August 24, 2017
API Discovery Meetup: Clarifai AI and Algolia’s Airplane Mode Search
In these talks from the API Discovery Meetup, learn how you can use Clarifai for convolutional neural net training, and let users search eve...
All right, good evening. Thank you for hosting me. My name is Tyler Jewell, and I'd like to talk to you a little bit about the science of content marketing.
First, a little bit about myself, I am a software guy. I've been in software for a couple of decades now. I started off my career at BEA. I spent seven years at BEA, and for three of those years I was their chief evangelist. That meant I ran their developer marketing programs. I was responsible for the technology blogs, the conferences that we held around the world, and through that period developed a number of books on Java technology, also contributed to conferences and articles, and really became fascinated with the ideas around content marketing at the time.
After that period, I became the executive in charge of a company called The Middleware Company. They had a web property called TheServerSide.com, which at the time had a half million readers. We eventually sold that to TechTarget, which is a large media company that has a variety of properties, and they continue to operate TheServerSide today.
After that, I worked with Floyd, who was the originator and publisher of TheServerSide and launched InfoQ. InfoQ is about nine years old at this point and has a million monthly visitors, and that was a business that we started on $150,000 of capital and was able to develop a very strong readership through principles related to content marketing. It's an item that is very near and dear to me.
Then, beyond that, today I'm the founder and CEO of a company here at Heavybit called Codenvy. We're effectively still a stealth-based company. We've been working, about 50 of us, on a bunch of technology for a number of years now that we're about to go out to market with, and our fundamental marketing strategy is yet again going to be content marketing for us. This is something that is very near and dear to my heart, and I was pleased to come here and talk to you all about it when they invited me to do so.
Let's talk about what content is, where content fits into a corporate entity, and then the definition of content marketing. Then we can talk about the different types of attributes that make for good content marketing and not so good content marketing, and a process around which you could implement a function if you wanted to pursue this.
At the most generic level, content is any piece of valuable and relevant information for an audience.
Valuable here is something that is reusable, it needs to be informative, it can be educational, be quotable. Value means different things to different people.
Relevant is usually topical, current. It has to be related to some other trend that's going on. The audience in a business construct is who's the buyer, the influencer, the user, your investor, your competitor, your partner, but in some audience in mind. That's a very generic definition of content that's there.
Interesting, this is a study here that was done about a year and a half ago by the Content Marketing Institute, which is this organization that's based in Europe. It's a collection of content marketing professionals. They just report back across all your departments and functions, what's the content types by use of business.
Obviously, what this shows is that there is a rainbow, high variety of different types of content that companies produce. We oftentimes think of videos or blog or audio as a very basic format, but there's a wide variety, and some are popular and some are not so popular. There's not a concentration, there's a lot of variety, and this is really interesting .
You have content, you know that you need to do something valuable and relevant to it. You know that there's all different kinds of types of content.
Where does content fit in a commercial entity?
There are really three functions that are primary content producers. There's product marketing, there's the communications department, and there's the content marketing function.
The value proposition in the types of content that each of these departments create is very different, and it's driven by the justification, the rationale for existence.
The first is you've built a product. Some company, you've built a product. That product has documentation. That product has a user interface. That is a form of content, but since you've built the product, it's tied to that product and the product is not complete until you have completed that. You've taken the product and you've shipped it.
Now you've got a product marketing function, and product marketing is justified as the cost of doing business, because if you don't do product marketing, people are not going to become aware of your product. They're not going to be aware of the functionality that it provides. They're not going to be aware of how they can engage with you to learn more about the product.
Product marketing is something that you're obligated to do. It's the cost of doing business. Its primary purpose is to provide resources for sales people, prospects, customers, and to help move them through the funnel and be a better conversion rate.
The types of content that you produce here are the things that you expect and that you see any time you go to a company. It's the demonstration. It's the white papers. You might have a data sheet or a spec sheet. You're going to have your sales presentation. You might have a product infographic that explains it, and a whole set of case studies that verifies and validates your value proposition that's there.
This is content that's part of doing business. You have to refresh this content when you do a new version of your product or you introduce a new product, or you sunset the product from your business. That's the first thing. That is not content marketing.
The second thing is communications, AR or PR. When should I write a press release? When should I do an analyst report? What's a boiler plate? Why do I have to do an investor or a customer annual update?
