Marketing Attribution: The Journey from Cost Center to Cash Cow

Editor's Notes

• #4: So, marketing has always been the butt of many a sales organization’s joke, and I’m sure we’ve collectively heard them all. T=Even the most recent episode of Corporate Bro, always funny any most often always equally sad, makes light of this contentious dynamic pretty well. Back in the small, quaint startup days at Bizible before we were bought by Marketo and then Adobe, I had no idea what this was about since sales always seemed to be homies with marketing, and we very much had a symbiotic relationship. However, this wasn’t what my sales and marketing buddies at other companies were telling me, and honesrly wasn’t what I was feeling even internally before Marketo implemented attribution themselves. And the reason why is pretty simple – marketing does a ton of work, but they get very little credit. And is this isn’t because sales or the rest of the org is out to get marketers or steal credit, well, hopefully. It’s simply due to a lack of visibility and tracking for marketing’s efforts, and an inability to make those efforts visible using metrics that speak the language of those outside marketing: how are you creating and enhancing our pipeline, and how are you contributing to bottom line revenue. Often times, marketers themselves aren’t even sure how to answer this to themselves, much less when asked from the outside. However, when you have the right data to back up your answers to those questions, the story changes pretty dramatically. And for the better.
• #5: And, enter the lush, revenue-rich utopia that is attribution: the other A word that isn’t analytics, but still involves charts, graphs, dashboards and the like. And we’ll go into what attribution is and isn’t, and how it compares meaningfully to analytics too. But, some key things to understand about attribution is that it’s being able to see everything marketing is doing, online and offline, across all leads and contacts, customers and prospects, opportunities and accounts, and holding each and every one of those engagements and activities accountable to…well, what? Views? Downloads? Shares? Conversions? Clicks? Nope. Again, holding them accountable to metrics that are spoken outside the marketing org is critical, in the vernacular of the purse string holders – metrics like pipeline growth, and revenue. Aka, the same metrics that makes Sales the darlings of the organization. Now, marketing can has revenue and pipeline too. That’s attribution.
• #6: But ok, so you may be asking yourself, what’s the difference between attribution and analytics? Is one superior to the other? Does one replace the other? The answer to both is no, one does not replace nor reign supreme over the other. But, one should never be mistaken for the other. For example, let’s say you are trying to understand how to optimize web traffic and page views on your website. Analytics will tell you that. How is traffic flowing from one page to another, what’s the fallout, stuff like that which conveniently Adobe Analytics is the market leader for solving. But, let’s say your CMO is told that the org needs to slash their budget because, I dunno, a global pandemic hits and there’s tons of uncertainty. Marketing, being marketing and all-too-often the ugly stepchild of the org, is told that only revenue-generating essential activities are going to remain funded. Is your CMO going to be able to answer that question by saying “but wait! Look at all of these page views, or downloads, or content shares, or conversions (whatever that means) that we were able to drive last quarter!” and expect the CFO to take those responses? Maybe you know the answer to this not-so-rhetorical question because you’ve been in these not-so-hypothetical shoes yourself. Conversely, what if she instead is able to say “but wait, look how much revenue we drove last quarter. Or, look at how many MQLs we drove, not to mention opportunities we nurtured to closed/won, or generated, or every single other step along our customer journey, backed up with data and reports inside the CRM.” That’s the difference. And this means that you not only get the coveted seat at the revenue table, you can also optimize your efforts according to those KPIs, so every month, quarter, whatever, those KPIs look more and more impressive, and thus marketing looks more and more impressive across every stakeholder and budget controller in the company.
