How to get B2B marketing measurement right

Early martech adopter reveals the metrics that matter most

By Published on .

Credit: iStock

The holy grail for marketers has always been to correlate spend with sales. In B2B, especially for larger purchases, measurement is further complicated given an average 14-month sales cycle involving five or more decision makers. Yet, a few chief marketing officers are getting closer, deploying a complex martech stack with more than two dozen separate technologies that help identify, track, nurture and qualify a lead before turning it over to sales.

One of the masters of this approach is Brian Kardon, CMO of Fuze, a cloud-based unified communications provider. As the former CMO of Lattice Engines and Eloqua, Kardon was an early adopter of martech and remains an evangelist for the science-side of marketing. In our conversation, he reveals what metrics matter most to him, specific details on the cost per marketing qualified lead, his testing approach and the acronym most boards cherish – the CAC ratio.

What are the metrics that matter to you the most?

There is this tendency of marketers to spew out hundreds and hundreds of numbers that have no meaning at all but focus is a key theme here. Because you can't focus on 50 or 100 metrics, it must be fewer rather than more. You also have to share them with sales and your CEO and your board. The most important metric to me is: how much of the net new pipeline did marketing source? I care about pipeline sourced, marketing influenced, and what percentage of the bookings that closed were sourced by marketing.

How often are you looking at performance data?

I'm looking at campaign performance all the time. So, while I can't figure out whether a campaign delivered closed business, I can know whether the campaign was successful in providing a marketing qualified lead (MQL) and then a sales qualified lead (SQL). I'm looking at early indicators.

Let's get into the weeds on MQLs and SQLs.

In the course of a year, we run a couple of hundred campaigns. A campaign could be an email campaign, it could be a physical event, it could be any number of things. But we're always looking at measurements of the campaign and we're looking at the cost for an MQL all the time, so we have a pretty high cost per MQL. I daresay it's over three hundred dollars for a marketing qualified lead. All these things sort of come together. I'm always looking at the performance of my campaigns, cost for MQL and conversion rates from increased MQL, MQL to what we call a "discovery call" which is sort of an SQL to SQO (Sales Qualified Opportunity) to the pipeline.

Where are you regarding attribution modeling?

We think we have world-class attribution, as good as it gets. It's not perfect. And so, we're looking at all the deals and what content they touch at what stages. So, we tag all our content whether it's top of the funnel, middle of funnel, the bottom of funnel. We also have some exciting things on our website now. If it's an unknown prospect we use reverse-IP lookup, so we'll know what company they're from—assuming they're not calling in or using Comcast or Verizon from their home but they're at their company.

Can you give an example of what specific web behavior means?

Sure. If they're looking at a pricing page or a product page, it would imply they're further down the funnel. If they're downloading content about the future work or something very high level, they're not as interested. We try to follow up with content based on the buyer stage that we infer from their web sessions. We tag different pages on our website and different content for a particular buying stage, which is something I think a lot of companies don't do, but it's valuable for us.

Do you test a lot of variables?

Yes. On our website, for example, we probably have 40 A/B tests going on right now. Little things; size of buttons, the color of buttons, calls to action, placement on the page. So, every time you refresh a page, it will look slightly different, and people will be presented with slightly different pages each time. We test for things like session length, bounce rate, click through rate.

Are there different metrics that matter when you're presenting to the CEO and the board of directors?

They care a lot about the CAC ratio—the cost of acquiring customer over lifetime value. They also care that sales and marketing as a percent of revenue is declining. They want to see the productivity of sales and marketing such that every nth dollar is driving incremental bookings or incremental sales. So, we want to try to show the productivity and a big part of a growing company is scaling. They want to see that your product is scaling, engineering is scaling. As you grow, you spend a lower percentage on that department.

Most Popular
In this article: