As a CMO, are you fully satisfied with the marketing performance metrics you’re able to report to the Board?

As a senior B2B marketer, are able to report all the ROI metrics that the Executive team is asking for?

For most B2B marketers, the answers to both of the above questions is a resounding “No.” In fact, according to John Neeson, Co-Founder and Managing Director of SiriusDecisions, four years ago, the #1 CMO priority was “the need for greater ROI.” Not much has changed in four years. In their 2016 CMO Study, Sirius Decisions found that CMOs still wanted to “improve their competencies through measurement.” But this time, it’s supported by a parallel organizational goal to create a marketing operations competency (I’ve previously written about the Rise of Marketing Ops).

CMOs have become aware that to achieve the promise of closed-loop reporting, they must build the marketing operations muscle required to align people, process, and technology to deliver marketing ROI metrics.

As they embark on this journey, CMOs will keep asking their teams, “Are we there yet?” And they will hear a common refrain as to why the answer is “not yet”: Bad Data.

But just what is bad data anyway? The answers from your team can be confusing, and not always internally consistent. So, as a former CMO, I’ve categorized 8 distinct types of bad data to help you understand the obstacles your team is facing on the road to Marketing ROI.

1. Account Hierarchy and Duplicates

Challenge: With the rise of Account-Based Marketing, accounts — always the primary focus of sales organizations — are making a bid to get more of marketers’ attention. But you may hear from your team that your accounts are a mess, and you have a lot of work to do to get anywhere near ready to do account-based ROI measurement.

The most common problem is that you have many duplicate accounts. A related problem is that if you’re going after large enterprise companies — like IBM or GE — you will have to make a decision about how to structure parent-child account hierarchies.

Solution: This is an area where you want to aim for good enough and not perfect. Account duplication is a problem you can — and should — tackle with a variety of tools on the market. On the other hand, I’ve often seen people fixate on edge cases regarding account hierarchy. GE and IBM are Fortune #8 and #31, respectively. Yes, they’re complicated organizations. But most Fortune 500 companies aren’t that complicated. If the Fortune 50 isn’t your primary focus, you may not want to get rat-holed on this particular challenge.

Summary: Get to good enough — not perfect — with respect to the quality of your account data.

2. Opportunity Segmentation

Challenge: In CRM systems, the Opportunity object represents a particular sales pursuit, and thus carries the pipeline and revenue dollars with it. In order to do any kind of marketing ROI reporting, you need to leverage this information. The problem arises when there are multiple opportunities per account, or when certain marketing efforts should not be credited for certain opportunities. For example, excluding renewals and upsells from being credited to marketing campaigns is common.

Solution: This data obstacle is primarily a mythical one. Yes, it requires some work to sit down with your team, Sales, and Finance and agree upon a set of rules that make sense for crediting marketing influence. But that is the work you have to do anyway if you want to have credible metrics.

For example, you may decide to exclude renewals and upsells altogether. Or perhaps you have multiple product lines, with very distinct target customers, and specific campaigns geared towards them. If it’s relatively easy to do, you can decide that only certain campaigns may influence certain types of opportunities. But an attempt at perfection here may betray your own team’s lack of comfort with ambiguity. With respect to marketing ROI reporting, most people are flying completely blind. So you’re probably aiming for a v1 — with directional accuracy — that will likely result in millions of dollars as a result of incrementally better decisions. Why wait?

Summary: You probably shouldn’t sweat this one too much. While you want your reporting set-up to be configured to your business, you can address all manner of special attribution rules after your initial reporting process gets going.

3. Orphaned Leads (Lead-to-Account matching)

Challenge: Most CRM systems were designed in an earlier era, before the rise of content marketing. They assume that a small number of leads will linearly flow into Opportunities. The reality is that most B2B businesses are generating large volumes of lead conversions relative to their Opportunity counts. What often happens is that Marketing works hard to generate 10 leads at Microsoft, let’s say, who are expressing interest in your solution. This reveals to your sales team that there is strong interest at Microsoft in your offerings, and that they should be prioritized. The Account Executive cherry picks the most promising lead, and converts that one into an opportunity. That leaves 9 “Orphan” Leads — the right people at the right target account with past responses to your campaigns, left hanging with no associated account.

