Archive for August, 2012

Introducing: Highlight

August 28th, 2012 by Alex Carusillo

We like to think of ourselves as internet super geniuses and pretend that all of our ideas come out of our heads fully formed and fully great. But - truth is - some of our best ideas come from the stuff we see you guys doing. Sometimes we see what you're doing and carefully weigh and combine it with our usecases and research to release incremental improvements to our dashboards (yeah. It's as entirely boring as it sounds.) But sometimes we see you doing ridiculously awesome stuff and we just sort of steal it because it’s so cool and that's all that matters.

So welcome to our newest thing - Chartbeat Highlight - by us, but mostly from you.

For years we’ve been running into images of your dashboards everywhere. Whether it’s screenshots on Twitter, Instagram pics, or Facebook uploads when you guys are having great traffic days or dinosaur-shaped spikes....

(Photo credit: http://twitter.com/roboppy/status/70947457894326272)

….you’re making sure the world knows they’re happening.

And you should. Because all of these moments - from the Jurassic to the practical - are celebratory events that you worked stupid hard to earn (or, you know, you just happen to be in the right place at the right time and you got crazy traffic... because the web can be totally weird like that) and you should shout it out.

So we built Chartbeat Highlight to make it super easy for you to share that joy with everyone

The next time something really great happens on your site just click the share button on your nav bar and we'll get you going. We’ll kick you over to the Highlight screen, and show you a live preview of what you’ll be putting out there. Specifically we'll show your current traffic, your total engaged time for the day, and a general trend line: all the fist-pumping information you could want. Then let us know long you want this thing out there. You can set your Highlight to expire whenever you want - keep it out there for a bunch of hours or cut it off after one. After that's all set up you'll get a link (with an adorable chart.bt shortener) that you can email on to Mom, share with your legions of Facebook fans, or - if you’ve connected your Twitter account to Chartbeat (and if you haven’t do it now. We’ll wait...), you can write a quick message and tweet it out right from your dash. It goes without saying, but to be crazy crazy clear - unless you explicitly go through this process we won’t share a Highlight on your behalf. Your data is as safe as it’s always been. Similarly right now only admins can share Highlights - so if you're not already one bug the person who set up your account to get it out there. So go share your Highlights with everyone and get yourself internet famous.

Data That Matters – Interbrand

August 21st, 2012 by Lauryn

Data is everywhere. Big data, ambient data, real-time, benchmarking – there’s so much that there’s no one metric or one way of using it that works for every company or every industry. The data leaders are the ones who take risks, who look at all the information available and decide what matters right now. We’re spotlighting these innovators in this “Data that Matters” blog series. We’re talking to people in various roles across multiple industries to see how they collect, make sense of, and act on their data. Read the full series. Next up is Nora Geiss, director of verbal identity and digital for global brand agency Interbrand. In advising brands like AT&T, Feeding America, and Johnson & Johnson, Nora suggests the biggest success when working with and understanding data comes from remembering that data is people.

I think people can get so caught up in the fact that data contains numbers that they forget that those numbers are really about people – and that’s where the story, the context gets lost.

Brands are collecting so many numbers from so many sources - their in-house web teams, their CRM agencies, their ad agencies; these guys are telling them the specifics about how certain campaigns are performing. Which is absolutely useful to tackling discrete goals. But looking only at this granular level doesn’t really tell you about whether or not people in general get your brand. Do you stand for something? And ultimately, do people actually care about what you stand for? And you, as a brand owner, have the option to not care about all that. I mean, you can look at just the short-term numbers to see if you’re making money off an ad. But if you want to be around for a while, to have real longevity and relevance through economic shifts and changing markets, then you need to care about the bigger picture. If you want to bring people back to purchase your products and services over and over again, to pay a premium for your product, to be an advocate for your brand to others -- you have to care about the people and the story behind the data.

It’s not just about the moment “before purchase” -- it can be more important, I think, to look at what happens after purchase.

What did your buyer say? Did they write a review? Did they post a picture of their purchase somewhere to share with friends? That’s where you figure out if you delivered something that’s actually valuable, something that your buyer is going to recommend to their friends, and their friends’ friends or even their not-friends through things like Amazon’s “people like me” type features. The “people telling people” part of the loop drives more purchase and has more impact on brand value than anything else these days. So if you’re 100% focused on how people experience what you offer, and how they can share that experience, then your reason for being as a brand – to deliver a product that people recommend, ultimately driving purchase and loyalty – is more easily fulfilled. Right now I notice a lot of brands looking at measurements out of context. Brands are only seeing “someone talked about me” or “4503 people are talking about me right now” and just focus on blindly increasing that number. It’s just traffic and number of likes or followers or whatever right now.

