AMA: Front Head of Digital Demand Gen & ABM, Moon Kang on Data and Analytics

OOH is admittedly something I am not too experienced in. In the past, I've run geofenced display and paid social ads but those were more or less attributable to that region, so it was a hybrid OOH and digital campaign (probably closer to a directly attributable digital campaign). However, my approach is always to be as close to attributable as possible. I would try my best to figure out where my target audience is and partner with vendors that can allow me to serve impressions there. For example, if I have a high concentration of target accounts in the financial district of NYC, I wouldn't run a subway ad at the 125th st station. I would work with a vendor that will let me rotate my digital ads and static signage in/around FiDi. I'd pay a premium for it too because I know the impressions there would be actually going toward my target audience. I would then take a look at direct and organic mobile traffic coming from that area period over period to determine the efficacy of that campaign. I'd also review branded term search impressions from that area and see if there was a spike. Overall though, OOH is about building awareness, and hopefully your OOH ad has a clear CTA; whether it's to ask users to visit a vanity URL, scan a code, or Tweet something with a hashtag, you will always want to measure your primary success KPI as the number of individuals that took action on your call to action. The secondary metrics would be web traffic from the regions you ran those OOH ads. Then lastly would be using a social listening tool to capture any mention of your ad or specific language you put on those OOH ads that you find to link back to folks' posts on social. OOH is something I'm eager to learn more of and practice. 

Yes. It's nothing out of this world -- typical UTM parameters pass all the way through to the contact record, which, when an opportunity is created on the same account, passes through to the opportunity record, which we follow all the way down to closed-won. We then look at LTV:CAC down to the utm_term level to see which paid search terms drove actual revenue/ARR. I do this for all paid initiatives. For non-paid initiatives, we tie it to the "journey" or asset level. What original touch point got the user to convert on our website? From that initial touch point, what additional marketing/sales touches led to a meeting? We follow those all the way through to revenue.

The demand gen team is goaled on inbound stage 2 Salesforce opportunities. Not only do we need to create demand, but we also have to capture it, nurture it, and progress it down the sale funnel, all while increasing velocity by removing friction at every stage of the prospects' journey. The inbound stage 2 Salesforce goal is determined by our revenue goals. We work backward from what revenue we need to hit our corporate goals. We then look at the Salesforce opportunity progression percentages to determine how many inbound stage 2s we need to hit that revenue goal, then I work backward to determine how many stage 1s I need with our given stage 1 to stage 2 conversion rate. I do the same to determine how many marketing-qualified leads we need and how many engaged leads we need to hit our goals. To keep each other accountable, we offer absolute transparency into how we are pacing toward our goals. We have a dashboard that shows our quarterly goals and how we are pacing toward those goals. This dashboard is refreshed every 3 hours. 

I break out the DG metrics into two groups: 1. Demand generation Here I focused on generating demand through "push" campaigns to our ICP accounts. I call this "push" because I am essentially shoving our ads in front of folks who never asked for it. I look at things like ad impressions, CTR, engagement rates, and conversions. From an ABM perspective, I look at the same ad metrics, but I also look at account view-through rates (the % of accounts who've seen my ads and then visit our website by other means), intent signal increases, and funnel progression. Holistically, I want to ensure we are driving the right level of engagement to accounts on our target account list. Within our target account list, I want to make sure, period over period, I am creating awareness & demand at the same accounts our outbound teams are working so they have a higher rate of turning emails and calls into conversations. 2. Demand capture Here I focus on capturing demand through our "pull" campaigns to audiences that already have existing demand for our solutions. These are paid search, review websites, and content syndication with intent signals. The metrics I look for here are pure lead capture and lead nurture into MQLs that I can hand over to our sales teams to work. I review these metrics daily, but only when I see anomalies that last more than 3 days do I actually take action. On a weekly, monthly, and quarterly basis, we review the demand gen machine by channel (paid search, paid social, organic, direct, etc.), then break it out by campaign, down to the utm_term level and ultimately follow these down to revenue and LTV:CAC. The intent here is to analyze the numbers to surface insights and propose action items for the month or quarter.

I would consider data and metrics to be synonymous in most cases, however, if I had to define it individually, I'd say data is the raw format of all the little quantitative metrics we are collecting on an individual, company, or campaign. Metrics are a bit more aligned to the quantitative data that I want to see because it's important for the analysis. For example, raw data might contain page views, sessions, new users, returning users, conversions, cost per conversion, and conversion value. The "metrics" I care about when looking at a website's performance is page views, sessions, new users, and conversions. Conversion value is more of a metric to look at the ROI or ROAS of a campaign. Basically -- data is the raw dump of quantitative metrics as a whole. Metrics are what I filter for to do a proper analysis. This brings me to analytics. To me, analytics is the effect of pulling the raw data, sifting through the core metrics, and analyzing the quantitative metrics to uncover actionable insights. Analytics, to me, is translating quantitative metrics into a qualitative story/narrative. 

