Bill Beutler on Why Wikipedia May Be the Most Overlooked (and Dangerous) Marketing Channel

Auto-generated transcript. Speaker names, spelling, and punctuation may be slightly off.

Mark Evans: When people think of AI, they talk about models, data, and often hallucinations. But they often forget one of the most important sources, Wikipedia. From ChatGPT Perplexity to Google's AI overviews, nearly every major AI tool relies on Wikipedia as a foundational knowledge base. But behind every page is a strange rule bound ecosystem of editors, guidelines, debates, and decisions, many of which happen without brands even knowing. And today's guest, Bill Bootler, is one of the few people who really understands how Wikipedia works. Bill runs an agency that helps Fortune 50 companies, universities, and even members of Congress manage their presence on the platform, all within Wikipedia's strict community guidelines. In this episode, we'll pull back, or attempt to pull back, the curtain on how Wikipedia works, the high stakes implications for brands, and how this free encyclopedia is now influencing the future of AI. Welcome to Marketing's BarkBelt.

Guest: Mark, thank you very much for having me here today.

Mark Evans: I think we should start, or I should start with a confession, is that as a brand and content marketer, Wikipedia is not in my wheelhouse on a regular basis.

Guest: Sure.

Mark Evans: It sounds like after this conversation, I'm gonna have to change my behavior. I'm gonna have to include it along with X and Reddit and LinkedIn into the marketing mix. And I would hazard a guess that my ignorance, my lack of maybe the fact that I don't pay too much attention to Wikipedia is probably common among the marketing community. And as we talk about the emergence of AI and how Wikipedia is one of those important but maybe low profile players right now, why do you think that Wikipedia isn't top of mind? If you think about it, a lot of marketers have now shifted their focus to Reddit. It's all about Reddit because that's where the true answers are. That's where you actually get real people giving real answers. But not so much Wikipedia. You don't see a lot of those conversations. Why is that happening either? And why should marketers pay attention to Wikipedia? And then we'll talk about Wikipedia and the whole AI emergence.

Guest: Absolutely. I think the simplest answer for why Wikipedia is not in the conversation, not in the mix as often as a site like Reddit, is a learned helplessness from many years of hearing stories about brands who tried to touch the third rail of Wikipedia and got themselves shocked. And if you've been in in the marketing or PR business long enough, you've heard some of the the scary stories, and there are many, like, public relations firms who just have a policy of we don't touch Wikipedia. It's too complicated. The Wikipedia community is they're not they are not looking to do marketing for anybody, and so it is difficult for a brand to find a foothold or a way to contribute to improve their own presence or to improve information that's relevant to their business as well. Different story on Reddit, different story on X, where X in particular, especially Twitter in its old days, and all the meta sites, they want marketers to come do marketing on their marketing platform. And Wikipedia is that's not what we are. We're an educational it's an NGO, really. It's a global organization that tries to pull together and organize world information and make it all available for free. It's like this altruistic project where corporate strategies are like oil and water in most cases, certainly not in all to the fact that my company has existed now for a decade and a half. But, yeah, no, it's Reddit's easier to use. You can create an account. You can go there and start posting, and it's just as lower stakes. Wikipedia, very circumscribed ways of contributing. And for most, it's just not worth the hassle.

Mark Evans: The third rail analogy is great because a lot of marketers I think they have an ignorance of what Wikipedia is it really part of the marketing mix? I would argue it's not part of the marketing mix, for better or worse. Reddit, as you say, is a lot more accessible, although there are lots of people who are very cautious when it comes to tipping dipping their toes in the red waters.

Guest: It can burn you too.

Mark Evans: It can burn you too. Couple of questions here. One would be, why should marketers embrace Wikipedia? And if you don't have Wikipedia strategy, obviously you're in the camp that you should have a Wikipedia strategy when it comes to marketing. What are some of the key steps to get started? You're starting from scratch. You don't want to make any mistakes. You want to leverage the platform. You want to get ROI from it, but you do not want to leap before you're ready. Number one question, why should we pay attention? Number two question is the key steps to get started properly.

