit about who you are and what you do?

Shane: Yes, so, obviously, I'm Shane. I run marketing right now at Intercom. I've spent probably the last – God – maybe 15 years working in marketing, variety of roles. Started probably the first-half of my career working in proposition development which is now called product marketing and more in the brand side as well. And then shifted, sort of, the second-half of my career towards more of the demand and now kind of running all of that at Intercom within, sort of, the B2B space. And so, I've had the pleasure to have a pretty varied background in marketing and really glad to be at a company that really cares deeply about all the different areas.

Glenn: And just to give people a sense of the size of your marketing organization today, can you give us some rough numbers?

Shane: Yeah. So, the marketing team, once we are fully staffed, will be 80 people. We are going into a hiring drive at the moment. So, anybody out there who is in marketing and would like to join Intercom, let me know. And yes, we're about 80 people. And, you know, we have a pretty typical split of the team across product marketing, brand, demand-gen, content. Those are kind of our four key pillars of the team. And then from a budget perspective, you know, we have over $10 million in actual dollars that we can use to execute marketing. So, like, media buying or marketing technology cycle, that sort of thing.

Glenn: And so, when you think about that, I mean, it's a sizable budget, it's something that you really have to think about, am I putting this to best use? So, what do you think about that and how do you even get your team to think about, you know, highest and best use of the dollars and the resources that you're allocating?

How Do Marketers Spend Their Budget Wisely?

[Minutes 2:12 to 3:50] To have a successful marketing organization built up, you need to have three core pillars:

  1. Demand Marketing: What do we actually want to say about our product and how will that resonate with the customer?

  2. Brand Marketing Team: Add the creativity, the emotion, and the spark to get people to actually care.

  3. Demand Gen: Take these great fundamental messages and can put them into marketing campaigns to drive reaction and conversion to customers. 

Shane: Yeah, definitely. I would admit that I probably spend most of my time over the last few years thinking about this question in differing ways. You know, and maybe I'll start with the team structure and how to figure out where to invest. Like, there is no one size fits all but I will say that generally speaking, to have a successful marketing organization you need to have three core pillars built up. One is your demand marketing team whose job it is to figure out, you know, what do we actually want to say about our product that will resonate with the customer?

The second team then is the brand marketing team who takes that and says, okay, that's great, totally in summary we want to say, how do we actually get people to care? Right? So, they're adding creativity, they're adding the emotion, they're adding the sort of spark to your marketing that makes people sit up and listen to the story that you want to tell that's been defined by product marketing. 

And then the third pillar is demand-gen whose job it is to say, oh, great, we've got this, like, great functional message about the product, we've got this great emotional message or sort of, like, creative message that the brand team have pulled together, now we actually have to put that in marketing and use demand-gen channels in order to get that message into the market so that people react to it and convert as customers.

So, you know, fundamentally, as you grow the team, you want to make sure that you are trying to balance investment of those three different things. And, like, different companies will, I think, have different philosophies on those three pillars. You know, I started my career in the consumer world where if you're selling perfume, you don't really need much product marketing. You know, it's like, much more of a functional message to say. Nobody is selling, like, this perfume is, you know, healthier than that perfume. It's 100% all about brand advocacy and the emotion, right? And so, if you're selling a product like that you would invest more heavily, obviously, in that brand pillar relative to a product marketing pillar. However, if you're traditionally especially in B2B, and B2B enterprise particularly, selling, you would have a much heavier investment in product marketing because you're going through a pretty detailed sales process. Once you're in that sales process, you know, those potential customers want to know truly function here, what can this do for me? Which is much more of a product marketing ask.

And then, you know, demand-gen, your relative investment there, I think, you know, plays off that a little bit depending upon if you want to go and build the scale of brand, you're going to need to invest money in raising – in channels to actually raise that awareness to get these in the funnel, whereas, typically more B2B type companies, and certainly, again, B2B companies that are more in the enterprise space, can have a more of a direct selling go to market motion, in which case, your investment in typical demand-gen channels doesn't necessarily need to be as high. Like, those companies will have a higher investment in sales relative to marketing.