Brands are just on the what, not the why – when the why is what gets into what people are thinking, doing, experiencing, saying.

You see a lot more when you get into the reasons why: you see that “someone talked about me because they saw this, they experienced this, they touched this” or “they talked about me to this person in this place.” Context is key. You can’t succeed if you don’t know the context. If, for example, my goal as a CMO is to drive premium pricing, and most people are talking about my brand in the context of coupons or discounts, then I’m failing – and I definitely don’t want the number of those mentions to increase.  So purely looking at numbers don’t help me much in a case like this. I think the brand world hasn’t moved past the numbers yet because, in large part, exposure to data happens every once in awhile and in an already numbers-heavy environment – as part of a quarterly report or an annual review of performance against financial goals. When you’re seeing the numbers day-to-day, the constant fluctuation makes you more apt to wonder what’s happening behind those fluctuations.

Brand owners need to live in an environment of habitual measurement.

An environment that makes it possible to take action that matters. You see successful brands taking that approach, like P&G and Dell and Starbucks. You see how they listen to what’s happening in the market, how they incorporate what they learn in their go-to-market strategy, how the audience is reacting to their reactions – the feedback loop. And I think you’re starting to see the change take root internally – it’s not just the marketing department participating, but the product teams, and the agency partners, which leads to much more agile internal process for everyone and everything. Feeding America actually did an amazing job of listening and delivering when they went through their rebrand. Every single person in that organization understands who they serve and what that audience needs most, so they know how to reach them, what to learn from the actions they take, and how to change the way they deliver accordingly. These guys doubled their revenue in one year after their brand relaunch. It’s amazing what can happen when you’re really paying attention to the right data – the data that tells a story about the people you serve. This is such a crucial shift. If your brand is only talking not doing -- using data and or feedback just in marketing and not using it to actually improve your products, quite frankly you’re screwed. 

Because your brand is not just what you say, it’s what you do.

 And if you set an expectation and fail to deliver – believe me, people notice.       We’ll be covering a new company each week – big & small, media & not, data junkies & analytics allergic – so let us know if there’s someone you want to see featured. Hit me up at lauryn@chartbeat.com.

Better API documentation coupled with an explorer

August 16th, 2012 by Etan Zapinsky

One of the goals of Chartbeat is to make really complex information really easy to understand - really fast. We keep this in mind when we build anything from a new feature on a dashboard to our heads up display and now our new API Explorer. Instead of creating static API documentation separate from the API explorer, which is confusing to navigate and takes way too long to jump back and forth, we bundled these two things together. So you have a seamless work flow when working with our API. Now, when you look at the documentation for an endpoint, you don't just see a unpersonalized template response, you can actually play around with your own data. It’s a preview of your data with our API. Awesome, right? Play around with a couple of these to see what I mean: These API explorer updates aren’t just an opportunity to explore the capabilities of our API, it also helps to directly integrate our API into your service. On top of providing the results of your API query, you’ll now see what URL would return those same results. From there, you’ll use the URL we provide to mimic the exact results you see in the response.

Now, let’s talk about collaboration. Instead of sending the actual API URL to show someone the response, we created a special syntax for specifying an endpoint and it's parameters within the hash of the URL of the API explorer page itself. This move allows you to show anyone a quick visualization of the information you plan on using without having to build anything. It’s basically magic. API wizardry. After you find the query you were looking for, just send the URL of the API explorer page to your friend/colleague/collaborator-in-crime, and they will be presented with an API explorer form just like you left it. We explain the response completely, so they won’t be bugging you about what a particular field in the response means. Lastly, we took care of a little housekeeping on the UX front for you: Since we want to make accessing your data as easy as possible, all your current API keys are at the top page, and by selecting one it automatically appears in the form. You can also manage your API keys directly from the explorer page. If you suddenly realize you need a new API key adding one is right there - where you’d expect it to be.

So what do you think? A little more helpful, we hope, especially seeing your data paired with the documentation. But you know us, always iterating, so let us know if you have suggestions on how we can improve our APIs.

