 Well, hello, everyone. Welcome. Thank you so much for joining us today. My name is Mickey. I'm an analyst and project manager here at Parsons TKO. And for those of you who are just learning about us are joining us for the first time here at Parsons TKO we work with nonprofits to improve their marketing tech business process and integrated data strategies. So today we'll be talking about social media listening what it is in the potential applications for the mission driven sector. So to start, let's start off by defining what social media listening is. So there are two terms that people often get conflated one is social media listening and the other is social media monitoring. So the easiest way to think about these is that social media monitoring tells you what, while social media listening tells you why. Another way to put this is that social media monitoring is looking at your own organization social media accounts and that's tracking things like likes, follows and mentions for your own or your organization specific account on any social media platform. And then we have social media listening. And this is listening to what a broader audience is saying. And so this is what people can be saying about a specific keyword or hashtag and there's different ways to measure this whether it's in a geographic location, or anything like that. So social media listening can be applied in many different ways and one is sentiment analysis. And so just to start off with some definitions that we can get all acquainted with what this is. Social media sentiment analysis is a computer algorithm that is used to determine if a writer's attitude towards a specific topic is positive, negative or neutral. And this kind of thing called sentiment analysis is created using another computer algorithm called natural language processing, which is designed to read and understand human text. And so while these two things can seem very complicated and very computer science see these two things play a large role in social media listening and they can be used to enhance the insights that one can glean from social media listening. So from here we have a handy little chart and so you can see that all the elements that we talked about on the previous slide like emojis hashtags and texts that are gathered from social media listening, all go into this NLP algorithm process to give a sentiment analysis results. And the reason that sentiment analysis can be so helpful, especially when used in conjunction with social media listening is that it can help us to understand the data to create actual action steps for organization, whether that's improving a marketing or fundraising strategy. So now that we got all the definitions out of the way, I'll give a concrete example and I want to show you something that I've worked on personally. And so in addition to my amazing work here at PT KO I'm also the co founder of activism always a startup that works to provide social media listening insights to the mission driven sector. So I'm going to share with you one of our first projects that we did a little over a year ago, called black lives matter always. And with this project we conducted a sentiment analysis of the black lives matter movement on Twitter immediately following the murder of George of George Floyd in May of 2020. Taking the usage of specific keywords and hashtags over a period of time, we were able to form a deeper understanding of how the world was reacting to this specific violent event. Okay, so lots of things on this slide, but to start with social media listening we were able to take this data and create a moving graph that showed the change of BLM always the movement on Twitter over time. So if you look at this moving chart on the top x axis, you can see the number of times each hashtag was used, and how that changed based on how the dates progressed, which is on that bottom x axis. And so by taking all this data into account we were able to distill trends over a period of time by tracking their rise and fall. So what we specifically found was that in the beginning so right in May of 2020 the most popular hashtags were hashtag George Floyd and hashtag I can't breathe immediately following George Floyd's murder. And then as the movement tended to progress hashtags like hashtag Juneteenth rose and popularity, but they also fell again very quickly since they were more focused around seasonal events. And this and then lastly, by the end of June so about a month later after we started this analysis hashtag George Floyd was no longer trending. So even though in the beginning it was by far the most popular hashtag within about about 30 days it quickly went back to the bottom. It's very significant because it helps us get a broader view of what's happening on social media. If we had just looked at a trending hashtag for one day, we would have thought that hashtag George Floyd was the most popular hashtag on Twitter representing the black lives matter movement, but that just wasn't true. So by kind of just looking at an individual, we wouldn't be able to understand kind of a mass sentiment on social media platforms such as Twitter around a certain topic which in this case was the black lives matter movement. What I remember is that because social media moves so quickly it's extremely important to understand not just a moment in time but the bigger picture of the topic that you're looking at. And of course social media listening can be really helpful for that, because it's specifically this holistic view that helps you understand what your audience is interested in. And as a result can help you better connect with them on social media. And so this is where the importance particularly about cultural context comes in. So this graph is one from our BLM always project and this is a sentiment