The major problem with the example from the previous post was that it only looks at a single threshold when computing accuracy. If we choose the threshold between the [A’s and B’s] vs the [C’s and D’s], we do not see a difference between the two models. if we had chosen instead the threshold between the A’s and [B’s, C’s, and D’s], we would have seen a difference.…
In the last post we introduced the idea of an evaluation metric for a classifier. In this post we’ll look at the accuracy metric.
The accuracy metric is one of the most common evaluation metrics. It looks at a model’s accuracy, or the portion of the data that the model classifies correctly. This is related to the error of a model; if a model has 90% accuracy, the error of the model is simply 100% – 90% = 10%.…
Missed our webinar on Wednesday? Well have no fear, in this post I’ll give you the same walk through I gave all the folks on the line on how to connect the dots from Fliptop to Marketo to Facebook in order to make Predictive Advertising work.
Let’s walk-through one real-life example of how marketers are already benefiting from predictive-enhanced ads.…
This series of posts will discuss several different metrics we can use evaluate the success of predictive machine learning classifiers. This first post will focus on the accuracy metric, and the next post will focus on ROC curves and the AUC metric.
In the first post in our series of machine learning blog posts, we discussed classification models. These models take as input a data point (for example, a lead we want to score), and outputs the predicted probability that the data is a member of the positive class.…
The explosion of big data has been both a blessing and a curse for modern B2B marketers. A blessing because, in one way or another, every time a prospect interacts with your brand—through your site, ads, emails, etc.—it’s now tracked. Because prospects interact with so many touch points along the path to purchase, even if an interaction doesn’t translate directly to a conversion, every touch is producing valuable data indicative of prospect intent.…
Predictive is becoming one of the most talked about marketing and sales technologies this year. According to TOPO, predictive analytics technology is the second highest buying priority for sales development in 2015 with 36.8% of companies planning to evaluate vendors this year. Everyone seems to have their own idea and opinion as to how the technology actually works
We’re here to set the record straight.…
The modern marketer’s technology stack is complex and diverse. Quite often MarTech tools are not easy to use or understand, requiring substantial training, support, and time in order to gain any real value. This leads to one of the biggest problems that B2B software companies face—the amount of time a marketer needs to invest in new tools, or even evaluate them in the first place, eliminates the software from consideration and purchase.…
Originally posted to CommPro.biz
As a marketer, you’ve heard the legend of marketing automation. The tale goes that there’s software out there that can move prospects along the buyer’s journey, delivering highly personalized experiences, all with minimal effort from marketers. The story is a promising one, and one that many marketers have already invested substantial time and budget in, but it’s an incomplete one.…
After many trials, we know a thing or two on how best to pick a predictive marketing vendor. If you’re new to the space, it can be difficult to pinpoint key functionality that will align with your company’s sales and marketing objectives. To help you out, we’ve compiled quick list of the Dos and Donts of evaluating a predictive marketing vendor.…
News flash – your best leads are already in your database. But do you know where to find them?
You’ve worked hard and spent precious money to bring leads into the funnel, but did you know that your most valuable leads could be sitting, untouched, in your database? If you’re in sales, have you ever had the feeling that you can’t fully trust your lead scores?…
Get updates on our latest articles
Thank you for subscribing!