You know that you can find information on just about anything you want on the Internet. But are you always finding all of the information that you need on a particular topic? When it comes to content marketing, you need to know exactly how often you should publish, exactly who is reading your posts, and exactly what you need to do to boost your readership (and, hopefully, your income). The key to finding these types of information is doing original research.Today’s guest, Andy Crestodina, is the co-founder and strategic director at Orbit Media Studios. He has done some great original research on publishing frequency, and he’ll be talking about that. He’ll also give us some tips on conducting our own original research. Stay tuned for some excellent hints that you can start using now to make your content marketing even better.
“If we produce something original, that will have a better chance in rising above the noise.”
“Ask [yourself], what statistic or what data point would strengthen my sales pitch? Then go find that data.”
“Basically, persistence wins. It’s the most powerful force in the universe.”
Nathan: Is there a correlation between how frequently you publish and your success? For a marketer like me, it would be amazing to know the answer to that question and it would be even more amazing if that answer was 100% backed by data. And do you know what it would take to find that answer? It’s called original research. That’s why I’m chatting with Andy Crestodina today. He’s the co-founder and strategic director at Orbit Media Studios, and he’s recently published some amazing original research about how much time marketers spend writing blog posts, how frequently they publish, and a whole lot more. Andy’s on the Actionable Content Marketing Podcast today to share some insight from his original research. You and I are also picking his brain on how to do original research ourselves to share data with our own niches. You’re about to learn how to gather research, analyze your results, and what some of those benefits are that you’ll get by publishing your own original research as a content marketing tactic. Let’s hear more about it from Andy.Hey Andy, thanks a lot for being on the show today. We’re really excited to have you.Andy: I’m glad to be here, Nathan. Thanks.Nathan: Andy, could you just start by filling me in on Orbit Media Studios and what you do there?Andy: Sure. I’m one of the co-founders, it’s a team of 38. We are quite simply a web design and development company in Chicago. This is year 16, so we’ve done more than a thousand projects. It’s a more strategic approach than most, we are very much focused on outcomes. It’s all about building things that are optimized for search and optimized for conversion and understanding analytics and aligning it with the content marketing and other marketing goals. That’s what we do. It’s a development shop here in Chicago.Nathan: I’ve been following what you guys do, we’re pretty excited to be chatting today. One of the projects that you just wrapped up was the Annual Blogging Survey and that’s what we want to talk about today. I was wondering if you could just give me some background on that survey.Andy: Sure. We know that there’s insane competition for content in marketing. We know that if we produce something original, that will have a better chance in rising above the noise because you will be the primary source for that topic. It basically involved deciding in advance that we wanted to fill in the missing stat, find the missing stat and sort of be a source for a data point that didn’t exist before.We just looked at the industry and said, what do often say but rarely support? And realized that of course everyone says marketing takes time, blogging takes time. No one ever really answered the question how much time does it take to create a blog post. We just decided to be statistically relevant, we had to get a thousand respondents at least, so we got a thousand respondents to the question how long does it take to write a blog post. We found out that the average answer was something like two and a half hours. While we were at it we asked 11 other questions, that was two years ago, so now we just finished the third, we have longitudinal data and trending data over time, which is all extremely interesting. You can see this and you can really see how the industry has evolved. It’s totally fascinating.Nathan: Yeah. I think that’s one thing that I wanted to ask you about. There has got to be some interesting stats, especially through the years, and you mentioned that two and a half hours on a blog post, that’s interesting to me because if time was an indicator of quality, I think two and a half hours would actually maybe be even a little low.Andy: You’ve actually just now touched on one of the main findings, that is about quality and that people are going deeper. There’s another really much less expected finding that we can talk about too but I absolutely agree with you. I’m in that category. Let’s say that you and I, taking this would probably answer the average blog post takes six plus hours to create. Nathan, are you in that range?Nathan: Yeah, definitely six plus. We write, we design and do research all of that. It definitely adds up quickly.Andy: Yeah. And your content is awesome. The editors write all of