• #7: But, being able to pull this off is easier said than done. Or, is it? Well, it depends on how you do it. If you’re here listening to me, then I assume you’re clued into why attribution is important for you to understand, and are starting to weigh your options on how to execute. I’d say that’s where many marketers are today in general, since frankly attribution is a newer kid on the block, not unlike where marketing automation was a decade or so ago. But we’ve heard from our customners that attribution today is now a need-to-have more than ever before. Or, maybe your team is a bit further along and you’ve started to put together an attribution play, potentially yourselves, using spreadsheets, having to pull in data from multiple silo’d platforms and third party sources, having to join that data to CRM and marketing automation data, then having to report and visualize it somehow, and then (often the hardest part) having to truly believe your solution is comprehensive with full data integrity, and accurately reflects the real truth of marketing contribution. That’s the funny thing about attribution calls I’ve been joining. They often play out like a therapy session. “Tell me about your attribution strategy.” 30 min later, and they’re crying on their home office sofas, or are ready to punch a pillow.” It’s not only really, really hard to do it yourself, but the results are (at least in my experience in and outside of many companies) ALWAYS dissatisfactory. Either because they feel they can’t track everything marketing is contributing and are getting sold short, or because the metrics they have to resort to are more of those marketing-speak vanity metrics around arbitrary engagement, which may have no connection to what actually moves a deal through a funnel with revenue as the goal. You know, that whole adage of 50% of the marketing dollars are effective and the other half ineffective, but no one can tell which.
• #8: So, just in case you needed another bleak dystopian reality to contend with this year, that’s what we hear marketers are up against. All the worse if they’re trying to expand into channels that are traditionally nightmarish to track using manual objects like campaigns, but that’s another story we’ll get back to. And if that weren’t enough, each persona and specialist in your team may want to consider different metrics relative to their own interests and pay grade. For example, your paid media specialist may want to know what keywords and campaigns are most fruitful to drive MQLs, whereas finance may want to know holistically what is and isn’t producing hard revenue from a higher level 30000 foot view, whereas sales wants to know what’s driving opportunities. Oh, and they all want to see this data in their relative systems of record, which is usually the CRM, or maybe it’s BI too, or maybe both, or maybe somewhere else? So you gotta somehow deliver that as well, otherwise your metrics are no good if the people who need them can’t access them.
• #9: Is there relief in sight? Yes. Attribution tools, and speaking of which Adobe happens to have a pretty good one, solve this in a variety of different ways with different approaches. I’m guessing that everyone here today is familiar with single-touch models, probably because you get that out of the box with pretty much every marketing automation tool – usually in the flavors of first or last single-touch models that attribute to lead creation or opportunity creation. I’m also guessing that everyone here today isn’t totally happy with that approach, and with good reason, since 100% of the revenue credit for a deal goes to just a single marketing engagement. That’s like thanking your entire professional career on a single course you took in college, or a single presentation you gave years ago. Or, pinning your entire personal domestic relationship on a single date. More likely, it was a series of events that took place over an extended period of time across (bringing it back to B2B) a large buying committee of people, involved across various inflection points in that deal. That seems like a more reasonable approach, but it does potentially (and I do emphasize potentially) involve a greater degree of effort to track all of those people, engagements, and times. Right?
• #10: So, let’s unpack some of this. Bringing it back to single touch, just to entertain this as a potential viable option for tracking a B2B marketing and sales cycle, which it isn’t, it does have certain advantages. It’s easy. Easy to implement, easy to understand, super simple to use. Too simple, in fact, and frankly, pretty misleading, as it way oversimplifies the complexities that are more the rule than the exception for a B2B journey - just like saying that in to make a great bottle of wine, you just need some old grapes that have been sitting around for a while. A B2B journey is more like a complex bottle of cabernet, requiring just the right care, touch, ingredients, and handling to bring about the desired result…of a closed won opportunity…ideally celebrated with an equally complex bottle of poison. And while every deal is indeed unique and will follow its own distinct path to winning, there are patterns and trends that emerge across each step of that journey. Patterns that will allow you to optimize by increasing and decreasing your spend at different points and phases in nurturing, depending on what needs to happen to move things along. However, you’ll never know this, if you’re just looking at the grapes and nothing else.