Data That Matters – Apartment Therapy

August 14th, 2012 by Lauryn

Data is everywhere. Big data, ambient data, real-time, benchmarking - there’s so much that there’s no one metric or one way of using it that works for every company or every industry. The data leaders are the ones who take risks, who look at all the information available and decide what matters right now. We’re spotlighting these innovators in this “Data that Matters” blog series. We’re talking to people in various roles across multiple industries to see how they collect, make sense of, and act on their data. Read the full series. Today we’re hearing from Apartment Therapy’s executive editor Janel Laban. Janel makes sure data is in the hands of those who are editing and marketing content to focus on publishing the best creative possible for the home design and lifestyle blog. We have one core mission at Apartment Therapy. It’s simply to help make people’s homes more beautiful, healthy and organized. That’s pretty broad, I know. We’re creating lots of content - around 40 posts every day across multiple channels, trying to cover the spectrum of everything related to life at home. The things we could talk about feels endless at times. But reader data has completely transformed our perspective on how to focus on exploring and showing how we really live now.

We’re writing an encyclopedia on home life.

So we need that data, the evidence, to back up our growing knowledge of what's important and meaningful to our readers. That’s what really helps us to better contextualize how we define home. And even the most minimal data matters, really. I learned to blog six or seven years ago looking at very simple analytics when we first launched our local Chicago site. I used to sit and compare the best and worst performing stories using Sitemeter  - which was a challenge, to say the least - and learned from those performances in real time and over time. It changed how I looked at content - to know that something is working helps to define everything. Now we have the ability to use data both to instantly understand the present and to better plan for the future. It keeps us constantly aware of what is working for our readers. When I first saw Chartbeat, I called my husband on his way home from work and told him:

“I need you to pick me up another monitor.”

Because I knew I’d need to have it up all the time. That doesn’t mean every single post we publish needs to perform to a certain expectation or be optimized in real time, though. Some posts are more reference-related, and so situation specific; those don’t always do as well in real time. That’s ok, that’s not their job. They can do much better than average over time, because people search for, for example, "how to hang artwork," and that sends them to our post on it.

That’s how we bring in new readers.

People will still be finding and visiting that post months down the road, and while that doesn’t necessarily get captured in real time, we have other data that shows its strong performance in the long term. The nature of our website doesn’t require us to break news. Actually, we never have to post anything in a certain time frame. We're able to flow out our content in ways that we have learned is best for our audience. So we have to consider both real-time and historical data, because our most popular content might not be popular the second it goes up.

Our data is a little more consistent than breaking news, but I find, more surprising. We’re always learning.

But like the more news-based sites, the hard part of our job, and what we also strive for, is finding the people who don’t know us or who don’t use us regularly and just happen upon us. We want to keep those readers and find a way to make them loyal regulars. So, it’s a balance of the more intimate posts about people’s homes that are very engaging to readers and provide the backbone of what we do, and the quicker posts that can go viral and introduce people to our site.

We aim for a balance of the more image-based and inspirational vs. more spur-of-the-moment and easily digestible content.

And I work to maintain that balance.

Believe it or not, we don’t regularly review the real-time data with the writers. 

What we’ve found over time is that writers really only need to focus on the quality of their posts. They shouldn’t have to pay so much attention to how their work is performing in the moment. I want them to be inspired, have a long term plan, and a passion and vision that builds their writing, not data they’re constantly looking at to mold their point of view on what content they feel they shouldbe writing. But that’s because our writers aren’t changing or adapting or marketing their own content. The writers generally deliver their work to us, we do some final edits and then queue them up across our channels, and then share them through social media and our email. Once the posts are live, real time changes are performed by the editors, not the writers, because we’re seeing the data and tweaking accordingly. To be honest, we don’t do a huge amount of real time edits to our posts, but rather look at the data in the context of how posts work within the entire flow of the day, and then, the week, month and year. We use the things we learn from performance data to brainstorm new ideas, as inspiration and planning for the staff and to shape the scheduling and flow of content on both the site and the social pages we run. We’re using the real time data less as a means for making direct content changes and more as a learning tool to discover the best practices of timing, packaging and promoting our content to better serve, inspire and engage readers.   We’ll be covering a new company each week – big & small, media & not, data junkies & analytics allergic – so let us know if there’s someone you want to see featured. Hit me up at lauryn@chartbeat.com.