Some of the insights I've seen that led to direct action items are misalignment on lead follow-up and poor multi-threading practices. If a lead comes in after seeing an ad about purple bananas, searches on Google for "purple bananas" and clicks on our ad that claims we have the best purple bananas in the biz, and the person requests a demo to see our purple bananas in action, we need to ensure our first-touch sales follow-up reinforces that intent and share how excited we are to schedule a demo call about our purple bananas. Sometimes the most obvious insights are here, when we follow up for demos about yellow watermelons instead. We see often times this breakpoint is where we see leads go unresponsive, or it takes too many touches to bring that marketing scent back into the convo to progress the sales cycle. Another insight that impacts changes between sales & marketing, is when we see Salesforce Stage 1 meetings get rescheduled or lots of no-shows. If we see this a simple action item is to inject automated reminder emails and/or inject a call to action or a teaser video in the calendar invite itself. When we see meetings canceled, rescheduled, or a flat out no-show, is when the lead was not properly educated on the exciting opportunity that awaits them on that call. By including a CTA in the calendar invite such as "please let us know what your 3 biggest KPIs are and I'd love to share at least 2 ways we can help you with those KPIs on our call tomorrow!" or a teaser video that will reinforce the value prop of your business to the lead as they inspect the calendar invite that they received last week and think about whether or not they should join. Think about how many times you've woken up to check your calendar and not remember what the meeting is about -- now imagine seeing a teaser video of that neat tool you were researching to help you hit your quarterly goals... that should eliminate reschedules and no-shows immediately. 

I believe a successful structure requires a robust data analyst on both Marketing and RevOps. The RevOps analyst is very important because they decide, based on the company's revenue goals, how many various SFDC stages are needed to hit that goal. They understand the stages and their progression rates (i.e., Stage 1 to Stage 2 rate, Stage 3 to close) to properly define how many opps are needed to hit their goals. They are too far removed from the marketing operations and marketing campaigns to determine how marketing can properly impact the growth and efficiency of these metrics. This is where the marketing analyst comes in. Based on the RevOps' assessment of revenue goals, they have determined the # of sales accepted opportunities, sales qualified leads, marketing qualified leads, and marketing engaged leads are required to hit our goals. The marketing analyst now comes in to properly understand the marketing funnel stage progression rates and the various campaigns that are running to give the numbers a bit of a massage to say, "based on our marketing spend and cost per lead, we will need $X,XXX to hit our goals." They can also understand the seasonality of marketing campaigns and determine when to set goals higher or lower through the year. They will also know that if paid search leads come in at $50 a pop, you can just give marketing $50m to generate 5m leads... there's the knowledge of diminishing returns, total addressable market, and efficiency scale that a marketing analyst that's closer to marketing would know. They then work together to come up with the numbers and tell their story to the sales and marketing leaders, who can both agree, based on their push, pull, and ABM strategies, that they are comfortable with the goal, and a partnership is formed. 

I think strategies like ABM are decided upon when you identify a strong, core ICP that you can trust both quantitatively and qualitatively. ABM requires a solid ICP to understand who to go after but strong qualitative feedback from sales leadership validates and supports that ICP. When you have that ICP and both sales & marketing believe in that profile, you decide that your tier 1 and 2 accounts are the only accounts that meet the ICP. Some things that I'd look at to determine ICP (which would lead to ABM strategies) include average contract value, deal velocity, and LTV. I would take those metrics and start to extract firmographic parameters to determine what makes up the characteristics of accounts with high ACV, the fastest deal velocity, and that have strong LTV. I'd take those characteristics and start looking at ABM strategies to only go after accounts that meet these characteristics since these will give me the most bang for my buck. As far as PLG goes, I believe that any/all software companies have an opportunity to build a little side PLG model that you can decide to blow up into a full-fledged program. Who you start with is what matters most. Some of the metrics I'd look at are your product's power users, sign-up to activation, and high-growth accounts. Once again -- take the characteristics of those accounts and determine who set of personas are prime for a self-serve model. From there, you start to analyze every step of their journey within the product. You look at metrics in-product such as # of sign-ins, pages with the highest session duration, the most amount of clicks, and exit pages. You then start to remove any/all friction points on these pages while simultaneously pouring more people into this self-serve/PLG funnel to get more data and identify more friction points to remove. Over time, you will have a well-oiled self-serve machine where users are going through the sales journey themselves and eventually helping each other out through the building of communities and forums.