Turning to Plain English to Make Metrics More Accessible

    by Jeff Weisbein
    March 31, 2017
    Mural from Unsplash

    The following is a guest post from KYA.

    I’ve been in the digital media space for more than 10 years and one thing is clear: the state of publishing and content creation is constantly evolving. Now more than ever digital publishers need to experiment with new tools and ideas if they plan on remaining relevant. In the simplest of terms, to find success in this digital landscape, you need to know your audience.

    "Did you know that over the past 6 months your Page Views are highest when you post 2x a day?"

    Jeff Weisbein, Founder and CEO of KYA


    If you know your audience you can tailor your content strategy appropriately. Then you can also use the data you have about your audience to run your business: to sell targeted advertising or to increase the paid subscriptions necessary to keep your lights on.

    But getting to know your audience is a lot harder than it sounds. Seriously, the majority of analytics products out there are inherently complicated. Analytics are inherently complicated. The fact is, while more digital publishers are hiring data scientists, most cannot afford to pay someone to make sense of it all. And really, they shouldn’t have to. Everyone who can benefit from analytics should be able to understand it.

    What if using analytics could be like reading a story about what’s going on with your site and its audience?


    At KYA, the analytics company I founded, we took this on as a challenge. We wanted to provide a better way to offer insights to customers, so they could spend less time staring at numbers and more time doing what they do best: creating engaging content. The feature we came up with we call Smart Insights and it works like this: we have come up with hundreds of different computations to crunch our customer’s site data with the goal of providing them smart, actionable insights in plain English.

    Analytics aren’t used as much as they should be

    Smart Insights provides publishers with unique and useful inferences from their data without them having to delve in directly.

    We started developing this feature by asking ourselves how we could better leverage all this data we’re collecting for our customers. What kinds of insights would be most effective and helpful? We ended up deciding that, for starters, the best place to start would be to identify big picture insights that we could offer up to customers. Insights that are not immediately obvious by just looking at our analytics — those are the kind we wanted to surface up to the top for them.

    Once we had a grasp on the types of plain English insights we wanted to offer to customers, we broke each type of insight into an actual question that we wanted to answer. For example:

    • What’s the best day of the week to publish based on engagement?
    • What’s the best time of day to publish based on time spent on page?
    • How often should you publish to maximize interactions?

    We just kept coming up with questions we wanted answers to.

    Learning how to ask questions of the data

    After we had a list with all our questions, we did some research, as there are a few ways we could have approached implementing this new feature. One of the things we spent a fair bit of time experimenting with was IBM’s Watson, which while cool, ended up not being the best solution for us (at least for now). In the end, we ended up crafting our own computations to generate each insight. That being said, Watson helped us see the data in new and unique ways — we were able to take some of the insights it extracted from all the data and come up with a way to compute those insights on our own.

    In the end, we had some incredible insights being surfaced. Take a look at a small sample of them below:

    We believe these types of insights will become crucial, especially as larger digital media companies expand access to analytics within their organizations. They will also prove immensely helpful for smaller companies and individuals who cannot afford to hire a data scientist to make sense of all their data.

    Each of these messages surfaced by Smart Insights is calculated using algorithms that analyze the data KYA collects on behalf of our customer, and the result is a simple sentence written in plain English that anyone can understand and act on. Now that we’ve figured this out, we’re planning to add more features to help digital media companies better understand the relationship between their audience and the content they create.

    We will continue to invest in our Smart Insight feature as well, plans to improve it are already on the product road map.

    As one of our existing customers wrote to us, “OMG this is so dope!!! Thank you.”

    Jeff Weisbein is the founder and CEO of KYA, a media analytics platform dedicated to helping publishers track audience, content, and engagement.

    Tagged: analytics analytics dashboard KYA media metrics platform

    Comments are closed.

  • MediaShift received funding from the Bay Area Video Coalition (BAVC), which receives support from the Bill & Melinda Gates Foundation, to launch the MetricShift section to create a vibrant hub for those interested in media metrics, analytics and measuring deeper impact.

    About MetricShift

    MetricShift examines the ways we can use meaningful metrics in the digital age. We provide thoughtful, actionable content on metrics, analytics and measuring impact through original reporting, aggregation, and audience engagement and community.

    Executive Editor: Mark Glaser

    Metrics Editor: Jason Alcorn

    Reader Advisory Board

    Chair: Anika Anand, The Evergrey

    Brian Boyer, Spirited Media

    Clare Carr, Parse.ly

    Anjanette Delgado, Gannett

    Hannah Eaves, consultant, Gates Foundation

    Alexandra Kanik, Ohio Valley Resource

    Ian Gibbs, Data Stories

    Lindsay Green-Barber, The Impact Architects

    Celeste LeCompte, ProPublica

    Alisa Miller, PRI

    Connect with MetricShift

    Facebook group: Metrics & Impact

    Twitter: #MetricShift

    Email: jason [at] jasalc [dot] com

  • Who We Are

    MediaShift is the premier destination for insight and analysis at the intersection of media and technology. The MediaShift network includes MediaShift, EducationShift, MetricShift and Idea Lab, as well as workshops and weekend hackathons, email newsletters, a weekly podcast and a series of DigitalEd online trainings.

    About MediaShift »
    Contact us »
    Sponsor MediaShift »
    MediaShift Newsletters »

    Follow us on Social Media