You can’t know how successful you are — or even if you’re succeeding — unless you find a way to measure performance.
I’ll be leading a workshop today on metrics for Digital First journalists in Connecticut. It’s scheduled for the Middletown Press, but if the weather turns nasty, I might stay in New Haven. Either way, you can follow the workshop on a livestream and live chat. The workshop starts at 4 p.m. (but if I’m traveling, weather could delay that slightly).
I’ll start with a process journalists should use to measure their digital efforts, then I’ll discuss some possible ways and tools to carry out that process:
What and how should you measure?
Here’s the metric process I recommend for every Digital First journalist:
- Identify an important goal for your position and discuss the goal with your editor.
- Consider what measurements would show how well you were achieving your goal.
- Explore what tools are available to provide that measurement.
- Consider how much money or time it would take to provide that measurement and adjust accordingly.
- Consider what the measurement shows about your performance (and what it doesn’t show).
- Repeat steps 2-3 to provide metrics in areas that the first tool didn’t measure.
- Repeat steps 1-6 for other important goals.
- Review your measurements periodically (monthly would be good) to assess your performance.
- Determine whether and how you should change what you’re doing.
- Determine whether and how you should change what you’re measuring.
- Share and discuss that review with your supervisor.
- Keep in mind that metrics aren’t perfect.
Align metrics with goals
We haven’t yet developed ideal metrics to show the success of our efforts at Digital First. Part of your effort to measure is to understand the value of any metric you use: the strengths as well as the weaknesses. You want to understand both the performance that a standard measures and the way that use of a metric might change behavior.
Many years ago, I worked for a publisher who wanted to measure my reporting staff’s productivity by counting bylines. If my reporters were not working hard enough and writing enough stories, counting bylines would be one way to measure whether they were working harder. But they were all working hard and writing plenty of stories. In fact, the publisher and I had agreed that we needed to do more enterprise reporting, and that usually means fewer stories. I noted to the publisher that counting bylines would encourage writing quick daily stories, puffing stories that should be briefs into bylined stories, rather than spending more time on enterprise. You need to tie your metrics to your goals. I proposed instead that I designate which stories counted as enterprise stories and we track the increase in enterprise stories. (On the other hand, if you have a reporter who isn’t writing enough stories, counting bylines can be helpful.)
You also need to understand the context of the metrics. If you measure reporters’ performance by page-one stories at the same time that you shrink and redesign page one to allow fewer stories there each day, you’ll have fewer page-one stories, regardless of how well reporters are performing. Or if you decide to emphasize local stories more than national on the front page, suddenly you’ll have an increase in page-one stories by the staff, whether they are doing a better job or not. And if you’re counting front-page stories anyway, you’re measuring print performance, not digital performance.
I don’t claim metrics as an area of expertise, but an area of continual learning. In more than 600 blog posts, only seven show up with the word “metrics” in them, and this will the first where that is the topic of the post. But I do evaluate which stories attract the most traffic on this blog (four months in a row, I did posts noting that I had set traffic records and attempting to analyze the reasons and draw lessons from them). When I was training full-time, I attended a seminar on measuring the effectiveness of training, and did some research and blogging about the topic. I think measurement of performance is important, even if it is imperfect. Even the experts in metrics are constantly working to develop better ways of measuring performance and to understand the limitations of the metrics they use.
The point of today’s workshop will be to help journalists start to measure their own digital performance, and to understand the importance of measuring our collective performance, in different areas:
Revenue
I make no apology for stressing the importance of business success. Revenue success means more jobs for more journalists. It means we can give raises and bonuses to staff members to reward performance. It means we can stay in business and keep serving our communities and our nation with quality journalism.
As with any measure, you need multiple metrics to truly understand Digital First business success. We want to know overall figures such as total revenue, operating expenses and EBITDA (earnings before interest, taxes, depreciation and amortization), also known as cash flow. But we also want to know how we got to the revenue figures. Print advertising revenue has declined nationally six straight years, falling about 60 percent from 2005 to 2011.