These are all things that companies do when they feel like they need to redefine the boundaries of the competitive market. It's when they need to change the definition of the market itself and set the battle lines that are there, right?
I think of communications as air cover. It's the air force of the business. You produce communications content and you drive communications out to analysts and press when you're looking to define the space and change the competitive boundaries so that it's more favorable so that your sales team is conditioned to be more successful with the deals that they go into.
The justification here is either an increase in the addressable market, making the addressable market a little bit more tangible so that you can get to it, and, as a result, diminishing the impact that your competitors can have in the market space.
It's hard to justify, if you're a startup, communications, AR and PR. It's an expensive cost. You almost always have to do it through brokers and proxies. You have to produce special purpose content that is designed to help you achieve this goal. The results are always a little bit fluffy because how do you really know whether your addressable market is changing or not? Content marketing is not communications and AR/PR.
What content marketing is, is that it's a discipline, a very particular discipline designed to attract, retain and nurture a very targeted audience that's there. The purpose for doing this is to increase your lead or lower the cost of acquisition and reduce your churn rate over time. It is a tactic that you employ when your purpose is to drive revenues, support a sales team, and increase your margins overall.
The types of content that you can produce are very similar to the types of content you might do for product marketing, but they are content that's designed to go out to a distribution channel that is going to be valuable to that audience. It's everything from blogs, articles, videos, e-books, infographics, surveys. It's whatever is valuable to that audience. That's content marketing: it's a discipline.
You may ask about where's docs? We have to produce a lot of technical documentation. That is part of shipping a product. The docs people may be content specialists. They may produce a lot of content, but it's not content marketing. It's just part of the cost of shipping a product there, and it's part of the feature set of that product.
What is content marketing?
Content marketing is a discipline to create and distribute content to attract, acquire and retain a very targeted audience, with the objective of driving a profitable customer or a profitable action with a customer.
If you cannot define how you're going to drive a profitable action, then chances are you're not doing content marketing.
There's really two types of actions that we like to see that drives profitability. Either something that lowers your cost of acquisition and brought awareness to increase the number of leads, and anything that can improve your conversion rate will help here, or something that increases the amount of revenue that you get per customer.
If you can provide training, or a lot of education that is very valuable to an audience, they're more likely to acquire more, or they're more likely to renew than they would have without that content that's there. It's content that's valuable to a particular audience.
Now, what I said here is that it needs to be profitable and it needs to be to a particular audience. That audience doesn't necessarily have to be your prospects.
The purpose of content marketing is to define an audience that's important to your organization, and go after that audience in its entirety, even if all of those people may not eventually buy.
If that audience is important to your organization and you can develop a content strategy that's meaningful for that audience, then they will want to hang around you more often, and then they're more likely to convert. It increases the number of leads. It might improve their conversion rate. It might also improve the fact that they might stick with you longer then.
Another great study, why have companies typically done content marketing? What this basically says is that in the red, 2012, the black is 2013, it used to be that people did content marketing purely for awareness, as a communication strategy or as a way to engage customers, but the market has swung dramatically and almost half of all people who perform a function, a content marketing function, now do it purely for the purpose of lead generation on that. It's very much of a sales-driven discipline.
You're saying, "Okay, I'm kind of interested. "I'd like to drive some more leads. "I'd like to lower my cost of acquisition. "This value proposition, this sounds good. "Tell me more. "How do I get into the content marketing game? "Is it my job to produce one piece of content, "or do I need to produce multiple pieces of content? "How do I get into that?"
Obviously, if you want to have the highest possible impact with your content marketing function, you want to get the maximum consumption of that content, and there's two ways to do it. Either one is you can generate one piece of content and it goes viral, and you get an infinite amount of consumption of that one piece of content, which is really high leverage. Or you could develop a library of layered content over time.
In the first case, in order for that to take place, the content piece has to be emotional, timely. It's got to be brilliant, funny, maybe entertaining. It's got to be injected into the dialogue right at the right point. In order for that piece of content to be just right, there's a lot of moons that have to line up. In other words, that is completely unpredictable.
Generating one piece of content that gets a huge amount of virality and a huge amount of consumption is a very difficult thing to do. It requires perfect timing that's there. It's a risky endeavor. You can pursue it, you can put your effort into it, but you don't know what the results are going to be.
