December 22, 2015
Lead Scoring Tips
How many sales teams have you overheard saying, “Our marketing is amazing – they have our backs and send us more qualified leads than we know what to do with”? Not many, I imagine. The response is usually more like “marketing has no idea – we generate all our own real leads.” If this sounds familiar, it’s because a lack of lifeblood leads is often the first place fingers point when sales hits the skids. Plus, leads by themselves aren’t good enough - you want those with the need and budget, who are ready to buy soon. So for more than a decade, marketing automation vendors have offered lead scoring as the way to auto-magically qualify those “ready for sales” leads from the stream of raw leads. And while lead scoring is widely used, results are decidedly mixed. SiriusDecisions found that 68 percent of companies who use marketing automation systems do lead scoring, but that only 40 percent of sales people get value from it. By school standards, this would be an “F”. Moreover, there’s growing consensus that traditional (i.e. manually managed) lead scoring is cumbersome, complex, even broken to the point of being BS. At the same time, predictive lead scoring solutions like Infer that use machine learning are quickly gaining speed. So what does this mean for marketers?
First, a little background on lead scoring
Lead scoring is a “methodology used to rank prospects against a scale that represents the perceived value each lead represents to the organization. The resulting score is used to determine which leads [sales and marketing teams] will engage, in order of priority,” according to B2B research firm SiriusDecisions. Vendors claim that lead scoring helps grow revenue faster – and more predictably – by enabling marketers to focus on filling the funnel, then use scoring to sift out qualified leads. The expected result is threefold – sales reps won’t waste time sorting through hordes of mis-matched buyers, early-stage buyers are shielded from pushy sales calls, and marketers better understand their audience’s profile. Lots has been written about how traditional lead scoring doesn’t work, but let’s now dig into five common hurdles companies face when trying to qualify leads with scoring.
**The biggest hurdle: lack of access to statistical support **
The first challenge lies in determining the right score values to assign to leads based on their behavior or demographics. This is really an attribution exercise that is best handled by statisticians and software, but typically falls to marketers who have to act on gut feel and assumptions. The scoring process starts with marketing and/or sales managers agreeing on lead qualification criteria. These criteria are compiled into a scoring matrix, like shown below, which adds incremental point values based on demographics (e.g. company size or job title), marketing engagement (e.g. email clicks, web url visits, form completions), direct engagement (e.g. sales calls, support interactions, webinar attendences), or app usage (e.g. key in-app events). Without the support of a statistician, or having personal experience in performing and interpreting multivariate regression analyses, marketers can’t possibly know how important each of the above signals are in predicting whether a buyer wants to engage sales and/or will buy. Even if they did know exactly which signals drive conversion, they can’t weigh up the relative importance of each. For example, if you are targeting CIOs of 5,000+ employee companies within the retail vertical, then you might add extra points for a “CIO title” in “retail”, but what if they are in a 3,000 employee company? And what if they request a sales call, but have never viewed any of your content or signed up for a trial? Which are relatively more important to your ability to engage? Moreover, do any of these signals really matter to a sales rep, who just wants to connect to a qualified buyer and begin their own sales qualification process? Lead scoring point values are typically made based on best guesses, like that a “Contact Sales” request is worth 25 points while other form submissions are worth only five. Or that certain key events - like a user completing a “request demo” form or signing up for a trial - should “automatically assign” leads to a sales rep. Well guess what the sum of a bunch of assumptions is? An even bigger assumption. Yet this score is often the handshake between marketing and sales qualified leads…talk about an situation set up to fail. Many argue that lead scoring is a process, and that you need to start somewhere then refine. Like anything, scoring gets better with practice and time. The thing is, you can be investing your 10,000 hours into many areas - launching new campaigns, generating leads from search, producing fresh content, engaging customers and partners, or doubling down on high-ROI ad channels. Very few people have the time to continuously test, measure, and incorporate the incremental effects of lead scoring improvements into their process. There are better ways to generate qualified leads - which is the subject of my next post.
Four other reasons why manual lead scoring doesn’t work
In addition to needing a statistical basis and skilled operator, there are numerous other reasons why traditional lead scoring is broken. Reason #1: Lead scoring models need to be updated whenever new assets or buying behaviors arise. For example, if there’s a new product launch, new sales territory, or a new free trial, the existing scoring values need to be laboriously re-created and reset. Anyone who’s spent time grooming a scoring system, only to have to hit reset after introducing new assets or process changes, feels this pain. Yet most companies are ever-evolving, which makes this a leading reason why most companies give up. _Reason #2:_ Accurate firmographic and demographic scoring requires lots of data. Best-of-breed lead scoring vendors have built platforms to scrape, acquire, and compile vast proprietary databases, allowing users of their technology to cross-reference and score their own contacts. In short, the traditional method can’t handle the amount of data needed to reach statistical significance. Reason #3: Traditional lead scoring typically only uses data the marketer can capture. It misses out on many other predictive signals of buying behavior, like current technology usage, VC funding, or management team maturity. All of these can be critical background knowledge to tee up a first sales meeting or prioritize an email nurture journey, but are unavailable without extensive data augmentation capabilities like mentioned in point two above. Reason #4: Contact-based lead scoring does not reflect the opinions of every stakeholder, even though most B2B decisions are reached on a group-consensus basis. Besides sole proprietors and very small businesses (e.g. under 25 employees), most companies require support from multiple-stakeholders prior to committing to new purchases. Traditional lead scoring cannot unearth behind-the-scene politics or exchanges, nor does it alert members when a new executive or buyer appears on scene who might impact the vendor relationship.
Companies like Zendesk realized this a while ago
At my previous SaaS company, Zendesk, we ran a quarter long-experiment with a sophisticated, cross-functional team that included sales development, marketing operations, data analytics, and email marketers. We were already running a fairly complete lead scoring matrix, and selected at random 400 “ready for sales” leads and 400 non-qualified leads, with the intent to understand how much better the scored leads performed when engaged by sales. Despite our best efforts, we found no statistical difference in our ability to connect with, re-engage, or win the “ready for sales” leads compared to randomized, non-scored leads. Admittedly, we only dedicated one quarter to the experiment, but the experience confirmed anecdotally what many report - that it takes a dedicated operator (with a Ph.D.) to manually tweak and test scoring criteria to yield any results - and who knows how to measure those?
**Putting traditional lead scoring to bed **
Virtually any company can benefit by creating a plan of action to more efficiently guide buyers through their journey to conversion. While qualifying and nurturing leads is a very important part of determining a person’s interest level, it needs to be harnessed and leveraged in a better, more targeted way than traditional lead scoring provides. Implementing lead nurturing aimed at producing “real” contact sales leads, that evolves based on reader engagement, that drives intelligent lead routing, and that identifies hot user segments based on key insights or app usage, are better ways to create qualified leads without using lead scoring and drive 10% - 20% top-line revenue growth. So it’s time to stop fighting your scoring model, and to start winning more customers. Read my next blog post to learn how.