everything that goes into outreach; the planning, using data to decide what to write about. You and I would be in that category of people who in 2014 there were 5% us that put that kind of time into it. 2015, it went up to 6%. Now that has doubled and the percentage of bloggers who spend six plus hours on a typical post is up to 12%. We might’ve just been a little bit ahead of the curve but that is a big finding. It’s just that people are spending a lot more time on each specific piece of content today than they were before.Nathan: I think that’s a really interesting finding. What else have you discovered or are there some other interesting trends that you found through this research?Andy: A lot of it aligns with that. More people are using a formal editing process, more people are writing longer content. The average blog post now is more than 1,050 words. More bloggers are using analytics to check results and measure performance, but the surprising finding was that we thought we would come up with less is more, quality over quantity, and it’s true that a lot of bloggers are publishing less frequently. The percentage of bloggers who said they are publishing daily has dropped by 50%. Meanwhile, there’s a spike in the percentage of people who said they blog monthly.But when we asked is your content driving strong results? There’s an imperfect correlation between frequency and self reporting of strong results. People who publish daily are saying they get better results than people who publish weekly, who are getting better results than people who publish monthly. There’s a finding in here that basically, quantity and frequency matter, they matter a lot. They’re actually strong correlations with strong results.Nathan: Yeah. I really like that finding. I mean, as marketers, we talk about consistency a lot and I think your data that you’ve just come up with backs up that claim.Andy: The weakest are the people who said that they blog at irregular intervals. Those are the people who are least likely to say, only 14% of people who say that they blog at irregular intervals also report strong results. I’m an SEO guy from way back, I pay attention to things like authority. I don’t know a single blog that has a domain authority of 80 or above unless it’s a daily blog. There is absolutely a correlation between the amount that is being published, the frequency, the consistency and results; it’s just undeniable. It doesn’t mean that we should be sacrificing quality at all but it can’t be ignored that there is a factor here, it does matter how often not just what percentage what swings our home runs but how often we go to bat.Nathan: I think that’s a really interesting finding too because the more you publish, if it’s really quality content, the more opportunities you have to reign for keywords to increase that domain authority to your point.Andy: Exactly. Let’s say 5% of your articles are those amazing, home run, unicorns that get far better than the average results. Assuming that that number doesn’t change, which it should, you should get smarter based on your data but let’s say that it’s only 5%. The more often you publish, the more likely you are to hit that one out of the park. Also, I think that there’s a really good point here that Sean Callahan, one of the big bloggers on the LinkedIn B2B blog, he writes for LinkedIn. Shaun did a post in defense of frequency where he says one of the benefits of publishing a lot is that get a lot more data and now you’re more likely to learn what’s working, what’s not. Imagine if CoSchedule or Orbit or anybody just published once a month, we would not have very much insights into what topics are working, what formats, which collaborators to work with, how to structure the content because we just wouldn’t have a whole lot of grease to the mill. I think that frequency gives us the benefit of more content, more data and more opportunities to optimize.Nathan: I love that insight. I think that frequency is just a really important part in consistency. That’s something that we advocate a lot at CoSchedule, especially because the tools that we offer helps solve some of that problems around consistency and getting organized to be able to publish frequently. It’s kind of fun to hear some research to back up that claim. Andy: Yeah. There again is part of the benefit of doing original research. This research for me is kind of a content play and it’s a little PR splash and hundreds of websites talk about it and link to it and it gets a lot of traction in social media. But in your case, you just brought up an excellent point. One of the real deeper benefits of doing original research is that you can publish a statistic that supports your sales claim, your value proposition. In this case, this survey just happens to support what you guys do so beautifully which is keep it organized, keep it consistent, be good at what you’re doing in a repeatable way overtime. What our listeners or anyone can do is to ask themselves, what statistic, what data point would strengthen my sales pitch? And then go find that data, go produce that, go do the study or do the survey or find that so when you go talk to people, you can bring that point up and back up all of your marketing claims with some type of evidence, in this case original research.Nathan: Yes. That’s what we wanted to talk about today. You described this tool to me, or this project to me as creating something that’s a content marketing play. I wanted to dive a little bit deeper into that and get your advice on helping other marketers conduct and publish their own original research. I know you did this research through an online survey, I was wondering if you could explain or describe some of the other really good ways to gather research for content marketing?Andy: When you do something like this, I’m getting 1,000 people to answer 12 questions and I’m doing a lots of outreach, we can talk about how we did the outreach. And then, you produce a statistic and that’s interesting. When you publish it, you can reach back out to those 1,000 people and you build a list. It’s based a little bit on this idea of quantity, a lot of people answered. But another really powerful way to do this is maybe you’re doing a survey, where in my case I might talk to 50 marketing directors or 50 chief marketing officers and I want to spend 10 minutes with each of them on the phone. In that case, it’s not as much about the quantity, it’s about the quality and I’m getting more of a networking benefit for my research which is perfectly legit, some people do it extremely well. It’s just a different play, it’s a different way to get benefits from research, it gives you a pretense to call people and talk to them and grow your network and interview and that’s a smaller group of more targeted people. There’s other ways to do it too. We’ve done research where it was like what percentage of websites have a certain feature? I just found a list of marketing websites, had an intern go look at each website, we publish just the stats to show this percentage of websites had these features; that got amazing results. We created a post out of that called Web Design Standards. If you search for Web Design Standards, you’ll find that piece of research. There’s lots of ways to do it, you can do it by observation, that last example, or by high touch outreach, which gives you a networking benefit, you can do it with a survey. You can conduct experiments, AB testing, you guys have published so many things on these topics, you guys had done lots of research yourselves. You can even do it just by aggregating other people’s research which is a several data points from several different places. Or research the research itself and you can produce new statistics that in some ways are more credible than each of the originals.Nathan: Just to dive a little bit deeper into how you did this research, you took the survey approach and you had mentioned before that you got results from more than a thousand people and I think that’s a really big number. I was wondering how did you share or promote that survey to get so many participants?Andy: This one is a brute force, basically. I’m not sure I recommend this for the listeners. Very few people will try to do this, although it is extremely effective. What I’ve done over the years is paid attention to people as I connect with them online. If they are a content creator, I think of them as a possible future respondent to the survey and I add them to this list. I used to do this in Google Sheets but now I’m using Ninja Outreach. I put them in there on this list of bloggers or content creators.When it came time to launch this year’s survey, I wrote a simple template email. But then Ninja Outreach is not email marketing because I’m not using it to mass mail. I just see the person on the left, oh, there’s Nathan, and then I see the template on the right, and I can actually customize the template a little bit based on who I’m talking to. It’s like, “Oh, hey, would you mind taking this survey, it would be super helpful. By the way, how’s life in Fargo? I love what you guys are doing in CoSchedule.” I did that a thousand times, basically. I did it for an hour in the morning, an hour at night, every day for more than a month before I got 1,000 responses. Other good things happen when you do mass outreach. It’s not the most efficient way but a lot of these people are close friends and it’s fun to reconnect.Nathan: I can say that that definitely worked on me because I got your email and I immediately went in and helped you out. Well done.Andy: Thank you. I’m very appreciative. It definitely took a big team to pull this one off.Nathan: I think a big part of your original research’s success would lie on some of the questions that you ask. Making sure that you get the answers that you want. I was wondering if you could kind of elaborate, how do you write a really good survey question?Andy: This survey was sort of the basics, the what, the when, the where, the why. There are experts who are really good at writing surveys. What they would tell you is you front load your survey with the most important questions so that you can get that data even if the person drops off. People are less likely to give you demographic data because it’s personal. Put those things at the end. Whatever you’re looking for, first you want to have your survey be short enough, you don’t ask questions you don’t plan to do analysis on or aren’t thinking of using in some way at least; lots of survey are just way too long. This was one page and the questions were pretty tight, how long does it take, I