The other approach is to say, "Okay, instead of doing one piece of content, "what would the strategy be if I had "a very systematic approach to building a layered set of content over time?" Instead of trying to have a maximum impact on any one piece, what I try to do is have a little bit of impact on every piece there and I try to produce those pieces in a regular rhythm and I predefine that rhythm. If I do that, then I can be very predictable in what my content does and the results that I get for that. There's a quality versus quantity tradeoff, but there's also a value, a metric benefit to this as well.
The metric benefit is that if you produce one piece of content, its value to you is a unit of one. If you produce a second piece of content, it also has a unit value of one, but if you publish them one after the other, their combined value is actually 2.1.
The reason that the second piece of content increases the overall value is that, one, you get a cross-linking benefit, and when you start having a cross-linking benefit, your value on the search algorithm start to go higher. If you put that third piece of content in there, now the aggregate value is 3.2, and then 4.3, and on and on and on.
What happens is that it's not that the hardest piece of content or the least valuable piece of content that you ever produce is the very first one, and then every piece of content that you produce after that actually increases the value of all previous pieces of content because of the cross-linking value, and then also your ability to take that content and start to interweave it into a variety of distribution channels that are permanent and sticky.
They become permanent records of the Internet, and then starts to attract people that you would have never imagined, who start coming in through one of those pieces of content that was sticky, and then they find their way through the rest of them and get an immense value from the library that's there. It's been proven, there's been a lot of examples of this, we can talk about that, that if you just have a very rhythmic, consistent approach of content that's thematically relevant, you end up creating a lot of very persistent and sticky value over time.
Okay, you wanted to do that. How often would I have to produce this content? This chart says that the more you do it, the more valuable it is, but there's actually been studies that have shown that most professionals are looking at one to three times a week.
If you do a meaningful piece of content one to three times a week, that's going to be the highest conversion rate.
The next thing is that when you do the content, it needs to be thematically relevant. If you are doing one piece of content on Tuesday and then the next piece of content on Thursday are somewhat topically irrelevant, you can't get the cross-linking benefit. You're better off taking a theme, mapping that theme out over a number of months, have overlapping themes, and then slowly build upon it.
That ends up establishing a brand, it ends up establishing almost like an encyclopedic library of whatever topic you're building, and it starts to build the readership that's there.
The next thing is, is that, "Okay, you got me. "I need to be rhythmic, I need to be consistent, "I need to be predictable. "It needs to be thematic, it needs to be chunks." If you're going to do all that and you want to do it well, then you absolutely have to have a methodology around how you're going to produce the content. The hardest part of content marketing is not actually writing the content.
Most of you in here have so many ideas that you can actually get the content produced in 15 minutes in some sort of caffeinated rage of keyboard typing onto the screen. I do it all the time every time one of my employees pisses me off. I'll have a three-page paper done in a matter of five minutes, right? I'm sure every one of you have done that at some point in time. Writing the content is easy.
It's the production cycle of the content that is the hard part of this. Just look at what these guys do once or twice a week for the speaker series. They do the headshots. For me there must have been like 40,000 meetings with Dana in anticipation, to prepare the content for this. They've got two camera guys. This guy's just taken my picture the whole time through over here. You've got the audio going, the two presentation forms. They made sure I have my water here.
That's way more effort than I had to put into banging out these slides here, and the resulting impact is, is that they wanted to create something that was valuable to the people who were going to watch it, and that in order for it to be valuable, it had to be highly produced. If you're going to do one to three times a week, you're going to need to have a production cycle.
You're going to have to treat this as a discipline that the publishing of the content itself is way more effort than the content.
That's going to be important. I'll talk about how you can go about doing that. After you've decided that you're going to do the production cycle, that you've done the content marketing strategy, you then need to figure out what types of content you want to do.
There is content that is really difficult to develop, and then there's content that's really easy to develop. Then there's an impact rate. As you would imagine, what this chart shows here is that industry news, blogs, and some e-books, like mini e-books, and graphics are the things that are the easiest to produce that have the highest impact.There's all sorts of other things that you can do, but those are great areas to start.
Then you get into the content cycle, and there are three audiences that play. If you want to develop this as a discipline, there's really three people that you've got to set up here. The first is the editor, editor in chief, call him whatever you want. This person ideally would be a domain expert to whatever the audience is, but they don't have to be.