analysis graph which we defined briefly earlier on. And so looking at this graph alone it's very natural to deduce that the majority of the tweets that we saw on Twitter about the black lives matter movement were negative since we see so many since a large part of that graph has those black lines. So what this chart doesn't tell you just looking at it at first glance is that the sentiments don't actually refer to attitudes towards a specific topic, but actually refers to the language being used. So the majority of Twitter users weren't against BLM or in support of ex officer Derek Chauvin, but they actually were using language at the computer algorithm the NLP algorithm, mark as a negative. So just as we saw with the hashtag bar chart on the previous slide looking at one specific instant isn't enough when it comes to social media listening. It's far more useful and also frankly necessary to look at a bigger picture and look at it holistically to understand it and then utilize it correctly. So we have some example tweets of the types of tweets that we use in this analysis. So if you look at the details you can see that it's not necessarily that the majority of the Twitter sphere was anti BLM. Instead they were using rhetoric that the computer happened to define as negative. And the kind of things that the computer defined as negative were synonyms of pessimism, pessimism, aggressive hurtful or crude. So this is an example of a tweet taken action from our actual analysis. The negative tweet uses words and phrases, such as brazenly and senselessly murdered. And those are the kind of words that the computer algorithm deemed as negative, which is true, they're negative rhetoric, but that does not mean that this Twitter user was against the BLM always movement. It's not as easy as it was for the algorithm to say that the majority of tweets were negative. And for us to understand that as anti BLM is just as easy for other all cultural context to get lost in the algorithms, and therefore not necessarily translate to the strategies that we form for our organizations as a result of these insights. Another example is at the top for the positive tweet. You can see that the strategy works that uses words such as synonyms of optimism peaceful and compassionate has words like happy and blessed and love, and in this particular context this algorithm does seem to be pro BLM. But of course it's always important to consider the wide of context of how these words are being used in this rhetoric, and what the larger social media audience thinks about this. So we've talked about what social media listening is what it can be used for the pros and cons. But most importantly it has a lot of room for error specifically when it comes to sentiment analysis just like any form of technology does. So dive in a little bit deeper about how we can overcome these challenges and understand how we can apply social media listing as a tool into the mission driven sector. We have Chelsea ins and a here with us today. Hi Chelsea ins and a. I'm so excited to introduce you both we have Chelsea Louie and Zanae Aquino here with us. And so in addition to her marketing work at PT KO Chelsea is a communication specialist. And she also is a co founder and chief communications officer at activism always. And we also have Zanae Aquino who specializes in computer science and is the head of product management and activism always. And she also major her work at Salesforce as an architect program specialist. I'm going to pull down my screen share for our conversation. So both of you are here today to help us understand how we can apply something as complex as social media listening to the mission driven sector. So to start, how would you each define social media listening and what's your experience with it. I think social media listening is again a complex idea. It's again the observation and collection of what's being said on social media. And again, what you mentioned before in the most important part of that it is how you're listening actively listening and applying it to your own case. So my experience with social media listening is seeing what trends are seeing what's being said and trying to cater my content to increase engagement, and I've been able to do it in a variety of contexts whether it be my own personal brand here at over at activism always with the context of being able to be inclusive to different audiences. And additionally, in the corporate sector to appeal to different architects and software industry. Thank you. What about you Chelsea. Yeah, following up on that. I think my experience with social media listening is a little interesting so I was originally on that BLM always project with Mickey way back. And that was actually sort of my first sort of technical project with sentiment analysis and like working with an op analysis and actually learning all that terminology. I think my background is definitely more in sort of the analysis of that works I've never worked with the actual like algorithms to develop the like social media listening. But my background is in sort of analyzing social media from sort of traditional more like communications practices. And I think in that project I was able to see like all these like sort of traditional methods of going on social media and just seeing what people are listening to collecting interesting posts. And that was made so much easier using these technologies. So I'm coming in with a bit of a non technical background, but very happy to be here as well with my experience with BLM always. Thanks so much Chelsea and it's a nice and to work in computer science and have a very strong technical background I know that we already kind of went over this but how would you personally define social media listening. It's, I guess, in the most