And so, you know, the first thing that I try to do is really think about those three pillars of product marketing, brand marketing, demand-gen and depending upon the company that I'm at, and figuring out the unique, kind of, the needs of that company, design the team depending upon the unique needs of the space.

When Do Marketers Validate Their Hiring Decisions?

[Minutes 6:15 to 7:10] As part of this process, you need to check the load balance. How and when do you do that?

  1. Plan at least a year ahead. A really good marketing plan needs at least 9 months of lead time.

  2. Have a 2-3 year plan for execution.

  3. Execute at a quarterly level.

Glenn: And so, how often do you check and validate the decisions that you made, do you come back quarterly and look at that and potentially load balance or do that annually? What's your approach there?

Shane: Yeah, you know, I think typically for marketing, we need to plan at least a year ahead. You know, if you want to do impactful marketing, the lead time for really great marketing campaign could be nine months. Especially if you're going to get an agency involved, you know, all the media buy-in and all of that. So, you definitely need to think at least a year ahead with probably a two to three-year general idea of where you want to go. And then we execute at a quarterly level, so we will have very detailed plans at a quarterly level on how we're going to actually execute and what we're going to deliver. And so, it's kind of like, I think, longer term, over two, three-year horizon. In one-year horizon, I think about what are the big initiatives that we have, the relative investments across those initiatives. And then at a quarterly level, we're actually executing.

How Do You Delegate the Marketing Tasks?

[Minutes 7:11 to 13:17] With all these plans, budgets, and tasks, do you just hand it off to the head of each department or actively jump in and try to help?

  1. It Depends on the Marketing Leader: Some are more extroverted and want to whiteboard it out with you and others like to take it away and think about it.

  2. Strong Pillar Leaders: Across each of those core pillars, you want a leader in place that is significantly better than you are at that core competency. 

  3. Understand Where to Farm and When to Sow: Understanding share of voice and the investment necessary to maintain share of voice helps to delegate where funds and tasks can be distributed.


Glenn: And so, how do you then take that approach and think about your team itself? So, you've decided what your buckets are, your three different buckets, and you're going to put so much into this bucket and so much into the other bucket. How do you then really take it down to the next level? Do you just hand that off to the head of that particular group and say, figure it out from there or do you get actively involved in trying to figure that out?

Shane: Yeah, I think that kind of depends on the marketing leader. And I'll admit that one of my personal, sort of, areas that I always need to make sure that I'm not doing too much is the risk I'm not delegating enough. And certainly, like, my natural position is to want to get into the weeds with the team and get in the room. I'm also like an extrovert, right, so I love, let's jump into the room, get a whiteboard together and start figuring this out. And depending upon the leader in my team, that can be beneficial or the complete opposite. Like, if I am working with somebody who I know likes to riff off stuff then I get into it with them. And if they are however, somebody who wants to take it away, think about it, I'll do it that way. So, I think it's about, you know, adjusting my style and how I'm involved to how they like to work. But typically speaking at a team of, kind of, 80 people, you want to have, across each of those core pillars, you want to have a very senior leader who is significantly better than you at that area. So, like, I would not hire myself to lead our demand-gen team, I'm not enough of an expert there, same for brand team, same for product marketing team, same for our content marketing team. And so, you know, I have to trust those people.

Having said that, there is, like, certain decisions I would sit at my level that I need to make sure that I do with my leadership team. And, you know, maybe I'll give a good example of a story where, I think, we got this wrong and then got it right. When I was working at Paddy Power which is our European online gaming brand. Traditionally, Paddy Power had very much been a brand invested in brand marketing. Our marketing budget was about €50 million a year. So, pretty scaled. And we had – God, I'm trying to remember the exact year – it must have been 2013, we had a really strong growth year. We had just started to get our demand-gen machine going and it was like we were growing exceptionally quick. The year after that, our growth grinned – not to a complete halt, I think we grew 4% that year and we couldn't figure out what was going on. You know, we were doing all of the same things, we couldn't figure out what was leading to that decline in growth.

