Hi there and welcome to Scale. I’m your host, Stewart Ritchie, the founder and lead developer at Powered by Coffee. Powered by Coffee is a web and software development agency focused on helping media brands with open source and solving general technology problems as they come up. Scale is a podcast about the media and how technology impacts media, and how media is, in turn, impacted by technology. Today, our guest is Neil Paul, a solutions engineer with Parsely. Parsely is a content analytics platform and part of automatic. Neil, thank you so much for coming on being with us today and talking about content analytics. Tell us about yourself. How did you get here? What’s your story?

Thanks, Stewart. Yeah, always happy to talk content analytics and so you know, I started with a publishing background. I was at a BWD financial journalism pub and that’s where I first got exposed to Parsely. We used it in our newsroom and then I later moved on to a product management focus role on like a news front end and I installed Parsely analytics there and so that kind of brought me full circle and now I’m at Parsely talking about content analytics day in and day out, with both our current client base and prospective users.

Awesome. So from that then obviously a lot of experience, kind of using Parsely as a product from your time in those newsrooms.

But for someone who doesn’t know what is Parsely, Sure, yeah, parsely is a web based content analytics platform, and content analytics and web analytics the difference can be fairly nuanced, right, if you’re not living and breathing that every day.

Okay, great. So then a lot of our audience and most people actually, there first port of call with analytics, is probably Google Analytics right to multuous, as that is today with the transition between Universal Analytics and Google Analytics four.

Yeah, I think that’s right and I kind of like in it. When we talk about Google Analytics, specifically in the broad appeal there, you know, google Analytics is like the Swiss Army Knights, right. There’s a ton of different tools in the toolkit there and where, when we talk about content and we’re content creators, you know, parsely and other content analytics tools are the scalpel, so you’re not going to go into surgery with a Swiss Army Knight, right, and that’s that’s how I kind of take that approach.

Okay. So it’s just, it’s a process by which you are able to get to the insights that matter to you very much quicker. Your stuff for maybe you’d have to go into Google Analytics and set up specific reports to get it into being a content analytics system is just here. It is, and this works, and this is a product where we’re iterating on and kind of making better all the time. That does that make? Is that about it, right? Right? I think so.

And then you know, specifically for parsely, because we’re part of automatic, which is, you know, a WordPress shop and the founder of WordPress is the ultimate owner there you know it also has closer ties to your CMS, Right, and so when you want to use analytics to inform things like content creation, content channels, and be directly tied where someone is actually putting in content and making the content, content analytics has closer ties to that as well. It’s really geared toward creating content.

Okay, awesome. So it sounds, then, like parsely and it’s kind of content analytics world. You could do a lot of it within general purpose analytics platforms, but the precision given by having a scalpel, which is like actually really like that way of putting like a really precise tool for the job, you the value there is, kind of getting to that return quicker and understanding that and that better. So for someone, then, who content analytics is maybe a new field, versus just general purpose analytics where what’s the starting point, like what are the general things that they want to be looking for, that maybe they haven’t thought about, that they could start measuring, to start tweaking, that they could see value in their business if that makes sense.

Yeah, yeah. So you know, I think when you’re you’re starting to walk, crawl and run with content analytics, you know the first thing, I think, is really having a tool that can measure some of those things whether you invest a lot of time in Google analytics and really try to hone that for your content and have things like your metadata associated with your content, or whether that’s to invest in a content specific tool and really start to run and make business impacts quicker. You know, obviously have been able to start to measure that some way right and have and set your KPIs. You know the metrics that you care about and so you know that could be something like engage time. So how often are people actually reading the content? That could be something like recirculation rate, how people are moving people through the session.

Okay. So is this kind of along the lines of you know your hypothesis of like, if we rework our headlines from you know this day forward to be 20% shorter than we would have otherwise been, is that going to have a measurable impacting click-through rate from social media or whatever, whatever source into the article? And then, conversely, if, if we up that click-through, it does that then mean that traffic is less targeted, so that has maybe increased the bounce rate? So it’s finding there’s like little changes that you were making that have one positive impact but also possibly have two or three negative Impacts that could caskey it down. Because I could see a world where, like great, we’ve got short in these headlines as an example.

it’s really nuanced and you know we could use the headline example or we could use the example of word count.