contextually appropriate definition. I think the difference between social media listening and monitoring is that when you're listening you're applying when you're listening you're actively taking the insights and curating your content. So I think it's a fair to adapt to that. And social media listening is just being able to take the raw data. So taking your likes engagements, being able to track that and see what works and what doesn't work. In terms of curating your presence because you don't want to overload our audiences with things that aren't necessary things that don't work. So by curating and making that happen. That's I think the most important part of it. That piece that you said about social media listening is applying I think that's so important, especially when we're talking about appealing to our audiences. And so why is it important to be talking about social media listening in context to specific audiences. Yeah, so social media, one thing to note about it is that it's not a single demographic. There are all different types of people who are using the internet we have people from Gen Z, ranging to people in like Gen X. People are being people buying different things to be attractive to them to be appealing to them. So there are different interpretations different backgrounds and experiences. And you need to be able to make sure that your target demographic is being appealed to you can't be having your target graphic be. It's important for people like nonprofit organizations and only be taking up a tone that's only appealing to profit organizations that's an important thing to notice. It's important to cater your information that you're presenting in a way that the target audience can understand by either simplifying it by going into detail about a specific topic. All those different kinds of curations and customizations to make sure that you're doing the right thing with your presence. And I think they set it super well already. I think really the point of how social media listening affects audiences is that those audiences already exist out there. Like, people are already using social media people are existing in these sort of bubbles of different audience types. And most people exist within many different bubbles so I think Zanay mentioned like generational gaps, like young people are probably going to hang out with other young people maybe you're interested in one topic like environmental justice for example, you might be interacting in a bubble of people who are interested in that topic. And these bubbles exist with their own norms with their own cultures. So, trying to like create a sort of communications message, trying to do something with social media. It doesn't make sense to just sort of like put out a message that's for a general audiences. Because most people aren't a general audience people have their communities they have their sort of specifics. So, crafting that message involves a little bit more thinking, and that starts with letting the listening aspect of it. Thank you. And so you already touched on this a little bit but if you could just go into some more detail about why the concept of social media listening is so important specifically to the mission driven sector. Like you mentioned Zanay we know that a lot of more like for profit companies and corporations have used social media listening for a long time but why is it so important to incorporate that into the mission driven sector. I think it's important that the concept of virality in terms of defining that as just high traffic high engagement, and the spread of ideas. It's not limited to just the commercial side of a lot of the world people aren't just appeal to like how much money's been being made or what's super cool and everything with the mission driven sector what I think is really important about that group is that a lot of the times the narrative that's adapted to that is a more empathetic kind of view. And especially as we can see after the whole Black Lives Matter movement took up and momentum and everything that a lot of people are appealing to the human side of the stories being shared. Social media isn't all about just irony and humor anymore it's about building connections understanding different types of perspectives. And in order to understand why it's important specifically the mission driven sector. You could just see how social media has impacted everybody outside of it. Think about go fund me think about like different sorts of like the fact that we're having this conversation as a result of the kind of momentum that social media has been able to you're able to spread the word out you're, and if you're able to increase the engagement for it you can spread so many different types of messages so many different types of missions. And it's important to see the concept of getting popular on social media as more of like more people are willing to hear what you have to say. And I think the goal of building momentum for a lot of mission driven sectors it's just getting your word out getting your cause to have importance in the eyes of others. I definitely want to add on to sort of like the popularity sort of aspect, or even just thinking about in terms of like reach like social media is fantastic for finding those very specific audiences and trying to spread the message and target that sort of word in these much more targeted ways so many people use social media and they use it for so many different reasons. I think, going back to I think the original question you had Mickey was sort of about the concept of social media listening and how it's applicable to mission driven organizations. I think it's incredibly applicable and I think while we're talking about it in like a semi technical way right now calling like social media listening and maybe