Guest: So the first part's easy, and there are more reasons now than there were even six months ago to, or a couple years ago, to take Wikipedia, and it's, like, family of sites seriously. There's Wikipedia. There's Wikidata. There's Wikimedia Commons. The Wikimedia Foundation that oversees Wikipedia also oversees a few other projects. Wikipedia is definitely the most influential, but the others have some niche informational purposes as well that can be useful for brands. And so for many years, Google Search surfacing Wikipedia pages to the top of search results pages was the reason why people cared. Mhmm. It was a place where everybody's familiar with it. It was always presented in the same voice and the same it's familiar. It's easier to read a bunch of Wikipedia pages than to go to a bunch of different corporate and academic and news pages where they're all written differently. Wikipedia is maybe not perfect, but it is, like, easy to consume. And its ubiquity has granted it a kind of authority, so it's a place to be seen. Now in the last couple of years, but really in the last six months, as there the kind of marketing community is waking up to the notion that people are using ChatGPT and Perplexity to form opinions and make purchase decisions. So all of a sudden, it's important to understand how the LLMs are describing you and to figure out what you can do about that. So it happens to be the case that not only is Wikipedia very important to how Google organizes information,

Mark Evans: but

Guest: Wikipedia and its sister sites, but really Wikipedia the most, are important to how and foundational fundamental to how the LLM models are built in the first place. Wikipedia is one of the kind of is one of the pieces of training data that's in every that's in every training run. It's in every model. It's free. It's incredibly dense. It's so well structured and interlinked. You would be foolish not to use Wikipedia in training your model. Now to the question of what's the why you should have a strategy. I would say this. You should have a Wikipedia perspective. Not everybody needs a Wikipedia strategy because not every brand qualifies for Wikipedia. So you should at least first understand whether you qualify or not. If you do have an article, size up how good it is, see how it's meeting your kind of information goals. I would say that every brand should have a wiki strategy. Notice I left the pedia off there. That's because of those other the sibling sites of Wikipedia. Wikidata is a structured database of facts. It's Wikipedia is meant to be read by people. Wikidata is meant to be read by machines. And you can go look at it at wikidata.org and look at a look at an entry. You can get the idea of what it is, but it's not for people really to spend any time reading. Wikimedia Commons, the image image repository, the media repository of Wikipedia, uploading images to that if you can release them under a free license, those go, like, right to the top of Google search results and now sorry, the Google image search. And and show up in info box nav boxes. And then increasingly, the LLMs are going multimodal, and they're bringing in images and other things as well. So it's like you want to be findable where there are opportunities to place your information. Wikipedia doesn't want every brand. It does a bit of gatekeeping. Wikidata, WikiCommons in particular, are a lot. They throw open the doors. They want if your content can meet its standards, it wants it. And so that's the way I would say that you should have a wiki strategy even if you're not quite ready for Wikipedia.

Mark Evans: Couple of questions here. Given the versus AI and the fact that when you do a search on Google, you get the AI generated summary, and I would suggest for the most part is that a lot of people go there and they never go anywhere else. They don't scroll. They don't click. Impact has that had on Wikipedia in terms of traffic generation? Because as you mentioned, used to be one of the top results for many searches, but maybe now not so much. And I think, obviously, it's afflicting every single website out So what's your take on on the traffic impact? And then we can talk about another thing.