And what we realized what happened was that, you know, I think about demand – like, brand building is like you're essentially sowing the seeds, right. You have a farm, you're planting the seeds, the crops are growing and that's your brand, you need to invest in that. And the demand team's job is like, they're like the farmers, right. They're going to come in and they're going to say, oh, look at all this great, kind of, brand awareness we have, we're going to go and convert that. So, they cut down the crop and they become revenue. And what we ended up doing was, we had years of this great investment in our brand and when we built up the demand-gen team, it was like, you know, these farmers going to town. And just like, there was just fields of crops everywhere for our demand-gen team to convert to customers. And so, what we ended up doing is, over-correcting and we pulled the money away from the brand and into the demand-gen team. And over time what ends up happening is that, suddenly you've got all these farmers and no crops left.

And so, in – I'm trying to remember the years. So, the first year, obviously, we have this great growth year. Second year really slowed to a halt, because we over-corrected. And then into the third year we rebalanced to get to a nice, healthy relationship between that brand and demand investment. And so, like, at my level, that's the type of thing that I need to be thinking about. And there are various methodologies that you can think about measuring your demand-gen impact versus your brand impact and are you over or undersaturated.

So, for example, when I was at Paddy Power what we did on the brand side was we did a lot of share voice analysis and we looked at what we recognize that on average, if everyone is equally good at executing on the marketing, if you have a 12% market share, to hold your market share you should have a 12% share of voice. Meaning that you are investing about 12% of the overall investment in marketing in that category. If you spend anything less than that, your share of voice is lower than your market share and that's an issue. And we had dropped from a share of voice in brand – our market share at the time I think was about 12% and we had dropped our share of voice 10 to 5%. And we didn't realize that because we didn't had that type of tracking in place. And so, once we had that tracking in place it meant that we were able to better measure, you know, if one of our competitors start to go really aggressive, we might have to react and invest more in the brand.

And then on the demand-gen side we had a – because it's much more measurable – we had a much more, sort of, technology driven, multi-touch attribution model to understand our relative investment there and what was working or not. So, kind of measuring your relative investment and the impact of it is, like, so critically important to knowing at my level, you know, are you about to make the mistakes that we made at Paddy Power. 

Glenn: It's interesting, Shane, because you bring up a point that a lot of marketers can run into, which is imperfect data or incomplete data. So, it's pretty easy when you look at a sheet of paper or you look at your spreadsheet and it says, okay, if I spend a dollar here, I get X back. But when you have somewhat limited information or imperfect information, how do you go about basically coming up with the best guest or do you basically say, well, we're not going to touch this with a ten-foot pole because there's not enough data to make a rational decision?

How Does a Marketer Learn Their Attribution Model with Imperfect or Incomplete Data?

[Minutes 13:56 to 19:08] You can study an attribution model, and marketers will say they have cracked it all, but with these three buckets you can learn more about how your data behaves:


  1. Directly Attributable Direct Response: We are able to tie this conversion directly back to a part conversion.

  2. Non-directly Attributable Direct Response: Say we buy radio or podcast advertising with a really great offer. We expect some people to hear it and redeem the offer, but we can’t directly attribute it to the activity. 

  3. Non-directly Attributable and Not Direct Response: This is your investments in Comms, PR, or other brand-building, possibly for other initiative but still helps contributions to the organization. 

Shane: Yeah, you know, any marketer, and there are many out there who'll tell you that they have, you know, figured out attribution budget.

Glenn: It doesn’t matter how big your budget is.