discussing some of the algorithms. I think it's definitely like tried and true like procedures for most people use who use social media most of the time you're trying to use social media to listen to what other people are listening to, or like talking about. So the idea of using it for your organization is taking those sort of practices that we may have in our sort of individual lives our personal lives, being able to listen to people in social media, and sort of amplifying at targeting it, expanding the reach of like what we're listening to and also listening to things much more specifically. I think this is definitely like procedures that are more common in commercial industries, like Zanay mentioned like organizations who are trying to sell things they're definitely like using social listening technologies, or having like staff members like be involved in social media and listening to what people are talking to about their brand. So I think there's definitely a room for mission driven organizations to capture these processes to maybe do something other than maybe like sell a product but maybe to spread a message about their cause, or the organization their services. I really love that connection between listening just how we do an everyday life in social media listening I think it's really helpful to see something as complex and technical as social media listening is just another way of what we do in everyday life specifically in the mission driven sector being empathetic listening being receptive to other voices I think that's a fantastic point. A audience question from Mary price thanks so much for your question Mary. And so Mary's question is, are there social listening tools that are better for assessing other languages on social. Yeah, to get into this question, but I think it's under, like it's important to understand how the listening happens. In terms of social media listening it's all data driven so what it's doing is that it's taking what's currently happening pulling that information so what I mean by that is that the current feed of tweets, that sort of thing, and you could specify, and then I'll say sorry. And then what you do from that is that you're drawing insights from the words that are being used like if you remember what Mickey was showing earlier in terms of positive negative neutral analysis. So what that is is that there's a trained model that's able to determine like what's positive and what's negative. And so, in terms of actual things in the market that are adapting to different languages, those models need to be listening to those specific languages. At the top of my head I'm not completely sure what in the market is currently appealing to different sorts of languages but that in general structure is kind of what's happening so looking for social media tools that are multilingual in terms of pulling different data from different different like regions different like language specific regions is just an important thing to look out for when you're looking for those tools. That's a great point that reminds me of one of my projects here at Parsons TKO where a client was looking at a specific keyword on social media. That was Spanish and there are so many different variations with accents and different spellings and abbreviations that weren't originally accounted for that they would have missed out on all of this insight so. Thank you for that point. And we have one more question from Mari, which is do you find that some social platforms are easier to listen to than others simply based on the structure of the platform. That's a great question. One thing that I think it's important to note is that once again going back to the sentiment analysis that works the best on and the easiest I think on text based platforms to think about like general Facebook post and tweets. That's like the easiest to run a sentiment analysis on compared to, like, for example, Instagram or TikTok, where you have moving images and like the analysis and the models that have to go into that have to be much more complex. I think just as a quick add on to that I think in addition to the tool or the technology or the platform you choose to listen to. It also depends on your organization itself and sort of like where have, where do you know that you have communities and where do you want to learn from. You're picking Twitter because it's easy or because there's text there, but you have no community there at all, or you're not sure even where to start. It might actually be a bit more difficult to find sort of relevant insights. I think similarly like people on staff of people on staff are really familiar with like looking at Facebook posts. You guys have been posting on Facebook forever. You know sort of what's the cadence of how people post on Facebook will be much easier to analyze at the end of it all. Then if you're picking something like TikTok that maybe no one on your staff uses. So that's also a little human aspect to keep in mind. I think that tangible action step and speaking of that do you have any tips or recommendations for how social media listening can be adapted to the mission driven sector. I think for tips for adaption so like adopting the technology. The biggest thing when adopting any new technology in addition to like social media listening is to remember it's it's never like a one to one adoption. I think I just mentioned that like, just because this tool might be really good and a lot of people use it. It doesn't mean it might be the right one for your organization you might need to tailor it a bit. So for the same example, like if your organization is really familiar with Facebook you know a lot of people on Facebook talk about the topic that you're interested in maybe like the geographic location you're in people use Facebook for