Solution: This is indeed a real problem. In fact, our own benchmark research shows that for every 1 Contact associated with an account, there are another 1.6 Orphan leads (BrightFunnel, 2016). There are several different ways to solve this problem. One approach is to programmatically convert Leads into Contact-on-Accounts in your CRM, based on some sort of fuzzy logic. There are several solutions on the market that do just this.

However, I have to caution that changing core CRM data just for the benefit of better marketing reporting is a slippery slope. There is a reason your Sales team left those leads unconverted. If they want to convert them so they have the info on the Account, they probably would have done that already. An alternative approach is to simply do the fuzzy matching as you are creating your marketing ROI reports. This is more expedient in most cases. You can always send the list to your Sales Ops counterpart to make their own decision.

Summary: Use fuzzy matching to associate leads to accounts, instead of universally converting all leads.

4. Missing Contact Roles on Opportunities

Challenge: We have all been trained by CRM systems — Salesforce Sales Cloud in particular — to believe that people (Contacts) need to be manually entered onto an Opportunity by a sales rep. When using native Salesforce reporting, if there are no Contacts on the opportunity, there is no marketing attribution, period. So many companies enforce the requirement for at least one Contact attached to an opportunity — as surely, the AE is selling to someone. But it gets more complicated. On average, sales reps attach 1.2 Contact Roles per Opportunity (BrightFunnel, 2016). But those same opportunities have an average of 5 influential Contacts and another 8 Orphan Leads, for a total of 13 people. Marketing, by definition, is your company’s air cover, attempting to reach those 13 people. Sales, by definition, is trying to develop a Champion, and who will reach those 13 people from the inside. So even if a good sales rep attaches one Contact Role, it is highly unlikely they will attach 13, let alone even know who all 13 are in many cases.

Solution: Some marketers go on the warpath, trying to change sales reps’ behavior to do more data entry. This is a losing proposition, and likely a career-limiting move on any B2B marketer’s part. Trying to redirect sales efforts is like trying to redirect a river. It can work temporarily, but it takes a lot of resources to keep up. What I recommend is to let your closers follow the path of least resistance. Instead of badgering them to attach Contact Roles, simply ignore that concept altogether. Instead, what makes sense for most people is to attribute campaign influences at the Account level. This was true long before the rise of ABM, but it is especially true now. This can be done through some fancy spreadsheet work, or a dedicated analytical platform.

Summary: You do not want to “solve” the problem of AE’s attaching Contact Roles. It’s reasonable to come up with a process to attach at least one, but don’t rely on this data entry for marketing ROI reporting. Instead, report at the Account level.

5. Lack of Campaign Data

Challenge: Some organizations are so new to adopting marketing automation tools that they have not gotten to a point where they have sufficient campaign data. The marketing team itself most likely hasn’t started assiduously creating campaigns for every marketing effort across all channels. Or if they have, they haven’t yet synced the data over to their CRM system, where it can be associated with revenue and pipeline.

Solution: This is likely a hump you will have to get over before getting meaningful marketing ROI. The promise of the last wave of marketing tools — led by Eloqua and Marketo — was that we would digitally be able to track every online and offline campaign interaction. To do so, you must first adopt such a tool (there are many more alternatives as well, such as Salesforce’s own Pardot), and ensure that your entire team gets into the habit of tracking their data appropriately. This is second nature to email and digital marketers, but for field marketers, it requires some extra legwork, so that may be an area where you allocate additional resources and training for healthy marketing operations practices.

Summary: Simply put, you’ve got to have campaign data to have marketing ROI reporting. If you’re using a marketing automation platform, you’re halfway there.

6. Lack of consistent Campaign “success” definitions

Challenge: Some organizations achieve initial success with rolling out their marketing stack, but find out a year or two into the journey that their definitions of “success” aren’t consistent across time or various channels. This can lead to some serious hand-wringing as to whether to go back and re-engineer the whole thing before doing any kind of marketing ROI reporting.

Solution: Rather than re-doing everything, the better approach is to create different definitions of success after the fact. For example, perhaps your webinar signups and attendees are both marked as responses, but going forward, you only want to count the attendees as being influenced by that campaign type. That’s a perfectly reasonable judgment call to make, and a very common one. But it makes more sense to apply that rule after the fact. But if you must, you can go back and fix all your old data as well. I’d recommend you do that after starting to report on marketing ROI.