Absolutely okay. Yeah, so that’s a really interesting piece to knock on from that then. So then, who and I mentioned this like varies kind of in Newsroom per provider, but who is? Who is this for? Is it for the writers that are like trying to, you know, craft stories and craft and get their point across in such a way, this to be like looking through, okay, well, this kind of thing works really well? Or for our audience here, are you X-men, zedrals? Or is it for the editors that are like looking across that, you know, yeah, that that’s Stanford’s, that whole group of people they’re looking after. Who generally is? Is this for? Who generally is getting the value from, from this kind of tool?

Right? Yeah, it’s great question. I’m a data nerd, so I believe in democratizing data, right, I think everyone’s, no matter their role, either, as you know, a beat writer, an editor, a Newsdust manager you know that, you know, encompasses the whole thing or your executive editor, I think your content analytics is available and valuable to each person, each role. But what you choose and your scope of observe metrics differs, right, and so If I’m the executive editor, maybe I care top line on numbers of our content engagement as a whole, right, you know how is the entire health of the system working? How is that impacting our revenues? Right, but if I’m a beat writer, you know I’m more, potentially more focused on improving my individual content, growing my engagement on the channels that I interact with and really creating the best content that I could create.

Yeah, I wonder if. So, when it comes to things like, I guess I always get a little like the data is one thing, the data is telling us one thing, but is that the right thing for us going forward? You mean, like we’ve spent a bit of time looking at personalization recommendation engines recently for a reason and you give it a lot of content. It spits back the options and that it thinks this user will pick, and you tell it which of those it picked and you give it like a yeah, you got it right or no, you didn’t get it right, kind of things. So it learns to like do whatever it is this going to do.

No, no. It’s an interesting topic and I think in in in my field there’s People when they’re talking about making decisions with data, we’ll say data driven, or they’ll also say data informed right to very, very similar Topics and like kind of use of language, but a little bit different.

Yeah, I like that distinction data informed over data driven. I think that’s like a really great way of looking at it actually. But then the other thing with them is and maybe this is more of a partially specific question, but you talk about having kind of like a content to view, kind of your content, everywhere it is is. Are these platforms in particular able to kind of pull that information from other places? And Google Analytics, for example, because it’s the one I know most. You have to have like installed your profile and like kind of given all the tags and stuff to be like get this information from here. But when we come to things like social media like is partially kind of able to reach in and say like your engagement with this tweet is, you know, higher or lower than average X, y and Z or other platforms such as YouTube, pulling in things from you podcasts and stuff like anywhere else that content exists, or is it very much a on owned property kind of tool?

Yeah, and that’s one of the distinctions from like a GA and more specific content analytics tools. There are tendrils into other places, right, so we can look to understand. When I create a post on a social network, you know we hook into some of their APIs. Right, so I can say, if I post on Facebook, for example, you know I’m getting this many comments likes on that post and so we can actually tie some of that social engagement together. And it varies on social network, right, some of those properties are more, you know, conservative in what they share, right, with other you know places some are not right. But then other things, like understanding how people are interacting with web applications or aggregators right, so, like you know instant, you know Apple news, that sort of thing. You know, we also hook content analytics and also hook into that partially specifically and give you a little bit broader of a purview and just your own and operated domain because your content’s out there, right, and so you want much of a complete picture as possible in this day and age. Yeah, absolutely.

Because it always strikes me as being incredibly overwhelming to have you know your content. You know how your content is performing on the site, how your content is performing in 5, 6, 7 other you know social media platforms, how it’s performing in your email provider assuming it’s kind of gone out as part of a newsletter and, I suppose, a good quality analytics. You know content analytics, so it’s going to be able to pull that all together and kind of show insight across across those different channels as much as possible.

Exactly, and you know it’s about giving you that understanding of your audience. Some people in your audience might, you know, visit your own and operated network. Very rarely. Some people might stem a headline right, and so there’s some ways that we can look to bring more of a complete picture together for you so you can make the best decisions about what your audience is doing, because, like you mentioned at the start, you know there’s unintended consequences, there’s knock on effects, and so you need to, you need to think carefully about you know how you want to change your content to meet your audience.