like finding things around the services you offer. It makes sense to find a tool that might be more suited for that rather than maybe get use like a big Twitter tool or a Twitter sort of like algorithm that everyone else uses. So I think that's just something to remember. Yeah, the value of the of the technology only matters as much as it's fit to your organization. So yeah, I think that also goes back to I think Mary's question earlier about like languages. So maybe like the English models are like the really popular one. But if you're like doing work that maybe is majority in Spanish, you might need to look for a more specific model maybe more specific analysts to understand that technology. Fantastic and keeping momentum of shifting a little bit from the conceptual theories of social media listening to how everyone here can actually apply it to their organization. So the tools have been used for kind of introductory social media listening, or alternate ways of listening or hearing from your audience even if you're not using these tools. So maybe you have any, any tools in mind. Yeah, so currently the product that we're working on an activism always is this product called Mary Alice. And what the thing that I think is the most important about it is that it provides concise feedback in terms of what sort of what kind of data is being pulled. So the best ways to have an increased presence is like knowing the timing of your post knowing like when is your audience the most active, being able to draw insight for that and timing your posts around that time, using keywords and hashtags, like if people are tracking specific trends or the algorithm itself is grouping specific things how can you make sure that your information is being captured into it. Well, yeah, with Mary Alice is able to track just your own personal trends but trends throughout the entire Twitter sphere and dictate how you can advance your engagements with those in mind. I've also seen one of the most important. I think what's one of the most important things to keep in mind with social media engagement one of the biggest trends I've been able to see right now is that personal engagement in terms of like, how can they interact with your posts that's something that has garnered popularity so whether it be polls, whether it be like retweets and like quote tweets or just answering questions that you kind of have there. Those are what I've been seeing as the major trends through the analysts, the analytics that we've been able to provide, as well as a different, yeah, different just sort of insights in that context. This is cool tool tool that we're working on. And I think, as we develop that I think a much more lo-fi method that you can also use is really just sort of like the manual staff member checking back on your posts. If we're going back to like social media listening as a name it's it's listening to your social media. Thinking about your day-to-day sort of interactions of social media if you're putting out a post on LinkedIn and you see maybe in the notification that you're getting five, 10, 15, 20 sort of like likes or comments. Are you actually going back and seeing who's messaging you, seeing who's liking those posts and are you actually re-interacting and re-engaging with them? I think that's a very basic way to be like listening to your social media and actually just like taking the step and taking action with your social media engagement. That's something that I feel like most times we think of and we're like, that makes sense. You should do that and may fall to the wayside because there's so many other things going on. But really sort of the manual labor of going in and engaging your social media audience and listening to them is a very lo-fi way to get involved with at least these procedures. Because I'm sure not everyone has a huge budget to be like getting an analyst that can do sentiment analysis in NLP like at the moment or even like adopting a new tool. So those type of procedures to go back and listen can be really useful. Thank you for highlighting that difference, Chelsea. For example, with the BLM Always project, we were analyzing 10k tweet today over the course of about three months. But that volume of course, the help of NLP and AI algorithms is very helpful. But if you're just doing your organization's tweets or just from one day or even just a smaller part of the sector, it definitely is super accessible just to use those in-house Facebook analytics or Twitter analytics and things like that. And I guess, and last thing to sort of highlight that difference, again, right, like the social media monitoring may be only looking at the tweets or the posts that you put out. And that can be really interesting in lots of ways. You can track sort of like your growth over time. And if we want to do listening, it's as Mickey mentions, maybe carving out a small community within sort of the sector that you occupy or the industry that you occupy and listening to not only like what other people what's happening around your posts but might be happening around sort of like peers happening around like activists in your organization advocates for your causes and trying to like patch those conversations together. And like, is everyone talking about this one topic this week. Why are they talking about it and doing that little deeper research. We have another audience question from Lisa, which is how can you use social listening to identify a target audience or how a brand is perceived by different segments. I feel like that's such a good question and so many people have this question so I would love to hear both of your cakes on this. Yeah, so one thing to note about when you're collecting information and data is that in order to find out if something is good or viral or popular