• #11: OK so, out of the dark ages and now into the world of multi-touch. You may as well throw in a greater than sign between the two. [next]
• #12: There we go. So, now we are working in a world that allows us to take off our blinders beyond what’s right in front of us, I.e. just what makes a lead or just what makes an opp. Now, assuming you’re using the right attribution approach, you can see everything that happens long before, and long after those milestones, and in some cases the first time someone from an account hits your website, even if they don’t fill out a form just yet. As if you can’t tell from my less-than-subtle already, this is what you should be using in a B2B or B2C considered purchase sales process. Singel touch works OK with B2C consumers since they’re quick, transactional, and without much complexity. Like a crisp sparkling water, to revisit the beverage analogy, refreshing and hits the spot, though rarely complex (unless you’re talking about topo chico twist of grapefruit, which, as an FYI, you would be drinking in person at a fancy hotel banquet room right now on Adobe’s dime, fingers crossed sometime next year). Where was I? B2B marketing and sales cycles are complex, and the require a solution that can wrangle in all of those complexities, and let you understand what’s making an impact and what isn’t, and where it’s making an impact, from start to finish. Only then will you be able to see what’s blowing it out of the water, and what’s a complete and utter waste of your ever-threatened marketing time and budget. But, there are cons too. Or as I prefer to optimistically say, challenges that need to be considered and approached intelligently and proactively. Which are – you gotta understand the different attribution models, and understand why you’re using whichever one you’re using. Some of them frankly don’t make much sense to use, and employ rather arbitrary weighting which doesn’t quite foot to reason, so you gotta watch out for that too. But - and this is probably the biggest gotcha - there’s the risk that you’re not actually tracking every engagement across the customer journey, and things that are making an impact, both small and big, are still going unnoticed, and thus underweighted, which conversely means other channels are getting overweighted and over-credited.
• #13: OK so, I agree this can seem like a bit of a tease. I would venture to say you may be aware of what you need and where you want to go assuming you’re not already there today, but now the challenge is actually getting there, and doing it right from the beginning so you don’t waste time and money with costly missteps into the wrong attribution solution. Part of doing this right is understanding what really makes attribution work in practice. Yes, you want the data. Yes, you want to see the full customer journey. Yes, you want to have it go broadly as well as deeply, so you can make precise adjustments not just channel by channel, but within each channel. For example, paid media, knowing what keywords are winning and losing. Great. However, since you’ve gone up an order of magniture or two in the complexity of what you want to track, just think how much work it will be to track this – especially if you have to do it manually. And unless you’re a robot, any errors in this will snowball into a cataclysmic data mess that leaves you highly skepti al of the reports it generates. So, I would go so far as to say that equally important to the data is the way in which it is collected. Manually, via say the campaign object, going into spreadsheets, requiring weeks or months to analyze – or automatically, avoiding all of that mess and headache?
• #14: Which is just the right thing to be wondering as we go into some options you have in terms of solving for attribution. Basically, most every attribution tool can be broken down into two categories. Campaign-object-based tracking, and touchpoints. Let’s see what each entails, and what you can expect out of each, as the decision you may here may be the most important one in terms of choosing the right attribution tool for your needs.
• #15: OK so campaign objects. In case you’re unfamiliar, and since campaigns are the most ubiquitous term in marketing, we’re talking specifically about the campaign object in your Salesforce of Dynamics CRM. Basically, a container that you can use to capture a static list of leads or contacts who participate in, well, whatever. In this case, a marketing engagement. For example, and I wouldn’t be surprised if this is how most of you are doing this today, you may host a webinar and receive a list of attendees who attend that webinar, much like we’re doing to you right now. You take that list of virtual badge scans, and upload them into a campaign in your CRM to track who attended. Pretty easy, and frankly, if that’s the extent of your marketing (i.e. things that are easy to track and upload into campaigns), a campaign-based attribution solution will get you by. And they’re cheap, as they should be. However, let’s say you’re a bit more sophisticated, and you aren’t just hosting one-dimensional yes/no/attended events but say, digital marketing like paid media, paid social, organic, seo, gated/ungated content, outbound bdr/sdr sales, etc. Things that have more than just a single dimension of data, and may involve different parameters, UTMs, keywords, dispositions, assets, conversion paths. Sadly, your great hope campaign object has let you down since it can only capture a single dimension of data, once, meaning if you want to be able to not just understand super basic marketing mixes, you’re on the verge of entering child/parent subcampaign spreadsheet manual tracking hell, which makes a cheap attribution option suddenly not so cheap anymore. Your time has a cost too, and probably not a cheap one, and I’m sure that price would increase if you were told you had to spend hours per day and weeks at a time living in manual data tracking dystopia versus, say, being able to use high quality, clean data to make heroic marketing decisions. Which is my theory as to what was really going on with this Adobe stock photo that kinda looks like this marketer is flipping the bird to someone telling her she has to track everything using the campaign object. Simply, unless you’re just doing events (likely illegal renegade ones at the moment) or very basic virtual marketing that a campaign object was designed to track, the campaign object approach becomes awkward, tedious, not scalable, inaccurate, and costly in many ways, very quickly.