Okay, absolutely Cool, it’s a really interesting one. I think, like that knock on effects is like is kind of where I get like derailed when thinking about these things. I’m like, yeah, we did this and it changed that, but I don’t have enough data set up to be like looking for the right metrics to see if that had those knock on impacts. And that’s where I personally like blown off these things of like it’s too complicated, I’m gonna. I’m gonna stop.

Yeah, cool, and that’s really what we’re using.

Okay, Cause that time from universal analytics is either until the session expires or they’ve left the page in it.

That kind of like exit event fires and you can say of yeah, you can set like a maximum session time and pieces like that in universal analytics and you can tweak that a little bit, but it’s not. It doesn’t really get that interactivity because it’s the clock starts, the clock stops, right. Yeah, yeah, absolutely.

Yeah, okay, I mean, those are, that’s really interesting and I imagine then that’s really easy with like a video that’s kind of playing on a page because you’ve got kind of starting, end events and things that are all firing. But that kind of content, textual content, like reading, like it’s good. So is that then able to tell, say, like that I’ve left this open in a tab but in a browser window, that’s just not mind me in focus. Like like I’ve got a text editor open here, so on top of my pay, top stack of my browser, but the window I’m talking to you in a second. So like how would it impact with?

obviously things aren’t perfect, like that’s a real weird edge case, but oh no, that’s actually fairly common, and we, we, we make that distinction. And so, if it’s not your active browser tab that that hard beat picks, those are running.

Okay, cool, that’s good to know, that’s really interesting. So then there’s a lot, there’s a lot here. So we’re looking at engagement time as kind of the primary metric that is like useful to sort of measure and kind of see what people are doing within, within your content. And then I suppose the game then after that is what do we do to improve, improve, engage time? Are there any more kind of like experiments that you, you see, work well for people just getting started with this kind of thing? You obviously mentioned the total length of that article. It’s a good one, but what? What are other common things that people, people try in this space?

Yeah, yeah. So I think a lot of it is around testing how impacts of your content are driving gauge time, but then there’s also a piece of recirculation rate and being able to think about how the structure of my content impacts people following the next story, the next move Right, and so that journey could be through. How we structure our homepage Right, so you know a lot of content analytics systems, personally included, like a, an overlay that allows you to track, slot tracking. So if I have something in the first slot, how is that compared to a piece of content in the fifth? What happens when I move content from the fifth? And how does that, you know? How does that perform compared to everything else that was in that slot?

Right, okay, that’s really interesting. So then there’s a whole element there. So when I hear that like my head and me to the goes, okay, so there’s like an element here of driving personalization of like this kind of person, when exposed to content types A, b and C, end slots one, four, six and eight on this page are more likely to to sign up. Or is that too much, too much depth to really be a useful thing to look at? I suppose that depends on how big the organization is and how much data they can achieve both in the click-throughs and their audience.

It depends on your audience.

Cool, that’s awesome. So maybe more parsely specific then. But so that element of like that curation piece, like tradition, that would be done inside the content management system and you’re building out you know the flow of that homepage. Someone is picking kind of what was going into those, those articles is given parsely’s relationship with a large content management system you know be a scum Is that? Is that able to work the other way, are you kind of able to put in place UI within parsely that does that like curation or really nodding along?

So yeah, you won’t use the dashboard and drive that curation, but we actually have, you know, tools and API that can power your front end design, and so you know it could be general right?

Awesome. And so then we’re talking kind of like and that kind of like oh, this person did X, y and Z. And that brings me to the inevitable questions that are in privacy and cookies and stuff. And you know, in the UK, in Europe, so there’s huge like data protection laws and stuff like that. And I know, please, you’re in California, no, you’re not, you’re the other side, but company or workforce be a set of California. So there’s huge like huge data protection laws kind of coming in California as well. How GDPR had like a big impact on marketers, particularly ability to like target and focus on their audience. Is there the same, was there the same impact in content analytics? And how important is knowing about the user to content analytics or is it able to be done well in a privacy respecting way?

Yeah, I think privacy is talk of everyone’s mind and I think the curation can be done in a good way while respecting privacy, especially if you think about your audience critically.