A key metric for our company now is the relationship between the decline in print ad revenue and the growth in digital revenue. As our Senior Vice President, Local Digital Sales, Adam Burnham, recently wrote, we are trying to reach our crossover point, that point where the growth in digital revenue exceeds the decline in print. So those two figures are key metrics right now in measuring our business success.
While the measurement of journalists’ performance doesn’t tie as directly to revenue performance as ad sales reps’ performance does, I think it’s important to start a metrics discussion with recognition of the importance of measuring revenue success.
Traffic
Just as you measure the newspaper’s print reach in multiple ways (total circulation, subscriptions, single-copy sales, readership, penetration), you need multiple standards to measure your digital reach: page views, unique visitors, growth in either of those figures, percentage of traffic coming from out of your market (in-market traffic has greater value to local advertisers).
You want to know traffic patterns: Web traffic tends to be highest during the workday, but iPad traffic is higher in the evening.
I think journalists should know the page-views of the stories we write. Not so we can skew our coverage entirely and start writing only celebrity gossip. But we should know which stories resonate most with readers, and this knowledge should contribute to our coverage decisions. If the stories we think of as important capital-J Journalism aren’t attracting readers, we need to rethink whether we need to promote them better, display them better or write better headlines to have a better impact with our most important journalism.
Sometimes the traffic numbers will help guide us in deciding what to stop covering and where to increase our coverage.
Social media
Lots of metrics exist for measuring social media: Your Twitter profile provides the number of followers, how many you follow and how many times you’ve tweeted. Facebook Insights provide measures such as likes, reach and how many people are talking about your page. Omniture or your blog analytics can show how much traffic comes from social media sites. (Twitter is always a leading source, if not the leading source, for me on days that I have posted new content and tweeted about it.)
Each of these measures has value as well as limitations. The number of Twitter followers, for instance, doesn’t tell you how many are in your community, how many are friends and colleagues and how many are spammers who were just trying to get you to follow them back and will never read one of your tweets.
Tools such as Klout and Tweetreach attempt to measure influence and reach, providing more sophisticated (but still imperfect) social measurements.
Engagement
Joy Mayer of the University of Missouri, published a report on different ways to measure community engagement.
If you are reaching out in person in the community, you may want to count the number of people attending events. You might count the blogs in your network, referrals to your site from the blog network or referrals from your site to the network. You can count comments on stories and blogs, people joining, replaying and commenting on live chats and liveblogs.
How are you measuring success of your digital efforts? What are tools you have found helpful? What are the gaps in your measurement efforts? How have you adjusted based on metrics?
Here are slides for the workshop:
Good stuff, Steve. Thanks. I shared something similar with my staff last week. Looks like we are running in the same direction. http://www.yorkblog.com/ydrinsider/2012/02/29/what-does-success-look-like/
LikeLike
Here’s five ways to measure Twitter traffic: http://www.searchenginejournal.com/5-tools-to-measure-your-twitter-activity/8116/
LikeLike
[…] The Buttry Diary: Introduction to Digital First metrics: How do you measure success? […]
LikeLike
[…] I blogged yesterday about metrics, and since Chris March mentioned my posts noting my monthly review of my blog metrics, I should […]
LikeLike
[…] CLICK HERE to read Steve Buttry’s blog post on the topic, including his notes for the workshop. Like this:LikeBe the first to like this post. […]
LikeLike
[…] final note: This blog post is a great example of how misleading metrics can be, if you only look at one measurement, or if you place exaggerated importance in a single […]
LikeLike
[…] Introduction to Digital First metrics: How do you measure success? […]
LikeLike
[…] as well as reporters with multiple beats. I’ve been thinking of writing about training and how you measure digital-first success. I’m planning posts about the digital-first approaches to breaking news and engagement. What […]
LikeLike
[…] I won’t pretend that analytics are an area of expertise for me, but I have blogged about matching metrics with goals and understanding […]
LikeLike
[…] my powers in analytics are less than super, but I know a little and I’ve blogged an introduction to metrics as well as posts about misleading analytics, digging below the surface to understand metrics and […]
LikeLike
[…] my powers in analytics are less than super, but I know a little and I’ve blogged an introduction to metrics as well as posts about misleading analytics, digging below the surface to understand metrics and […]
LikeLike