Shane: They are talking absolute – excuse my language, they are talking shite, as we say in Ireland. You know, at Paddy Power we spent about three years building a multi-touch attribution model for our digital marketing. We employed DoubleClick and tracked absolutely every click, every impression. I've actually never seen a marketing team with a better understanding of the part conversion than we did at Paddy Power and yet you still need to use a certain amount of judgment. Like, you know, I think about the marketing investment in sort of three different buckets. And this is going to get a little bit, sort of, like, textbook lecture here, there's like – so bear with me a little bit – three buckets-

Glenn: You like three buckets of things, I've noticed.

Shane: Honestly, I remember someone said to me once, if you got to have list it's got to be an odd number. And maybe I just, like, don't have a great memory and so I like to keep my list short. So, like, maybe there should be seven, but I'm like, whoa! So, like, the three buckets would be, directly attributable direct response is one. That means that we are able to tie this conversion directly back to a part conversion. So, they clicked on a paid search ad, they clicked on an email, they viewed this banner ad, etc. So, that's directly attributable. And it's direct response, meaning that, when we do that marketing, we expect the person to respond in the very short-term. And so that – you know, typically like trials paid search would be direct attributable direct response. And the way we measure that, you can rely much more heavily on the data for that type of marketing than you should have a multi-touch attribution model. You should track the path conversion. And in there, obviously, is big debates around, should you use last clear, should you use first touch, should you use linear? The answer really is that, you need to hire a data scientist to do essentially uplift analysis within that. And like, we can talk about that for about three hours. But within there, you can use data. And you can be pretty scientific about your decisions.

Then the second bucket would be non-directly attributable direct response. And what I mean by that is that, let's say we buy a ton of radio advertising or podcast advertising. We would expect that to have a direct response. Like, we would expect somebody to listen to that and let's say, it's got an offering, to directly go to our website and covert. But it's not-directly attributable. Like, I can't see somebody click on that podcast. I can't see somebody click on that radio ad. And typically speaking, in that bucket, you should be measuring based on uplift. And so, you should be able to see site traffic, once you launch the campaign, increase. The difficulty would be knowing which of the podcasts or that media investment led to that, but you can at least see the overall campaign impact. And so, that one is like a combination of art of science, whereas the first bucket was like 90% science.

And then the third bucket is, it's still non-directly attributable and then not direct response. And that would be things that your investment coms, like, PR, and that type of investment where we don't expect somebody – you know, if we're doing some PR, we don't expect somebody to necessarily read that article and immediately buy the product. It might be about building our brand as we, you know, move towards IPO or something like that. There might another goal for it. And we also can't directly attribute it. So, we don't really – you know, we can't track if somebody reads an article on AdWeek about Intercom. And so that bucket is now 90% art and 10% science. You're using your judgment and leading indicator is, like, okay, we've done this big push around, like, you know, conversational marketing as coms focused, are we seeing people, like, on Twitter associate Intercom more with conversation marketing? When they mention Intercom, do they mention conversation marketing? And so, you can use leading indicators like that, but you don't have like that kind of output metric to revenue in the way that you do in the other buckets.

And so, anyway, I got a bit theoretical, but I think about the world in those three and apply varying degree of art and science depending upon which bucket we're talking about.

Glenn: You know, it's interesting, the third bucket, which is the most art-like, contains things like – not just kind of the basic PR of going to the press but also probably some analyst activity. How do you all think about the analyst community? And are you a – let's spread it out as much as we can or let's focus on one or two and really have an impact with them?

How Do Analysts Affect the Non-Directly Attributable and Not Direct Response Efforts?

[Minutes 19:10 to 22:00] Analyst activities can impact your results quite a bit. Here is what we’ve identified to be the most successful:

  1. Mid-Market Businesses and Enterprise Offerings: That third-party verification is key  because it shows community buy-in.

  2. Identify the Most Successful Relationships: Determine where your organization is most effective to seek out and create relationships with the analysts responsible for covering that space.