Guest: I can speak to the Wikipedia traffic because it's been been a couple weeks since I looked at the stats, but nothing's really changed in a couple weeks. Wikipedia traffic is steady. It is not declining. There's a study that someone did last year that suggested that at most, the LLMs, again, ChatGPT is the biggest by far, are slowing Wikipedia's growth more than they are it's not actually declining. One reason would be that I don't know. And no one knows all the reasons exactly for why traffic goes in which directions or not. I will say this, though, that because ChatGPT now does cite its sources when it goes and does the real time search to build an answer, And Wikipedia is so often in those results. People are not really in the habit of clicking on those the way that they used to be in the habit of clicking on Google links, but it is another avenue for people to find Wikipedia information. So it's it's not completely hiding the fact that it's getting information from Wikipedia. I also think that Wikipedia is still a destination site of its own. Even if you don't start out searching on Wikipedia, if if you're using iPhone like I do, and you start typing in the name of a movie or an actor, like, iOS is gonna recommend a Wikipedia page even before it recommends a Google page for certain proper nouns like that. So there are there's multiple ways to access it. Also, the way that many people go to Google, type in their search term, and then add Reddit to the end to get a real person's opinion. Yeah. I it is also not unheard of for people to add Wiki to the end to make sure they find the Wikipedia article because they know how that they know how to read that. They know what they're gonna get. And Wikipedia's value remains considerable at this point in time and for the foreseeable future. I do think what you alluded to, the traffic being taken away from news publications that depend on advertisements from eyeballs in order to create keep creating more news is potentially a long term problem for Wikipedia, though, because Wikipedia articles need to have sources, reliable sources, citations, citation needed being the the common phrase. Yep. You might have a lot more citation needed if there are not news articles out there to verify information. So there also could be fewer Wikipedia articles written if there are not enough news publications. And I would go as far as to say that there are, like, certain categories, particularly in business, entertainment as well, that the decline of publications serving those markets means that there are fewer reliable sources for Wikipedia to build articles out of. So it may have already impacted it, and we just don't know. It's the dog that done the dog that didn't bark is the Wikipedia page that didn't show up when you search for that topic.

Mark Evans: Right. One of the biggest focuses for a lot of marketers is generating content for LLMs or fixing their content, updating their content so that they can have a bigger presence. So that when people are using ChatGPT or Anthropic is their brand names will come up as one of the sources. A lot of time and effort is being spent, a lot of pieces of that. A lot of AI is being used to generate content.

Guest: The same For better or worse.

Mark Evans: Yeah. Is the same approach being applied to Wikipedia? Are marketers starting to maybe I should back up, should marketers start to audit their Wikipedia presence and then proactively look to either fix the content on Wikipedia or create new content so that their Wikipedia presence is improved, and then it gets Yes. Crawled by the LLMs and, like, in this sort of virtuous circle that they may wanna create.

Guest: Yeah. So especially for brands, companies, people that already have existing entries, which have long argued that they should care what it says. I think that is even more true now. I think in the past, there might have been a little bit of an attitude of, I don't trust that site. People don't trust that site. Whatever. I don't wanna be seen as being too vain as to come get involved there. But I think that increasingly in this LLM I don't know LLM first, but, like, in this current media environment, in this current information discovery environment, you are doing yourself a disservice if you don't pay attention to what it says because the machine doesn't know the difference. The LLMs are good at facts sorry. They're good at patterns. They're not so good at facts. And so they'll just regurgitate what they find. And so if you have a page and it is inaccurate or if it is not telling the full story, then those inaccuracies and those gaps will translate to whatever platform is making use of them, whether it's Google, whether it's ChatGPT. So it's just it's good information hygiene. I would also say, I mentioned Wikidata before, whereas Wikipedia doesn't want a brand. Then there's not one article about every brand in the universe. Wikidata, maybe not every brand, but every established, verifiable, going concern where you can show that there is that this topic could be linked to other topics and where you can show, like, citations on Wikipedia need to be, like, news articles. Citations on Wikidata can be press releases, which they can never be on Wikipedia, really. Wikidata is much more expansive in terms of what it wants to cover. It's also more welcoming in terms of what kind of sources that it allows. It still has a little bit of Wikipedia's kind of corporate skepticism, so there are there there are new invisible tripwires with Wikidata. It's also harder to use in the sense that as a structured database, you're not writing sentences. You're matching statements and properties and properties and values, the form statements, and it's a little complicated. It's a little more math y than Wikipedia is. But these are important information sources. And so at the very least, you should have a perspective on them even if you choose not to do something about them now. But I will say this. Since q one at my company, we have received we've seen, like, tripling in the inbound interest in creating articles and improving existing ones. So this is definitely a time where people are waking up to Wikipedia as being important. They've always known a little bit about it, but now it has a renewed focus, I would say.