A lot of the great blogs that are out there, the editor in chief actually doesn't really relate to the audience, but what the editor in chief does, they run the editorial committee. They make sure that there is a calendar of content always ready to go. They recruit the authors. They make sure that the author does their mad 10 minutes of content creation. Then they coordinate the production cycle, all the editorial reviews, all the graphics, the web design, and so they need to coordinate this web of people who they are working to make sure that the backlog is always filled and always ready to go. That's a non-zero effort.
That person is also the person who is responsible for measuring the impact of the content marketing, and their mission must be making sure that each piece of content that is produced gets maximum distribution out there, and so they're quickly developing the blog, they're developing the distribution channels, building the relationships with other syndicators so that each new piece of content that goes out reaches just a little bit further each time.
Then they track and optimize and report back to management about how they've made you so much more money so that they can get a raise and continue what they're doing here.
Authors can be any technical experts, any experts in the company or out of the company. Content can come from anywhere. It doesn't need to be authored by you. It just has to be relevant to the audience you want to go after. Then you want to put together a professional face, so I highly recommend outsourcing the graphics and asset creation if you can't do it in-house.
You do all that work, you produce the content, then you need to build a distribution channel. Distribution channel is equally hard. It's not as simple as, "Hey, put it up on my blog. "Let's put out a couple of tweets "and get my friends to retweet it." That's very inauthentic. Tweeting is a very inauthentic way, unless it leads to a back and forth dialogue. What you're looking for is you want to have persistent and permanent placement of your content in the places that the audience you're looking to target will find valuable.
If you can do a Quora post, that's a Q&A about the piece of blog content, that's a permanent sticky place, a Reddit conversation, a permanent sticky place. If it's done right, it feels like a very authentic, very natural type of conversation which makes the audience feel connected to you, like they're beginning to develop a relationship to you and they're going to be drawn in to it.
The content marketer, the editor in chief, is responsible for figuring out where these content elements lie and identify these channels and work them. Work them, work them, work them.
There is a structured way in which you can build a channel plan, have this in the chart, but you identify the channel, the name, the structure, the tone in which you want, and then the desired call to action that you have.
You produce one piece of content, you have a list of 15 channels, whatever you are, and the editor makes sure that it gets driven through each of those channels each time.
Some success stories here. I told you about InfoQ. I'm just blown away by Floyd and his organization. They started the entire company, a for-profit company, on $150,000. They knew that they wanted to go after senior decision makers in the technology organization, preferably architects, team leads, some managerial types, but these were people who were making day to day technology choices within their organization.
They did not and have not ever hired a single content author. Every piece of content that has ever been produced on InfoQ has been done so by people who are not employees of the company. They developed a very concise audience. They developed a great culture and process around the benefit of people contributing content for them, and then they managed a very rigorous editorial calendar and created a revenue sharing model where they calculate at the end of the quarter all the revenues they get from their media impressions, from their sponsors, the events that they do, and all the authors get to participate in that.
That made each of the authors so invested in the success of the site itself that they were the evangelists to make sure that the content they wrote was, one, maximum distribution, and, two, highly effective. Really great model, and now they have a monthly unique visitors there.
Rebel Labs is a little bit more relevant. This is the company ZeroTurnaround. It started in Estonia. They are in the Java market. They make a very elegant utility that makes editing Java files a lot easier, so you don't have to recompile and redeploy, and it takes that edit cycle from two minutes down to nanoseconds here. A point product tool, high volume sell, and they basically said, "We're going to go after JVM developers."
They started their blog in the beginning of 2013. 18 months later, that blog accounts for I think 60% of the overall traffic for all their properties combined there. The statistic of 80% of the traffic is organic, this means that 80% of the traffic is not coming from people who went to ZeroTurnaround and then clicked over. These are people who registered to the RSS feed. These are people who typed in "Rebel Labs" directly. These are people who found the content through other means that were out there.
That's a reflection that their distribution channel is actually way more effective than the product marketers, who don't seem to be very effective at product marketing their company. Otherwise, you would see a better balance of that.
A third example is Cloudant. They started, and got bought by IBM earlier this year. Their idea was that they were trying to do lead generation for new projects. They were selling into departments. A lot of their deals were $100,000 to $200,000 recurring a year.
They wanted to go after the core CouchDB community. They went and started developing a lot of stories about how CouchDB was being used in the community, and it had nothing to do with their product but they just wanted to be the story masterminds, very stretched cases at high volume, and so they curated those, and they would do about one every two weeks or so.