Absolutely. I think that’s an way to spot on, because that user has not just opted in, but offered up and made themselves known to the organization of. I’m this person. I have opted in to your personalization. I have an account here as my data, rather than the very aggressive. If you viewed this one article, it falls into this bucket.

Now everything you see will be this Now I think that really sets a high bar for us and makes content analytics that more important, because it’s up to us to help encourage that person to become known, to give us that email to register. That’s a higher bar than where it was when we first started to digitize content. You could have cookies that would track across multiple platforms and third-party cookies. So I think that makes content analytics even more important and it makes us, as content creators, really set the bar in terms of getting someone to do something. I like this. One of my old editors would say your job isn’t done when you press publish, it’s when you get people to actually read it. That’s where content analytics can come in and say are people reading this? Do I need to make changes in the moment? Do I need to think about how I create future content to get better engagement?

Yeah, absolutely. We’ve just recorded an episode with Brian Alvey, who very much is on, and I’m not sure if it’s going to come out before or after this one. So if you’re listening, maybe you’ve already heard it or maybe it’s the next one. And if you don’t know who Brian Alvey is, he was very important in the start of digital media. He was a key player in weblogs and getting modern blogging and what we do in modern media happened out.

Unsurprisingly, brian and I are in lockstep there and I think that applies to content quite heavily. Right. It’s a you know make your best judgments. You know, create quality content, but don’t try to take down every single screw and have something that’s you know home perfectly, because you’re not going to be able to do that. You need to have things out there. You need to let your audience start to tell you thanks, and I think personally, like you know, locking something down and focusing on something, so putting it out into prohibitive, and getting the right feedback and matching to what your audience thinks right. Like your audience’s preferences and your idea of an audience’s preferences, they don’t always align. Let them tell you, right, they’re trying to tell you what they want.

Absolutely, that’s awesome. And then you know we’re kind of getting towards time, so kind of like Russian this me a little bit, but we’ve talked a lot about how this is so useful for you editors, writers Getting the right content, working on engagement, building the audience, everything like that for the business side of of a publisher. So you traditionally Google Analytics For some stuff, then you, whatever their particular platforms they use for monetization, will have some insights and that’s probably all pushed up depending how Sophisticated organization is into you know some kind of business intelligence platform, but I feel like overkill. So is you know content analytics for for that upper management, for the commercial site? You know I imagine there are uses, uses there to the UC, yeah, and I think that’s where the conversion element comes in right and being able to tie content to the conversion, right.

so you know, maybe In the past we’ve had the capability to know, just like our subscriber numbers, and we’ve got like a last touch conversion model which says, okay, people sign up on the subscription page.

It was fantastic. So then, and just new work coming to, and every time, just to kind of like summarize for our listeners, I suppose Content analytics distinct from you, universal analytics, general, like whatever, in that they are very much like. Here is the specific things around your content, like how is it performing, what is the engaged time, you know, what is the recirculation of retention, how many people are coming back? Leading on to this and this, and that’s you to enable publishers, writers, content production team to do better work, to better communicate and better engage with their audiences is not strictly about finance, but Doing that work well will lead to financial rewards, incentives from the business side. It’s more about like this content led to this conversion through a funnel, getting away from a single touch, like they were on this page and the subscribe is more. They were on this page and then this page and they generally read this lot of stuff and we seeing a trend on this.

Yeah, if people are interested in learning more about content analytics, specifically parsely – parse.ly is our website. Got great resources and blogs where you can learn more and start to dip your toe in, and then you can also check out the parsely LinkedIn, which is posted on a lot. So check those things out.

I always get it wrong. I always put in parsely the herb and they’re spelled very slightly different.

I’m like I need like an automatic redirect some high staff in my browser.

Thanks again, so much. Thank you for listening and if you’ve enjoyed the episode, please make sure and leave a review on iTunes or Google Play or wherever you happen to pick up the podcast. If you know anyone that you think would value from this episode, please share it. We would love to have it and let us know what you think in the comments on Twitter. Wherever you’re on social, you try and find us and let us know, and we will hopefully speak to you again in two weeks. Thank you very much. Goodbye.