Mark Evans: Great perspective. So if we go back in time when it came when SEO emerged as the force Yeah. On the web and people were keyword stuffing, manipulating content and searches. And Yeah. Now you've got marketers that are, for lack of a better phrase, LLM stuffy. They're write they're writing content. Some of them writing content at scale around particular words so that this that the LLMs can pick it up. When it comes to Wikipedia, because the rules of engagement and the community guidelines are so strict, what's like, how do you work with clients to make sure that they can improve their Wikipedia preference, but they're ethically on side

Guest: Yep.

Mark Evans: And they can engage with the Wikia Wikipedia community in a very credible, authentic way. Because the last thing you want is someone to call you out and go, brand x, you're clearly trying to you know, you're playing dirty pool here. How do you work with clients to make when they come to you, as they as it sounds like they are, and they say, no, listen, Wikipedia is not getting much attention right now. We know we need to be there. Help us do better. What do you do with them to do better? And maybe give me an example, if you can, of a client that you successfully helped do better on Wikipedia.

Guest: Absolutely. So there are two tracks for how we help clients, and the first one would be most people come to us asking to create new articles. And we can't create all of them, so even though that's most of the interest, most of the actual work is in improving existing entries. So we work with a lot of large companies, and there are, like, communications departments to improve the existing pages about them. So I'll focus on the latter just because that's a little more straightforward. And, yeah, I can definitely talk about a good example of a client we've worked with a number of times over the years. The first thing I want to do is find out what it is that our client is looking to do. What is their problem? What is their gap they're trying to fill? Just to size it up. What's the scope of it? And then we will compare that to what is feasible according to Wikipedia's content guidelines and the sources that are available to verify information. Gotta have high quality sources to verify, or Wikipedia doesn't wanna have that information. So then my team will with that altogether, we will write a new draft of maybe it's just a few sentences, maybe it's a full section, heck, maybe it's the entire article. Every topic is different because every company is different and the situation they're in is different and the sources about them say different things. Project we take on, we learn about it as we go along. And so to the question of the engagement, which is very important, very good question. Wikipedia has a set of rules around paid editing, conflict of interest being the key term they use. Basically says, if you are connected to a brand or a person and you especially if you have a financial connection, they kindly ask not to make direct changes to the page, instead to create an account and have this account be disclosed as being either an employee or contractor or a consultant for that for that brand. And then instead of going in and making direct changes to the live article, go to the discussion section that's attached. Every Wikipedia article has a talk page. Mark, have you been on Wikipedia talk pages before? No.

Mark Evans: I have not. No.

Guest: Okay. Most have not. Like, 1% of people ever go look at a talk page, but those are vitally important, and they are prominently linked at the top of every Wikipedia page. And right underneath the name, it says talk. There's a link to the talk page. And go in there, and you can see this is where Wikipedia editors talk things out, so to speak. They hash out issues. They argue about this should be on there, and this shouldn't be. This is where we are invited to come raise our issues. And we can say, hey there. I'm so and so. I'm representing brand ABC, and we have noticed that the new CEO's name is not there. It hasn't been updated. Here is the name of the CEO. Here is the link verifying the fact. Will someone make this change for us? And that's kind of the stylized version of it. In in in practice, there's more it can get fiddly in certain ways. But that's basically it. You out a gap and ask a volunteer to make the change. If it's a real issue, if you've made a good case, then they should make that change for you. What we do is to do that at a a a massive scale. Not a massive scale. A very large scale for working for dozens of clients at a time.

Mark Evans: So you talked about a client that that you worked with over the years. Yeah. That would be great just to get some sort of real world example of a client that has leveraged your services to improve their orthopedic presence.

Guest: What was their problem?

Mark Evans: How did you help them? What has been the results?