you're grouping things together. So what that means is that in terms of determining different graph demographics you're not necessarily assigning an identity to each individual user, you're more so tracking like okay what's being used popularly is a common trend that you're able to garner from that. As Mickey mentioned earlier, like with the senselessly murder that's high, high volumes of people, typically, not to generalize the people who are having like a higher lexicon with those sorts of terminology is part of a specific demographic compared to Gen Z for example that happened to ironically use like the skull emoji. So, like again different context, kind of are important in order to determine that and from like the data analytics side, we'll get into this later. It's important to have different demographics, or different trends catered to by by having a diverse background of like data that you're kind of filling in. Anything to add to that Chelsea. I don't think I have anything to add to that specifically. I think identifying target audience I think as the name mentioned, you're going to get a lot of data and it's going to be a lot of sorting through. That sort of like procedural practice that you hone in on that target audience and it might take like like Mickey mentioned we collected a lot a lot of tweets over a span of many months. It also requires you building up sort of that's like building up the amount of data to actually be able to sort of like parse out those trends. So if you're doing a more lo-fi method, it can be more difficult. It means more, more tweets that you have to like look at manually. But I think it's part of like, are you documenting it? Are you keeping track of like the tone of the post that you're getting? Are you keeping track of like the sort of sentiment analysis you're getting? Are you getting overwhelmingly positive? Why so? And like following those leads in order to create that sort of target audience? Or seeing at least what audience you're filtering out the most. That ties into a question that Jamie asked about our BLM Always project. And so Jamie asked how we re-identifying the hashtags overall and was that ongoing or all after the fact? And so Chelsea and I will let you both add on to this since you worked on the project as well. But the first thought that comes to mind is that we mostly did a lot of word association. So hashtags that were used also with BLM, such as Black Lives Matter, or Stop the Hate, or anything like that, that were associated or used in conjunction with that BLM Always. It was a lot easier to identify them that way. And then in terms of like the flow, it was all mostly after the fact. Chelsea, is the name is that about right? Yeah, I think it's about right. Yeah. And we sorted both on sort of like major keywords that were used in like different tweets and we also looked at trending hashtags. So those were sort of ways that we were able to do our analysis. I think you mentioned directly sort of the hashtag visualization. But it was really just like from all the posts we were getting and then sifting them out in relation to sort of like the BLM Always or the BLM keyword. And that was sifted out mainly using NLP. I have a follow up question for you from Andrew Courtney and the observation about manual and hands-on. When would you suggest that someone or an organization invests in tools versus conducting it manually? Is it purely based off of volume or some other factor? Yeah, I think similarly to my answer earlier, it really depends on where you're at with your organization, right? If it's sort of the first time you've ever heard of like social media listening at all, I'd recommend maybe doing a little more research, maybe trying to do it manually and seeing how it feels to actually engage with that practice. I would say to not shy away from like contacting sort of like PTKO, we do a lot of work around this type of these type of conversations, right? So contacting someone who maybe have a bit more practice like presenting, consulting, presenting sort of like strategies for using social media listening. That's like one method. If you're investing on tools themselves, that really, really depends on your organization. Like, do you have the budget for it? Do you have someone that can actually like take that data and analyze it? I think if there's someone in your team that's already like, I know what social media listening is, we definitely have the budget. We definitely know what to do with it. If you have like a goal, then to like go ahead and invest in those tools. Invest in sort of like higher end consulting, whatever you need. But until then I'd recommend sort of thinking about doing it manually and then maybe contacting someone, calling someone that can discuss strategies to move forward. Yeah, it really depends. Just to add on to that, I think probably the best application of using tools is in conjunction with that strategic part. Like you mentioned that we do a part since TKO Chelsea. But a lot of time the data is just gathering the data is half of the battle the other half is actually applying it to strategic insights. And Joel has a similar question about that. Could either is an air Chelsea could you provide an example of how the insights that have been gathered from social media listening led to a newly informed strategic or tactical action for an organization. I mean, I can talk a little bit in terms of the bill I'm always project that we worked on. So yeah, I think this project was a bit of a passion project for our team. It wasn't for any specific client it was something that we were just, it was, it was a situation in which we were very passionate about what was going on. We saw sort of the context of it, a lot of what was happening was on social media. It was right at the start