  3. Smaller Business Differentiators: Being active on G2 is a great way to establish your market presence and a place where many businesses go to learn about the available options in their market. 

Shane: Yeah, we're pretty early days, I think, we've been focused maybe for about nine months on analyst relations and it's connected to Intercom's move upmarket. Like, historically Intercom has very much been focused on building and selling to small business, like, startups and now we are increasingly – there are mid-market and enterprise businesses coming to us to buy Intercom. And because of the strength of our brand, the strength of our product, and we realize that if people are coming to us proactively, we should really do more to sell and grow our demand in those segments. And so, as part of that analyst relations becomes really important because within mid-market businesses and larger businesses like enterprise, there's a much larger buy-in community and analyst and some sort of third-party verification can be a really important way to engage the broader buy-in community. So, like, you know, if the head of support in a company totally gets the value of Intercom, they want to buy it. They've got to go and get signed off potentially from the Chief Revenue Officer and the Chief Financial Officer and maybe from the CEO depending on the level of investment. Those buyers, it's critically important for them to have some sort of third-party verification like analysts.

And to answer your specific question about how broad and deep you go, because we're relatively new to this, we've been pretty focused on identifying what are the categories that we can be most, I think, successful within and then identifying the specific analysts who are covering those areas and creating relationships with them. So, for example, Gartner has been a big focus for us. But not just Gartner, like, specific analysts within Gartner who are covering our area. And then also because of our, sort of, startup legacy, the likes of or previously G2 Crowd, which aren't typical, like, “analyst” of kind of what you might think, they are having a growing influence as sort of a third-party verification. And so, we have a lot of work with G2 as well.

Glenn: Shane, you've covered a number of things here. If we were to boil something down to one thing that our audience could put into action today to really have an impact with either how they're thinking about or how they are doing their marketing, what would that one thing be?

Shane: You know, I think the one thing is the measurement question, because I think the measurement thing was, we talked about those sorts of three buckets and making sure that you have an understanding of those three and measuring the impact of them. That is the thing that leads to your understanding of what you should invest across, sort of, product marketing, brand marketing and demand-gen. And, you know, I told the story Paddy Power how we were popular company at that time and to go to having 4 to 5% growth year for new customers, that's a big mistake to make. And so, you can get it really, really badly wrong if you don't understand that. And so, I think, yeah, really, truly understanding how you're going to measure the marketing that you are investing in and to what degree are you going to use science versus – science of the kind of the outcome of that thing, like, revenue versus using the leading indicators – like I talked about before around, like, association with certain categories, whatever, as your way of measuring. But either way, you should have a way to measure that and make decisions otherwise you just sort of at risk of running after the new shiny object.

Glenn: Yeah, lack of focus can be a challenge. So, Shane, how can people get in touch with you? And let's also, if you can mention, if people are actually interested in one of your marketing positions, where can they go and find those as well?

Shane: Yeah, exactly. So, you can get on Twitter, is probably the best place. My handle is @shanemurfy – it's S-H-A-N-E-M-U-R-F-Y – the PH was taken. So, you can get me there. And then, yeah, if you go to you'll find a bunch of jobs there across the company and particularly in marketing. So, we'd love to hear from any potential marketers out there.

Glenn: Awesome! Shane, thanks for being in the show.

Shane: Thanks a million, Glenn. Have a good one.

Shane’s Bio: Shane Murphy-Reuter is the Senior Vice President of Marketing for Intercom, a customer messaging platform. To reach Shane, you can find him on Twitter.

Want to appear on AMP Up Your Digital Marketing? Contact Us and let us know how our listeners can benefit from hearing your message.

Social Media activity employee advocacy

Ramin Edmond

Ramin Edmond

Ramin Edmond is the former Content Strategist for GaggleAMP. Outside of work, Ramin likes to run, hike, and take pictures of Boston's best views. You can get in touch with Ramin by connecting with him on LinkedIn.

Thinking About Starting an Employee Advocacy Program?