The Top 7 Marketing Metrics for a QBR or Board Meeting

The other day an old friend, a highly accomplished marketing executive, asked me a simple question: if you only had five metrics to summarize marketing performance for a quarterly business review (QBR) or board meeting, what would you pick?

In this post, I’ll share my answer to that question. (Hint: I cheated and used seven.) 

I made my list from scratch. In order to avoid any anchor bias, I refused to even look at the draft list she sent me before coming up with my answer. 

I kinda cheated a second time because I grouped each metric under a heading. I like to remind people of marketing’s priorities, hopefully demonstrating alignment in the process. And, if marketing is not aligned, taking a grouped approach provides a clear opportunity for someone to speak up. Note that unlike some Kellblog posts, I won’t talk much here about formatting the metrics [1]. Instead, I’m going to focus on the metrics themselves. What should we measure?

If I were to present these, I’d preface this by saying, “Good morning, great to see everyone, and as a reminder, here’s what we do here in marketing. In priority order, we …”

We Make Pipeline That Closes

1. Marketing-sourced pipeline generation. I prefer measuring pipeline generation using opportunity count, not dollars, both because it’s more visceral and, particularly when there is a broad range of opportunity values [2], it can be impossible to know the value of an opportunity without getting fairly far into the sales cycle [3]. (And don’t worry, we’ll cover dollars below in metric three where it matters even more.) This metric is about count. Think: we agreed that marketing needed to generate 110 stage-two opportunities [4] during the quarter and we generated 120.

2. Marketing-sourced pipeline conversion. Because we understand that the point is not just to generate pipeline (which is really only a leading indicator), but to generate sales, we measure the conversion rate of marketing-generated pipeline. The trick is that this is an inherently lagged measure and the longer your sales cycle, the longer your lag. To make this concrete, the table below demonstrates an idea I call time-based close rates. If you generate 120 opportunities this quarter, while 23% of them may close in the fullness of time, 2% close this quarter, 4% next quarter, 10% the quarter after that, and so on.

Because sales lives quarter-to-quarter [5] and will die waiting for the fullness of time, we must factor this progression into our planning. We must also account for it in our metrics and the only good solution I know is to use trailing twelve month (TTM) conversion rates [6] [7]. Note that the CMO is stuck on the horns of a dilemma: either face criticism for using a lagging but highly sales-aligned indicator or face criticism for using a leading indicator that might not result in sales [8]. I’ll take the former in this case, particularly because so many other marketing KPIs are only leading indicators of sales. 

We care about pipeline that closes, not just pipeline that gets created or advances to demo stage, but pipeline that closes. I show that caring with this metric.

We Tee Up Sales for Success Each Quarter

3. Day-one pipeline coverage. This ties to my idea that the CMO should be the quarterback of the pipeline. Not just the marketing-sourced pipeline, but the whole pipeline. Most companies have four sources of pipeline with specific targets for each. For example, 60% from marketing, 20% from partners, 10% from outbound SDRs, and 10% from sales. The way most organizations are structured, the only person who owns all four sources is the CEO. Thus, insanely, in most organizations there is no natural owner for the overall pipeline other than the CEO. Because the CMO should always have the CRO’s back, because the CEO should delegate this important responsibility even if there is no natural owner, and because marketing is usually the majority pipeline contributor, I believe that the CMO should be the official owner of the overall pipeline. 

This means it’s the CMO’s job to ensure that sales starts every quarter with aggregate 3.0x pipeline coverage and, as a key part of that, to forecast next-quarter starting pipeline. That forecasting process should start early enough that you can still do something about forecasted problems, e.g., no later than one full quarter in advance. ”Doing something” might mean asking for more demandgen dollars or asking one of the other pipeline source owners (e.g., partners) to step up to higher targets. Worst case, it means escalating the forecasted and as-yet unresolved problem to the CEO.

The metric here is simple. The philosophy behind it is not [9].

We Generate Pipeline Efficiently

4. Demandgen cost per opportunity. Because we understand that SaaS companies effectively buy customers and that most SaaS companies require more than one year to recoup their investment in customer acquisition, we continually seek to reduce our demandgen cost per opportunity [10]. I pick demandgen cost per opportunity rather than overall marketing cost per opportunity because I want to put emphasis on the incremental (aka variable) cost of generating opportunities. If I want to measure overall marketing efficiency, I can use the marketing contribution to the CAC ratio. Here I want to focus on demandgen cost because it excludes fixed marketing costs that aren’t linked to generating opportunities (e.g., CMO salary, PR firm retainer) and because the primary business question I want to answer is: how much will it cost to generate 50 more of them? To do that, I don’t need to hire a second CMO or increase the PR retainer. 

If you take this approach, someone will eventually criticize you saying, “you’re deliberately understating the total cost of marketing-generated opportunities by including only demandgen costs,” to which you will reply: ”No, I am correctly stating the demandgen cost of opportunities because that’s the business question I’m trying to answer, and if you want to talk about overall marketing efficiency we can look at the CAC ratio and marketing’s contribution to it, including the sales/marketing expensive ratio.” [11]

We Get the Word Out

5. Awareness. Important as it is, demand generation is not the only thing we do here in marketing. We’re also responsible for getting the word out, making sure potential customers have heard of the company, have a positive opinion of us, and would consider us if and when they go shopping for a solution [12]. Towards that end, we run a number of programs to drive awareness/opinion/consideration in the market including public relations, brand advertising, and content marketing. Demandgen itself generates awareness as a by-product. 