of the pandemic people are at home and social media was such a sort of like potent communications, sort of channel I guess. So I think for us, using nlp and using sentiment analysis helped us understand what was going on as individuals as sort of a team of like analysts and researchers to understand what was going on is sort of like the wild wild west of Twitter, talking about these massive protests, and being able to take insights that where we can sort of talk to our friends about it, we can talk to other teams about it. When we're developing a we were part of everyone in our team was part of a mentorship program. So we could share that information and really make sense of sort of like the big conversations out there. This is maybe not particularly strategic or tactical for for an organization because the project was rooted in sort of like personal interests. We were able to sort of take that information get that those like big ideas was big conversations, and sort of create it into a visual like the one you saw with the hashtag graph that was moving. Those big conversations and make it something a little bit more palatable a little bit more targeted. So when we do lead those conversations. We have something to show and something to share. I think in terms of your organization if you did something similar, you can use it to discuss things like major topics in our industry that we need to focus on more. It can be used to help support budgets, it can be helped to use sort of like support where your team wants to target your social media messaging in the future. And can be some type of tool that you can use to sort of understand where the trends are going potentially. So little all over the place wasn't a specific example, but I think really the power of social media listening is being able to listen and take all that information and create something a bit more targeted. Yeah, I'm interested to hear your perspective on this Chelsea specifically as a communications expert. How can social media be applied to like marketing or branding strategies. Yeah, I mean I think the biggest thing is you're able to, I mean we talked a little bit about audience already so able to like see the type of language that your audience is using see sort of like the trends and topics your audience might skew towards like the basic examples like are they talking about this really positively, like is everyone talking about it filled with, like at least generic examples of like love and positivity and happiness on the topic, or people are really angry for valid reasons or is this anger in part because that is the language being chosen by your audience members. When you're creating a message, you can take sort of those inspirations from your audience and how they're talking and being able to create a message that's a bit more appropriate. I think it's so easy to have like really good intentions and being like, this is a great post this is a great message that I think my audience will love. And then you put it out and it's completely tone deaf, even if you have the best intentions, it's being able to like kind of check your work, make your work more targeted. And I know that's something that you mentioned earlier about understanding different generational differences so how could NLP and sentiment analysis and other algorithms like that that are part of social media listening become more culturally inclusive. Yeah, in this case I think it's important to note, like as you mentioned before with the BLM always project you're pulling like 10,000 tweets I believe over the course of like months. When you're pulling all of that data and pulling all of the information from the Twitter sphere that allows you to draw insights of all different kinds of people. So you're searching for like just using a specific keyword and not necessarily considering what other people are saying. That means that the model that you're building the insights that you're creating are going to be curated to a specific audience when you need to be able to look at it as a whole. So yes, again, that's why it's important to see what kinds of words are being used by a lot of people, like be able to be conscious of spelling miscommunication, and also different like contextual differences and different cadences and different kinds of terminology. And that comes again at the diversity of data analytics whether it be the team that's listening to it whether it be the types of information that you're considering is correct. That's what you do when you're looking at data is clean it as in like this term and what isn't contextually correct and what is is, and that grouping once again comes when you're able to keep in mind the different kinds of context. So in terms. In terms of being inclusive in our data collection, making you have to make sure that every person all the people that sometimes fall into the cracks are being listened to, and that you're turning those trends as a priority when you're being inclusive like in the case of sometimes an NLP jokes or references that only apply to a certain community are misconstrued pretty often, or like terminologies being used to intensify a term but it isn't necessarily like captured so I think it's important that the algorithms are just conscious of that kind of data. Yeah, computer struggle sometimes with understanding context, because it's again, you're it's being trained solely by what you tell it. And again it's kind of hard to teach a computer a joke. But overall I think algorithms are working towards building more inclusive models because people are being conscientious of that especially with a new generation coming into the workforce. Anything to add to that Chelsea before you move on. Anything to add on to that specific things and I said perfectly like it's, it's hard