In their case, they thought it was a mediocre success. It was a mediocre success because they were putting in about 100 hours per case study. They're investing way too much time for the results that it got. In order to do that, it required really talented individuals to develop that, so the skill set was way too high for them, and they could never get the rest of the company to help them develop the distribution channel.
While they had a huge impact and they were happy with some of the results, they felt like they could have come up with a better model on that. That's kind of more of a mediocre success sort of thing.
What can you do to get started? Very clear recipe here: set your objectives, clearly define some measurable goals that you want to target. Is it RSS feeds? Is it number of page visits? Is it number of content pieces? Very simple things, right? What do you want to achieve as a company with content marketing?
Define your audience. Audience, audience, audience, so important in this. Then do a psychometric analysis on that audience. What is it that they want? What do they need? What do they read, watch and attend? Figure out where they're at before you do this, then create your content plan, do a content calendar. Try to do a long content calendar, if you can. Then define the tactics of how you're going to get it drafted and published, and then test and analyze.
Generally, to do everything as I've described it here, you've got to have to have to budget $100,000 to $150,000 a year to get it going at full speed and capacity. Doing it half-baked is not going to have a lot of great results. Doing a blog post here, doing a blog post there. My advice on a lot of these things is this is a discipline, it is an investment. It needs to be part of your defined strategy. If you don't see how this strategy is going to help you drive revenues and leads, then it's just not a good use of your investment and your time.
You can definitely explore it, but think about it and make sure that before you dive into this world, that you recognize the role that it plays and that you're committed for a good six to nine months to see it through on that. If you're not committed to see it through to that degree, then it may not be the sort of system that you want to pursue at this point in time in your company's growth and you might want to wait till a little bit longer, when you can do this to its totality there.
Thank you. I'm residing here at Heavybit, here most days, when I'm not running around like a chicken, dealing with crisis number 10 of the day. You can reach me at email@example.com. I hang out on Twitter just a little bit as well. Other than that, thank you, and I hope you got something useful out of this.
He asked the question of the $150,000, what's that split into on an annual basis? It's a combination of syndicated content. You might have to pay authors, dedicated authors or content specialists to do that.
If you're going to commit to, say, 70 pieces of content in the first year, you're going to have to budget about $50,000 to $80,000 on the production cycle for that, graphics, blog design. If you have to do some translations or nice treatments, there's money that goes into that. It's enough money to cover a full time editor in chief as well, who's the dedicated person to just own this 24/7.
If you think there's a way to do it with on staff personnel, that all reduces the cash out of hand.
Yeah, that's awesome. The first question was how do you target the headlines, link bait versus some sort of other strategy that's there? The second, what's the relative percentage of doing something direct with your blog versus going out to some other distribution channel there?
On the first one, I come back and say, if you're going to do content marketing, it's got to be thematically-driven. The theme can come from topics that are hot with your competitors, topics that are hot with the industry, or topics that you're driving that you think are relevant to yourself. As long as that topic is meaningful to the audience, then it's going to be good.
Oftentimes, the audience that you're targeting is going to be behaving rationally about that, so if there's something that you would call link bait that's being discussed often, providing value on top of that conversation is probably going to be a very good thing, as long as it's thematically related there.
In terms of how much should you do on your blog versus say another, like a contributed post on another blog, I kind of define this as saying what are your distribution channels against your content calendar that's there? If you have a theme that you're going out with, you're going to get that rhythm of that.
Ideally, when you build out your distribution channel, when you publish that piece of content, it goes out to all of your channels at the same time, but some channels are special purpose channels, like partners that are there.
What you can do is you can either say, "All right, I have a partner co-marketing event. "They've asked me to do a contributed blog article for them. "That's going to fill up my slot, "my number two slot for the week," and you just go with that. You have to make the decision on how often you want to do that or not.
The beautiful thing is that if you do the contributed content smartly, again, you're layering the cross-links and the SEO back, so that's just bringing that readership back to the core library that you've developed there. I think that it's just part of that calendar, and it's more important to have a rhythm there, then to have a fair mix of the two at the end of the day.
I think that that's a good use of your time. Once you've picked up on the theme, the content author gets his head wrapped around it, so the effort to publish a derivative of the same theme becomes really low.
