The Cost and Hassle of Predictive Lead-Scoring Drift


Companies and their data scientist consultants work feverishly setting up lead scoring—gathering, entering, and analyzing data to create detailed customer and prospect analytics.

Then comes drift, the fundamental shortcoming in predictive lead or opportunity scoring in marketing.

Things change. Data points become outdated. Lead scores are no longer accurate. Inevitably, the data scientists must re-engage so the data can be examined anew and the scoring updated.

The problem is that hiring data scientists for periodic re-tuning is both expensive and time-consuming. In the US, these mathematical geniuses earn upwards of $ 300K annually. Translate that into an hourly rate, and factor in profitability for the data scientist’s employer, and it gets costly very quickly.

Most midsize to large companies—particularly those in competitive markets—should re-examine prospect and customer scoring quarterly. But what most do is delay. And delay again. And they drift.

The Need to Stay Up to Date

Companies lose business when their scoring is not up to date—and they know it. No one wants to send out their sales team to meet with prospects using a GPS that doesn’t auto-correct when there’s a detour. Still, there’s that big expense associated with being properly steered.

The solution is finding a way to stay on track without breaking the bank.

Predictive analytics technologies for marketers offer a lot of promise. They are gaining popularity in the marketplace. Knowing how, whether, and how much to automate sounds simple, but those are important considerations.

Companies that see their potential sales falling because of data drift are prime candidates for automation. Not surprisingly, many companies that determine they require a better process for analyzing their scoring start with baby steps. They automate some elements of the sales process, such as automated calling agents and email campaigns. Initially, they may see an increase in conversions, but usually the initial lift is quickly followed by a decrease.

That’s neither satisfying nor beneficial. And, significantly, it signals a problem within their lead scoring.

Make Drift a Thing of the Past

Rather than giving up on automation or taking baby steps that don’t lead to any changes, embrace algorithms and use them to their fullest. Putting one’s trust in algorithms may seem scary, but remember that the algorithms are pulling from existing prospect, customer, and sales data while pulling in outside data sources to provide a more complete and much more robust understanding of leads than you could easily get on your own.

With so much data being analyzed—regularly and most importantly automatically—drift becomes a thing of the past. The sales trajectory can continue to grow while the valleys level off. Lead scoring becomes more meaningful, and the sales team can focus on those leads most likely to result in new sales and those that will help grow relationships with existing customers.

The best solutions offer both real-time intelligence and the benefits of simplicity; the sales development reps team doesn’t need to deal with the complexity of analytics. It gets the score, the reasoning behind the score, and a feedback mechanism integrated into the team’s CRM. There is very low level of training needed, so deploying and launching such a solution is extremely fast, providing an immediate ROI.

Worth the Effort

Make no mistake: A fundamental change like this likely will raise some eyebrows. Senior management comfortable with a manual, rules-based system might fret it is losing control to automation. People used to being able to define their targets may balk at technology solutions that can follow their targets and define lead scoring for them.

Overcome the fear by talking with other companies that have implemented automated lead-scoring solutions. See whether your vendor can offer you a paid pilot program using your own data and assess the results for yourself.

One customer my company worked with faced the dilemma head on. Though the customer had confidence in its more manual means of scoring leads, a change in its target audience left its lead-scoring system completely out of whack. Sales plummeted, and the company knew it had to do something. By automating the process, the system quickly found leads that met its new goals, and got the company back in business in four weeks. The company found its true north and is sticking to it.

MarketingProfs All In One


Don’t Let ‘Geographic Drift’ Steal Your Marketing Budget



Location data can be a very reliable indicator of where to spend your marketing budget. If you’re a fan of Foursquare, or other check-in services, then there’s a wealth of information data to inform targeted  marketing. But not everyone is honest about their location on Twitter. And while that may cause some of your ad dollars to miss the mark, the key to success is to develop a better understanding of how users signal their location.

The first problem for location-targeted advertising campaigns is “geographic drift.” The 140 Proof blog defines geographic drift as “the difference between your stated location and your observed location.” Many users imagine that a physical area is much larger than it is. For example, people consider Chicago to be huge. It’s circle of influence covers parts of four states, including most of Illinois and Michigan.

140 Proof cites four key reasons for the drift: users choosing ‘vanity’ locations for their bio, check-ins are more likely in denser areas, users elect to share GPS data, and IP addresses can be fuzzy when it comes to precise location. If you lived near Chicago, but not in it, you might list “Chicago” on your Twitter page — it’s just easier.

Other factors, like commuters, tourism or L.A.’s driving culture can also impact geographic drift, making it trickier to market to truly local audiences. But what about those who don’t use the methods mentioned above? Exactly how hard is it to pinpoint user locations? According to research from IBM [PDF], not that hard at all.

The researchers at IBM created an algorithm that compared the content of geo-tagged tweets to untagged tweets. Hashtags, text references to geographic locations and even Foursquare check-ins all gave away a user’s location quite readily.

To test the theory, researchers gathered 1.5 million tweets and used 90 percent of them to train the algorithm, and the results were pretty good. The algorithm could predict users’ cities 58 percent of the time, their state 66 percent of the time and their time zone 73 percent of the time.

So what does this mean for marketers? To users of social media, location is a pretty flexible concept. But if you’re able to read between the lines, you can find the right information to target the right consumers.

New Career Opportunities Daily: The best jobs in media.