Data That Matters – The Blaze

August 6th, 2012 by Lauryn

Data is everywhere. Big data, ambient data, real-time, benchmarking - there’s so much that there’s no one metric or one way of using it that works for every company or every industry. The data leaders are the ones who take risks, who look at all the information available and decide what matters right now. We’re spotlighting these innovators in this “Data that Matters” blog series. We’re talking to people in various roles across multiple industries to see how they collect, make sense of, and act on their data. Read the full series.
Today we’re hearing from the president of The Blaze, Betsy Morgan. Betsy's previous experience as CEO of The Huffington Post and her current role building The Blaze - through an audience-focused, risk-taking, iterative approach to content creation - has been driven by the use of transparent, democratic data by every newsroom team.
I’m a big believer that unless the whole company understands your audience, then nothing is going to work for you.
At CBS, Huffington Post, and now The Blaze, we’ve always concentrated really, really closely on audience. Who are they? How are they interacting? Who’s new and who’s loyal? We think in the context of the audience.
At The Blaze, we’re looking at a potential audience that’s about 50+ million people; how do we retain those who are seeing us and how do we get new people every day? We have to have a really good sense of who’s coming to the site, when, and on a very immediate basis.
That’s why Omniture didn’t work: it wasn’t real time enough, and it was too backward looking. We need to understand the flow of things right away.
That’s why only having an audience measurement team doesn’t work. Everyone has to care about audience measurement and content, all the time, and in real time.
If tech is sitting alone, and product is sitting alone and editorial is sitting alone, and they’re solving problems discretely - then editors largely ignore the needs of tech, and vice versa.
It’s so key to take editing teams and tech teams and put them on a level playing field;  no hierarchy. And, everything is completely collaborative. It drove the growth at HuffPost, and we do pretty much the same thing here.

Having every team within your organization work together to solve a problem changes everything.

For example, in the last 24 hours, we’ve had two stories do uncharacteristically well with traffic coming primarily from Twitter.  The person who noticed the spike on Chartbeat wasn’t even the author or editor of the piece -- he was someone on the tech team. Which is great! Everyone is looking at our data at once. It’s no one person’s responsibility.
And, then you can try a lot of different stuff. You can take risks. If it’s just one person’s eye on things then you’re not solving big problems, you’re getting little things done.

The big problem with legacy media companies is that they’re trying to preserve audience. We wake up every day with the goal of growing our audience.

Everything is additive. The more iterative you are - and the better you iterate - the more your audience will reward you. It shows that you care about improvement, be it big or small. The iteration is the easy part. The hardest thing about data is that when a lot is thrown at you, how do you determine, what matters?
So, we’ve always concentrated on some very basic metrics. 
How many people come to the site in a month? How many in a day? How many people are on the site over a given hour?
If we’re doing things right, then everyone in the company knows those numbers. After that, it’s a drill down to find the problems. We try to figure out what we’re not doing right and how we can do it better, obsess over it, and then make some changes.
Eventually, that particular problem won’t be interesting anymore, and we’ll move on and fix something else.
This week for example, we’re obsessing over Facebook geographic data from a few specific cities.
It may be useless. It may lead to nothing. But usually it leads to something -- some other problem to solve or question to answer.

Data just starts the conversation, and that’s what matters.

One of the reasons companies don’t share data externally is because they treat data as punishment, which isn’t useful at all. Data as a pat on the back is useless too. Data needs to be used as a conversation starter - a way of getting people to think about things in a different way.
For example, our weekend traffic since the spring has been enormous, and that started a conversation about our audience.
Are they catching up on news over the weekend more? Do we need to add more recap and analysis stories?
One of the worst things an editor said to me a long, long time ago was: “Betsy, it's a slow news day/week/month.” What is that? Slow news day? Weren’t there tons of interesting stories published today?
That’s just not how we think about things here.
And, it’s not how sites that we admire think about things - like the Gawker or the Buzzfeed families. I've always loved them (especially Jonah Peretti, my former partner in crime at HuffPost). Both sites have always been really ahead of the curve, and we watch and learn from them all the time.
Of course, we also learn from HuffPost, who continue to do smart and innovative things that are clearly always informed by the data they have.
And, Flipboard, also, gives such a customized and relevant user experience for content. And, as a result, their audience reach is really far and diverse. They understand their audience.
When you think about innovators, you think about these guys: Companies that make decisions based on what matters for their audience, are hyper transparent about the data they collect, and who take risks - big and small.
We'll be covering a new company each week - big & small, media & not, data junkies & analytics allergic - so let us know if there's someone you want to see featured. Hit me up at lauryn@chartbeat.com.