had timeframes, and how many words, and I had ranges of word length. The other questions were really about where are you, what time of day, things like day lead, more than day lead, two to six posts a week or weekly, several posts a month or monthly. The most important one thought at the end and the best question, we didn’t add this until this year, is that final question about do you report strong results, moderate result, weak results, strong, some disappointing or I don’t know. We asked that question about what kind of results do you get. That allowed us to do the correlation and determine people who gave this kind of answer were more or less likely to say they got this kind of results. All of the data became much more interesting because we got that final piece of insight that connects it to outcomes. Nathan:I bet writing those questions and organizing that survey, you have some personal hypothesis of what the results can be as you draft those questions. How do you avoid bias or generalizations when you’re analyzing those results to report on them? Andy:You’re always looking at the amount of the total amount of data. There are examples where I had like, wow, this looks like a very strong correlation, but there were only 15 people who gave that combination of answers. So, okay, I’m getting down towards single digits here. I can’t draw strong conclusions from that. Even with a thousand respondents, the number of people who answered this way to both these questions gets smaller and smaller.I had strong pre disposition toward thinking okay look it’s showing once again, the trend toward quality, the trend toward greater investment, but there were several times that I was actually surprised which helped me confirm that I didn’t have a strong bias, surprised to see that people who had higher frequency reported better results. Really in this case as the content play, we’re looking to create a sound bite almost. You’re trying to create a unique statistic that you can then portray visually and get lots of traction in your marketing because you have a piece of data that no one else had until now. There could easily be biases in all kinds of surveys impulse. The more subjective the question, the more likely that is. Fortunately for me in this case, I was like, do you use email marketing or not? Are you doing SEO or you’re checking your analytics? They’re pretty binary responses.Nathan: Yeah, almost keeping it black and white gets you better results and easier to analyze.Andy: Exactly.Nathan: Andy, you had mentioned that you published this stuff and you get it out there and that these results get a lot of attention for you guys. I know that you put a lot of work into making that work that you just put into it pay off. I’m wondering how do you maximize or promote that original research once it’s published?Andy: Here’s a great trick for anybody. If you’d start doing this and you keep doing it, you’ll find that it gets easier, more efficient, and more powerful. Obviously there’s a lot of tools, I don’t need to name them. The tools that will show you who shared a piece in the past. That’s your list of who to share the new version with. Put in the old one, see what influenced the people share the last one, make them aware of the new one. You could also put the old address, the old URL into a tool that shows you who linked to something. That’s your list of people to promote it to who might be interested in covering it again and linking to it again. Basically, persistence wins, it’s the most powerful force in the universe. If you just keep making a new version of your research and make it an annual thing, it gets easier and easier because the people that you used to gather the data, you list gets stronger, larger, easier to access, the people who shared it, the people who mentioned it, wrote about it, linked to it, contributed to it. All you had to do is to look at what worked before and try and repeat that by letting each of those things be a trigger for you to do again. That is something that I realized like a flash of lightning, like wow, I know exactly how to promote this thing now because I can see what worked before.Nathan: And I think that’s a really good play and that ties into something that I wanted to ask you. You conduct this original research and published it three times already in three years. Why do you do that research? Why do you publish this survey data regularly?Andy: Without that, the data only gets more valuable if you gather it again and again to get the trending overtime. This is a kind of thing, it’s a survey, it’s showing what people are doing, and it’s data that you can’t really get in the other way. I’ve seen you guys do studies where it’s like length of content and what gets what kind of traction, those are things that you can get in a quantitative way, just through tools. But this data about, for example, what percentage of bloggers are putting more than one image into their content. To see that number creep up year after year, okay now I’m showing a trend. It just becomes more insightful, more interesting to do it repeatedly because you get longitudinal which becomes another reportable piece of insight that we can talk about just like we did just now. The percentage of people, like you and I, who put a lot of effort into their content