Guest: For sure. The one the first thing I'll mention, I guess, would be Mayo Clinic. Well known brand, hospital system here in The United States, where I'm in The US. So we've worked with them for multiple times over the years, mostly on their primary article, Mayo Clinic. Although being a large organization, there are multiple pages. In some sense, it is the same story every time. The existing article is missing key information. For Mayo Clinic in particular, it was their main article was missing information about research they had done, like, how they had advanced the kind of medical field over decades. It's like that story just wasn't being told. Nobody had got around to it. Wikipedia editors, they focus on things they're most passionate about, and there were just no no Mayo Clinic stands ready to pile in. And it's not like the the Taylor Swift article is probably one of the most, like, perfect, detailed, up to date Wikipedia articles on the planet. Right. And most brands can't hope to compete with that, or most artists can't either. So for Mayo Clinic, just as I said, we figured out what were they missing and what things were inaccurate. We prepared new content with them and shared it with their communications team. And sometimes we will work with a client and help them to own a user account and go to Wikipedia and have a conversation with editors. Sometimes we will do that for them. It's a little faster if, say, we can represent the company or organization using our own preexisting disclosed accounts that we use for multiple clients. And so I think that's what we did with Mayo Clinic was it was largely led by our team. And we followed all of the content rules in creating high quality informative content that gave good information to Wikipedia's readers about Mayo Clinic. And then we posted it was a series of edit requests over now we worked with them on, like, at least two or three occasions over the years. And so each time was a matter of a series of months. We never know how long something's gonna take on Wikipedia because it is a volunteer run platform. The editors just sometimes they're slow, That can be frustrating. There's actually a page in Wikipedia, like a guideline called, there is no deadline. And let me tell you, corporate PR managers don't love to hear that, but it's just a thing that we have to work with.

Mark Evans: So the question a lot of marketers, the question I get asked by my boss all the time is, how do you know that your efforts are being successful and how do you measure ROI? So someone comes to you, a client comes to you, Mayo Clinic, and says, listen, Bill, I want to spend x to improve my Wikipedia presence. Given the fact that everything moves in Wikipedia time as opposed to real time, what are the metrics that you can how do you justify or quantify your success? Yeah. Like, how does a client know that, yes, working with Bill is great, and I'm getting my money's worth?

Guest: So I I think the answer to this question is changing a bit right now, again, because we're in this LLM era. Like, I will say, for the longest time, we have positioned our Wikipedia services, you know, actually more as public relations than as marketing, where it's a little more defensive in that if who is looking up information about you? It could be customers. It could be investors. It could be journalists. It could be your own employees. It could be prospective employees. And so when they go looking for you, you wanna make sure that they're getting good information. So that's a little more PR than marketing. And so over in the PR world, the ROI is not so much the conversation because you're planning ahead for that future point where you got covered accurately in a news article because that journalist read Wikipedia. But now we are in this more marketing oriented LLM world, and there are a crop of new tools that are popping up to show what citations are influencing the LLM answers. There's a whole bunch of these companies popped up overnight,

Mark Evans: practically. Right.

Guest: And so they do show that Wikipedia is a big part of the part of that. ProFound is one of the one of the best funded, best known of these firms, I think, right now. And they put out a study in June showing that GPT is cited in, I think, 48% or so Wikipedia is cited in 48% of GPT answers that include citations. So it's just it's super relies upon it. So now it becomes more quantifiable. Whereas in the past, it was like having an article or not was a kind of a binary proposition. You do or you don't. And then if you had one, for most brands, it will show up in your top two, three results. That's the Google first world. In an LLM first world, it doesn't so much matter what exact position you are in a search results page so long as your information is helping to form the answer that the LLM is giving. So I think the answer part of your question would be tracking it's tracking Wikipedia's appearance in those if you have such a tool, Found or Xfunnel, then they can tell you. They can tell you how much it's influencing. Yeah.

Mark Evans: A lot of my biggest focus when when it comes to marketing, understanding how customers behave and how brands behave is triggers. And trying to get a handle on what makes someone or an organization do things. Because no one does things simply because, hey, this is a good idea to do things for a reason.

Guest: Yeah. For sure.

Mark Evans: A lot of the times, it has to do with a problem that they wanna solve or an opportunity that they wanna seize. Mhmm. So when it comes to Wikipedia and we'll go back to the beginning where the reality is a lot of marketers aren't paying attention to Wikipedia. They may not understand the influence that Wikipedia is having on AI and the LLMs right now. So what are the signals that a CEO or a marketing leader would get to suggest that, hey. Wikipedia matters. We're not doing very well on Wikipedia.