I think the more important thing is when you do pick a theme, stick with it for three to six months, as long as the theme is catching the audience and the attention of the audience that you're targeting that's there, and really think through when you're defining your calendar, where are all the companies and all the places that my audience resides that are going to get value out of it?
The editor in chief should be knocking on the doors of all those places and say, "Hey, would you like to link back to what we've got, "or would you like us to do a contributed piece "of content on this thematic topic as well?" It starts to play off, and it's almost like an accelerating momentum after a while. The first piece of content is always the hardest on any new theme, and then it gets easier and easier and easier.
Any benefit or loss from placing a whole library of themed content at once or over time? In the Zero -- I couldn't tell you, frankly. Let me take the ZeroTurnaround example.
When you look at their content calendar, they commit themselves to two times a week half page blogs. One time a month, full page kind of analysis article of some trend, and once a quarter detailed report. The detailed report is 60 pages long. In the detailed report, they go out and interview through surveys somewhere between like 1,000 to 2,000 developers, and then they get professionals to take all that data and do an analyst level breakdown of it.
When they started a year and a half ago, they took those reports, which were about 16 chunks long, and they would piecemeal them out. They got a negative backlash from the community because the thought was not connected all the way through.
They eventually switched to the point of just saying, "Let's just publish the entire report, all 60 pages." It's free, there's no gating factor. It's just published on one big PDF document, and you can go to town on it, and then their readership spiked because the thought followed through even though it was a compendium of stuff. I think you have to really assess your audience and how they're going to react to it.
Her statement was you have this headcount for the rhythmic content that's there. What do you do about thought leadership pieces which are more timely? I personally think of thought leadership as either a product marketing or a communications responsibility, and so it's budgeted through that.
If you have an AR/PR person, then the management team needs to work with them to get that thought leadership piece drafted, and get it into the channel where it's going to benefit.
That thought leadership rarely benefits a content audience, so if I'm at Codenvy, and for me, Codenvy, one of my buyer audiences is DevOps, there's not many thought leadership pieces I can write about that a DevOps guy cares about. The DevOps guy is a practitioner. He wants practitioner information, so that should not go in my content marketing blog. That thought leadership is for investors, it's for competitors, it's for partners, it's for the press, so I want to channel that through a completely different function and get that out there.
It's not my number, although I can tell you that it feels about right from my previous experiences. I can give you the data from the people who surveyed this, who basically these are professionals who do this way more than I do and they say, "Look, it's this."
Effectively, the thought that happens is that if you are posting five times a week or multiple times a day, two things. One, you just get fatigue with your audience. No one person wants to hear from anybody else that often, no matter how charming they are, not even George Takei. I really don't want to hear from him more than once a day. There's a fatigue element that is just human nature on that.
The second thing is that it's probably very difficult for one entity to produce more than a couple of pieces of really interesting content in a week anyways, so the higher the frequency that you get, chances are the potency of what you're having to say is going down, and that dilutes the rest of everything that you've got.
When you get into this, once every other week is too infrequent, because people are going to forget about you, and so you end up somewhere in the middle, which is a couple of times a week, where you're still current in their head but you're not fatiguing them at the end of the day.
That wasn't the data. I understand that the data said otherwise, but the pros, the people who do this for a living, will all -- I've put a link down there, I had the -- The pros will all say, "Don't do that. "Only do a couple of times a week."
His question was InfoQ is really great because they do the rhythmic content and they do the periodic events. The way we rationalize that in that business is that it's all one community, and there's two types of ways that that community can interact with each other.
One is virtually, through digital marketing techniques. The other way is in person, and that you actually are not going to get the maximum reading of the content that you have unless you orchestrate a way for these people to come together and communicate in person, because when people communicate together, they build an affinity for each other, they build a relationship to each other, and then, as a result, they get more value out of the content that's there.That's what leads you to why you need to do the events.
The frequency of the events has a very simple algorithm. Events are incredibly profitable. I mean, it's like 20X more profitable than all the other stuff that a media company does.
The thought of what you say, "Well, why don't you do "an event like that every week if it's that profitable?" Because the speaker availability just isn't there. When you look at the types of speakers that you have to get to target your audience, we're all targeting some niche audience there. There's only going to be a handful of speakers in the total universe that can speak, and they can only participate a few times a year, and that dictates the rhythm of the events that you can do on it.