To get an aggregate measure of these activities, we run a quarterly survey of buyers that measures:

• Unaided awareness. Name vendors that come to mind in the XYZ space.
• Aided awareness. Have you ever heard of vendor? [13]
• Positive opinion. Do you have a positive opinion of vendor?
• Consideration. Would you consider purchasing vendor?

While we’re happy to share the full report with anyone interested, in the QBR meeting we present aided awareness for ourselves and our top competitors.

6. Organic web traffic. The other way we measure general awareness is through organic web traffic, specifically how many unpaid visitors we get per month on the website. Are people finding us on the Internet and visiting our site? This is a coarse measure, but it allows us to keep an eye on how we are doing over time and relative to our competition [14]. 

We Care What Sales Thinks

7. Internal marketing CSAT. We view sales as our internal customer and our overall mission as to make sales easier. Towards that end, we run an internal customer satisfaction (CSAT) survey of sales each quarter and report back sales’ overall CSAT rating with marketing at the QBR. In order to inform our OKRs, we ask about many things (e.g., priorities, challenges) in this survey and the full report is available for anyone who attends the QBR.

I’ll conclude with a slide that summarizes this post.

# # #

Notes

[1] There has been some great content produced about this and in great detail of late — e.g., the Iconiq Go-To-Market Reporting Guide. While I’ve not yet reviewed it with a fine-tooth comb (because it’s both long and brand new), it looks quite good on my initial skim.

[2] Thus you end up using a placeholder value for new oppties which is effectively a proxy for counting them. If you create new oppties at zero value in such situations, I don’t pollute the pipeline with lots of proxy-valued oppties and, if I want to, I can always create “implied pipeline” by substituting 0 with the overall ASP or segment ASP. It’s impossible to do the reverse, because if your proxy value is $50K, you won’t know if a $50K oppty is a real value or a proxy value. 

[3] The other problem is that opportunity value is not single-valued but changes over time. So if you want to do pipeline metrics on value then you immediately beg the question: when? When the oppty was created? When it was 90 days old? When it hit stage 3? The world is much simpler if you just deal with counts for pipeline generation targets.

[4] Aka, sales-accepted opportunities. Generally in the sense that two keys have turned: an SDR thought it was an opportunity and passed it to sales, and a quota-carrying seller concurred.

[5] A favorite quote: ”I want salespeople who live in 90-day increments.”

[6] The simpler approach is to look at the TTM close rate of the year-ago cohort of new opportunities. The more complex approach is to look at the TTM close rate of all oppties generating in the past year, effectively stacking and sliding the progression (close rate vector) above. 

[7] At the 1Q24 QBR in January, I would say we generated 120 oppties in 1Q23 and 24 of them closed during 2023, for a 20% TTM close rate. (Unbeknownst to me at the time, 3 more will eventually close per the last column but that hasn’t happened yet. I could mention as an aside that 3 more are in the forecast for this quarter if I wanted to, assuming that none have close dates beyond that.)

[8] Such as stage 4 oppties or, while I don’t like demo as sales stage, oppties that reached demo. These are further down the pipeline than stage 2 oppties, but they are nevertheless still leading indicators and not sales. Because getting to these intermediate stages happens faster, the conversion rates are less lagged, but they are alas still leading indicators. I’ve talked to many CEOs who hooked everything to demo as a key stage, only to find that they were doing lots of demos, but not making many sales.

[9] People indoctrinated with a silo mentality may find it illogical or impossible to be accountable for something they don’t fully control. Think: how can I own the overall pipeline when I’m only responsible for generating 60% of it? I challenge such people to change their thinking to: I have two jobs. One is to generate 60% of the pipeline. The other is to make sure sales is teed up for success every quarter. I do that by forecasting starting pipeline coverage, leading a small team of leaders to decide what to do when we’re forecasting below target, and when needed escalate the problem early to the CEO.

[10] And the other way we try to reduce customer acquisition cost per dollar of ARR is to provide programs, tools, and training that increases the s2-to-close rate. We need to think of this as reducing demandgen cost per opportunity while holding quality constant (or improving it) where quality is measured by the s2-to-close rate.

[11] For die-hards, I’m often guilty of conflating incremental (i.e., marginal cost) with fixed vs. variable cost. The CMOs salary is a fixed cost. Demandgen is a variable cost in that it varies with volume. Total demandgen spend / total oppties generated = average cost per opportunity which is the actual calculation I’m encouraging. A true marginal cost would be the incremental cost of generating 1 more oppty, e.g., the cost of getting enough clicks to generate enough leads to generate a single opportunity. Here I think the average cost works fine and the real improvement is excluding the fixed costs that blur up the incremental cost of getting 10 or 50 more. But I’m sloppy in my language sometimes.

[12] And trying to accelerate that shopping trip is another thing we do in marketing, but the specific focus here.

[13] For most early- and growth-stage startups, <vendor> and <product> are synonymous. For bigger companies, you need to separate them. It’s not: have you heard of Salesforce? It’s: in the CX space, have you heard of Salesforce Experience Cloud?

[14] There is a nuance here but I do think companies should track this for both themselves and their competitors. The nuance is that for your own site, you can “know” how much traffic you get, but for the competition you can only “guess,” using tools like Ahrefs or SimilarWeb. The trick is when their guess for you is off, there can be a tendency to dismiss the competitor data as well. That’s a mistake. Present your own data for you over time (that you “know”) and then, when doing competitive analysis, compare the “guesses” using only the guess data, basically hoping for compensating and consistent errors in the process.