Guest: There's signs that there's red flags

Mark Evans: coming up. Like, whether it's not Asheboro or anecdotally, but somewhere, somehow, they wake up one morning and go, holy cow. We suck on Wikipedia. We better fix this. What are the signals? How do they know?

Guest: And may it may

Mark Evans: be profound. It may be ex funnel, as you say. What are some of the other sort of sources of trouble that they look at?

Guest: So the stylized story I like to tell, the anecdote that I like to imagine, and I've heard variations on this from clients over the years, is let's say the company CEO was at Thanksgiving dinner with their family, and their niece said to them, hey. Have you seen what Wikipedia says about your company? Is it true? Did that really happen that way? And the CEO, who themselves is not searching Google, searching Wikipedia, all of a sudden is motivated to go have a look themselves, and they're like, oh, that's wrong. What can we do about this? And then they turn to their marketing or their communications team and be like, are you aware of this? And usually, they'll be like, yeah. But it's just such a ball of wax. We don't know what to do about let's figure this out. We it's time to do it. So many times I hear from new inbound prospective clients, it's this thing that's been on the back of our mind for years. We just know it's there. There's always some inciting incident, and so Niece example is one potential like that. Also to say that very often, it does tend to be driven from the c suite. That somebody with the ability to drive an issue at the company, particularly the CEO, maybe the CMO, maybe the the communications chief, has decided now is the time to do it. And that could be driven by external events because Wikipedia is very much driven by what's in the news. If you need to have a good citation to add information and the news is a good citation, then what's happening in the news tends to get reflected on Wikipedia, especially in the former journalists like us know the phrase, if it bleeds, it leads. It's not the happy stories that tend to get the most ink. It tends to be when you screwed up or when something bad happened. And so Wikipedia also tends to reflect that tendency that it's easier to get on Wikipedia for bad reasons than from good reasons. Good reasons sometimes can be accused of being promotional. If a fact is too positive That's too promotional.

Mark Evans: Yeah. Yeah.

Guest: Oh, but if a fact is negative, important public information. We must include this recall. So these are the challenges that we deal with, and I think they're fun to deal with, but they are a pain for those we're working with.

Mark Evans: And so here's the self promotional part of the podcast. If people are looking to do better on Wikipedia and they check out what your company offers, how do you help them? What are the different ways that you can help them with Wikipedia strategically or tactically?

Guest: I would encourage someone to look us up and drop us an email and just give us a quick summary of what their issue is, and someone from my team will reach out and have a conversation with you about what we've seen it all before. So we'll have a perspective on what it's usually and what could the contingencies are. We just always have to have a conversation in the first place to see what's really feasible. I will say this as well. If someone's just looking to get some more information, I would really recommend, again, going to our website, beutlerinc.com, beutlerink.com. And we have on our site there what we call the Wiki resource library. It includes a very extensive FAQ, just answering every marketing and PR related Wikipedia question under the sun. And then it also includes a series of DIY guides on things that maybe don't rise to the level of a good project for my team. But a common thing that people wanna do is update the logo. From Wikipedia perspective, it is pretty simple. If you've never done it before, you don't know where to begin. And Wikipedia's, like, instructions on some of these things can be a little bit overwhelming, they and may not be written with a business audience in mind. So our Wiki resource library is written with a business audience in mind, focused on common issues that we see. And, yes, if you want to change the headshot of your CEO on their page, we have a simple set of instructions for that. If you can get the thing solved by just reading what we have put out there, we are very happy to have helped make Wikipedia a little better and solve a small problem for a marketer who just needed to do this one thing. If your needs are bigger, if you have a more challenging problem, that's when I'd say reach out. Let's have a conversation.

Mark Evans: Great. Thanks, Phil. If you enjoyed this conversation, subscribe by Apple Podcasts, Spotify, or wherever you eat your podcasts, and leave a review if you like what you heard. I'd love to hear from you if you're a CEO, entrepreneur, or marketing leader with a unique perspective or an interesting journey to share. You can connect with me on LinkedIn. Until next time. Thanks for tuning in.