• #16: Here’s a few bullets, just in case I haven’t sufficiently forewarned you. Campaign-based attribution: cheap and simple, though unless you can say the same about your marketing mix, it’s probably not a fit. Oh, and since it’s a single dimension, if anyone ever asks you for more than a channel-by-channel attribution questions, such as what campaigns and versions of ads are working and where they’re working, you’re in a tough spot since you can’t drill down into channel performance.
• #17: And if this whole presentation has felt pretty dreary and hopeless so far, that’s because frankly I’d rather you felt the impact of poor attribution during a 30 min webinar than over a year of a poorly choosen attribution vendor contract you’re trying to escape from. Unless, again, if you’re using very basic marketing channels that can be tracked using campaign objects, which at this point you can probably bounce out, and watch the last 10 min of this later when you start to grow into digital. But, if you’re already there or just starting to dip your toes into digital, I’m going to finally let the clouds part and give you a very viable, very scalable, and very elegant way to get all the data you need, as deeply and drill downey as you want it, without the manual effort which still falls short on reporting results. This, is the touchpoint approach. Instead of being reliant on campaigns, everything is pulled in automatically, down to granular asset and channel levels, to keywords and ROI, with in most cases all of this data being populated right in the CRM for ease of reporting. If you’ve got a marketing mix that spans online, offline, and maybe even outbound, touchpoints give you a scalable and granular architecture to answer pretty much every question you could be asked in terms of what channels and content assets are driving pipeline growth and revenue, down to the n’th degree of detail. And if this data lives in the CRM, all the better since everyone who needs it across the org has access to it, using the reporting tools and chops they already know.
• #18: And just as nicely, they’re easy to set up, and they take A LOT less work to maintain and scale than campaign objects. And also unlike touchpoints, they can capture everything important across every channel of your spend, since you don’t have to anticipate every single possible way and channel someone could visit your website, or engage with your marketing efforts. And because you can drill down, and down, and as far down as you want, again, this means you can optimize not just channel by channel, but within each channel, so you can understand every parameter, dimension, and optimize for each and every one of them. Honestly, this is where much of the real ROI of using attribution happens, which is being able to make these intra-channel optimizations and decisions, knowing where every dollar should go to drive revenue and pipeline growth, wherever and whenever you need to. But again, this doesn’t make touchpoints right for everyone. It all comes down to what your marketing mix, deal cycles, and overall deal complexity looks like. Since you can’t do this out of the box with marketing automation, it also requires investing in an attribution platform, which of couse has an associated price tag with it.
• #19: Oh and footnote, Adobe sells an attribution solution that just happens to be touchpoint-based, just in case you didn’t already put that together ;) It’s called Bizible, and I’ve personally gone through two acquisions, pretty much every job title, and the better part of a decade working and evangelizing this platform, and to this day still continue to extoll the virtues because that’s how much I believe in the results it has provided almost a thousand customers. So, if what we’ve covered here sounds relatable in terms of challenge, and the touchpoint approach sounds like it’s worth considering, I’d be happy to be a part of your conversation since, as you can tell, I rather enjoy the subject. And of course, as you can imagine, many of the conversations I have are about, well, is it worth it? Can we justify the expense for the returns? And I’m here to tell you firsthand that, well, I don’t know. I can take a guess if you’re a B2B organization with a long sales cycles into buying committees with an omnichannel marketing strategy, and I can plug some numbers into our Forrester Research formula designed to help evaluate the ROI of Bizible to give you some more data and confidence, and I can tell you exactly how customers have transformed their marketing universe across a variety of industries in a way that may just tug at the heartstrings of whoever controls your purse strings, but ultimately it’s a conversation we would need to have first, and of course, a conversation I would love to have.