has jumped. I wouldn’t know that it jumped or changed overtime had I not done this more than once.Nathan: And I’m sure since you surveyed that niche so regularly, you’ve kind of conducted a similar survey year after year. I’m just curious to know what results have you experienced from publishing this research.Andy: We make money when people hire us to build their website and our target audience is Chicago. If you work backwards from the question how do I rank number one for Chicago web design, Chicago web developer, Chicago WordPress and every other phrase, you conclude quickly that you need to have a very high domain authority, much higher than these other people who are also pros at this. Then the next step is, how do I get others sites to link to me. Then the next step is how do I get something on my site that is eminently link worthy. The next step is what do people really link to? And pretty soon you conclude I need something that is totally original that is the primary source, that a link to it is just a friendly thing, it’s an actual citation. The last two versions of the survey, 2014 and 2015, were linked to a combined more than 350 times from different websites which is very high for just two individual URLs. If you understand search and authority and competition for rankings, then it’s a pretty short line between that and their conclusion that you need something amazing, some powerhouse piece of content that just attracts links and mentions all day long. The survey has helped us maintain a search engine ranking that drives 70,000 visits a month, between 5 and 10 leads per day, $5,000,000 per year in business and 38 total employees with on-going needs including bonuses and health care. It’s all part of it.Nathan: I think that’s a really interesting part of this too. You mentioned that yes you are looking at the local scene but the work that you’re doing is spreading nationally and likely global. I’m talking to you from North Dakota right now. I don’t know if you want to elaborate a little bit on that but to not limit yourself when you’re a local company.Andy: It’s the fact that we are focused on just Chicago for our business audience as a service provider means that we really don’t sell to people outside of Chicago very often but that doesn’t mean that we can win in Chicago without having a much larger presence. If you want to win in your geography for a huge value service related phrase, like Dallas, Texas, Personal Injury Law, whatever, you probably need to go big on your content, publishing something that is useful to a much, much larger audience, national or international audience, to attract enough subscribers, followers and links to keep your site enough credibility. That’s why we all have to publish this mini version of Wikipedia on our websites.Nathan: I like that analogy and that’s really good advice, thanks for sharing. I guess Andy I have one more question for you about this, I think that some of this original research would be a really big strategic advantage for you guys at Orbit. I was just wondering how do you use it to improve your own marketing?Andy: There are a lots of little ways in which the work flow for something like this is more sophisticated than your typical piece of content. It forces us to go through a series of editors I have to work with, our technical director to crunch the data because I’m not good enough at Exel. I have to work with the design team and plan ahead to create all the graphics for it. I have to do outreach much earlier to share the preliminary data with the influencers who are going to be contributing and providing quotes for the research. It forces you to not be that little just running gun kind of content shop. You’re thinking mugs out which is certainly a stretch for someone like me who normally doesn’t have to do that much forward thought for content. That’s a useful thing to get the marketing team more integrated with the developer, the designer, the other outside audience. It also becomes a little more outpoint because when we launch this, we may have our own company meetings and we point it out and everyone sees how we really did just contribute in a totally original way to our field. It’s a good feeling, it’s a good thing, right? It’s like we’re not just a me too place, we are adding something new to a very loud and noisy conversation that rises above by being actually significant. It’s not a small thing. It’s one of our little attempts to provide value in a much larger way.Nathan: Yeah. I was actually thinking that in my head while you were talking, provide value. I think that your original research has been really influential and beneficial for us here at CoSchedule. It’s fun to look at how people are doing content marketing right now.Andy, thanks a lot for sharing your thoughts on original research, what you guys have done with your annual blogging survey and all your advice on this topic. Thank you.Andy: My pleasure, Nathan. So great to collaborate with CoSchedule. I have been a fan for years.Nathan: You just learned a lot about publishing original research with Andy Crestodina, the co-founder and strategic director at Orbit Media Studios. Andy, you crushed it. Thanks a lot for sharing your insight on not only your survey data but helping all of us do our own original research too.