Summary: This is a common problem with a fairly simple solution: create and apply campaign success rules after the fact rather than wishing for a time machine to take you back to your initial MAP implementation. Do not let this be an obstacle to measuring marketing performance.

7. Changing funnel stages

Challenge: Many B2B marketers have adopted a multi-stage approach to demand generation. For example, they may have several lead stages identifying degrees of engagement, such as Inquiry, MQL, SQL. This helps align Sales and Marketing efforts more closely, working off of a common definition of various stages of lead engagement, just as there are various stages of opportunity engagement.

However, organizations, especially rapidly growing ones, often evolve their stage definitions as they grow. Moreover, while the average Opportunity-to-Revenue cycle is 108 days, for high growth tech companies, the average Lead-to-Revenue cycle is 512 days (BrightFunnel, 2016). So from the time Marketing initially engages with someone at an account to that account generating revenue is 17 months. A lot can change in that timeframe, including what you deem to be a qualified lead.

Solution: Many organizations are tempted to change all their historical data to the new definitions so that they can have accurate marketing ROI reporting. I would caution you against doing this, as generally any data changed in your sales system of record — purely to benefit marketing — will raise some concerns. You’re potentially taking up scarce cycles related to your CRM to benefit marketing. Most mid-sized to large companies have some sort of governance process related to CRM systems, and limited Sales Ops bandwidth to implement changes.

The alternative, non-invasive approach is to apply those stage definitions retroactively, for reporting purposes — when comparing previous cohorts with more recent ones, you define those stages to match the current ones. You can accomplish this with an intrepid analyst and Excel, or through a dedicated marketing performance measurement platform.

Summary: Everyone has this problem, given that it takes 512 days from Lead to Revenue, and your business — and funnel stages — will change in that time period. Optimize for the future, but don’t let past definitions get in the way of accurate analysis.

8. Missing Cost Information

Challenge: To have the “I” in ROI you need to track the cost of your marketing efforts. However, most mid-sized to large marketing organizations have poor data completeness with respect to cost. There are a few reasons for this. For one thing, you likely have hundreds of campaigns, each needing an associated cost. In a 50-person marketing org, this might be spread across 30+ campaign owners and 5+ directors. Finally, not only are the responsibilities diffuse, the return (if you’ll pardon the pun) is unclear. What good is calculating detailed cost figures if no one ever looks at marketing ROI? There is a chicken and egg problem.

Solution: Don’t overcomplicate your cost calculations for marketing ROI purposes. Come up with a simple policy that ensures that the total sums up to your marketing budget. Decide if and how to allocate your payroll costs. Ensure that campaign managers include things like travel costs for trade shows. Ideally, you’ll want to run what you’re doing by the Finance team. Chances are, they may even enjoy helping solve the tricky cost accounting issues associated with marketing spend. Lastly, ask every director and campaign owner to report on marketing ROI. Cost numbers will magically start getting entered in the right places. (Warning: Do NOT attempt to sync with your financial backend. No good can come from this — certainly not in Phase 1).

Summary: Cost is very important to track, but its allocation can be a seemingly subjective exercise. That’s why you get paid the big bucks. Make a judgment call on a consistent policy. Does it really matter if that show cost $30K or $35K? Probably not. It’s more important that you show that you’re tracking costs and holding people accountable to do so.

These are the 8 demons you will encounter on the path to Marketing ROI. Be careful which you choose to engage directly, and which you choose to deftly bypass.

Remember: the length of your tenure as CMO can be highly correlated to how long it takes you to produce credible, meaningful marketing ROI metrics. And there will be no end to data problems. That’s simply the era we live in. Instead, aim to understand the potential data limitations, and get your whole team — and the executive team — comfortable with those limitations. Present the problem as choosing between putting your head in the sand while waiting for data perfection, and choosing to make probabilistic, 80/20 decisions, which requires comfort with ambiguity.

By tackling the problem of marketing ROI in an iterative, agile manner, CMOs can quickly show the value of their efforts, optimize their campaign execution, and align with their sales counterparts by minimizing data entry and CRM change requests.