to teach a computer a joke I think when it comes down to it. Your computer can get a lot of information and then from there you actually have to make, make the decision of like, do you believe what the computer is saying. How are you going to edit the computer to make it make more sense, and also think about sort of the staff members on staff the analysts you have that are actually working on it. There's so much around like, we have inclusive technologies or we have like inclusive algorithms, or more powerful algorithms but when it comes down to it it really is sort of like the person making that strategic decision. We're getting all this information do we actually like set out a campaign with this language, because the computer told us to or because like the people of color that are on our staff. They didn't give us like a go, or does our staff is our staff actually representative of sort of the messages that we're putting out or we're putting out for like to like put out a face. Yeah, I think when we're talking about like diversity and like messaging, or to be culturally inclusive and messaging it goes way beyond the technology. Sure. And following on that similar vein I know that the industry standard for assessment analysis is positive, neutral and negative. But our audience is wondering if algorithms can recognize things like sarcasm nuance justifiable anger, sadness, etc. To answer that I think you know go back to remembering that a computer only knows as much as you tell it. So, in terms of senselessly murdered that negative kind of context, sometimes when you're doing that cleaning itself you have to be able to catch those things in the track, or in like the cracks. And that once again comes from like a more manual sort of thing. So in terms of capturing sarcasm, like, in order to do that you need to be giving the computer different kinds of context telling it what's right and what's wrong, training the model and being like okay this specific wording of this is right, the specific one is wrong, and that sometimes comes at like the cost of more time and processing that needs to happen. Otherwise, it is possible, but in the industry right now it's a challenge. I love that piece about AI that you touched on we didn't really cover that in the presentation but all the things that we're talking about our artificial intelligence and so just as in a said you have to train the models. So they have to be given that information in order to understand those different things and so if these models are trained explicitly on Gen Z humor which can be a mess and we can say that because we're all Gen Z. then they may understand those types of things more easily. Right, and jumping to our next question so part of the purpose of social media listening is to expand your reach, because there can be such a wide range of users and audiences, but because it's not very common for the mission you've been sector to have diversified internally, it could be hard to gauge for what that cultural context is like you were saying for on staff members Chelsea. So how can social media listening help mission driven organizations better understand their audience but also vice versa. So I'm taking this question is to see like, how can social media listening not only help like the organization craft the message for the audience but maybe how it can get the audience to like interact more with the organization itself. Awesome. I think it goes back to what I was saying about like the tone deathness of a lot of social media messaging, especially if you're an organization that doesn't have a lot of experience, like crafting social media messages, or crafting sort of like very targeted communication messages. So being able to create a message that resonates with your audience that uses language that actually catches our attention, or keeps their attention, or actually feels comfortable for them to engage with is huge. I think it's easy to be like here's my standard sort of press release type language that I will I will put out into the world. And to expect people regular everyday people, like most of us out here to to feel like compelled to actually reply to that post, or feel compelled to like do something with that post other than seeing it as like, Oh, there they posted a new blog. Is is really, really powerful I think it's it's sort of like the organization taking a step out to understand the audience for the audience to take a step back, or take a step forward and to understand the organization more. And when we're talking about sort of like inclusive messaging or culturally sensitive messaging it's thinking of like, when we're crafting these messages we have so many different people out there. It's really, really rare that you have a audience that's totally like, that's totally like, what's the word. I can't think of the word but the word to say like everyone in the audience is exactly the same right. So creating a message in which like everyone's getting a little closer together the audience is taking a step forward the organization's putting their foot out to really try to like be closer and actually like have a better more valuable conversation that goes beyond the sort of like, I'm an organization that does cool stuff but I will share it in like a bland general audience kind of way and expect audiences to, to take like five steps forward to engage with that. Yeah. I'm not sure if I answered that Ricky so tell me if there's anything you want me to add. I think that was fantastic thank you. All right, so our last question I have for both of you is, how can we imagine the future of social media listening. What do you wish could be done differently and also how could it be improved. I know. Yeah, it's a heavy question. Like, how are we going to advance this as a whole. I think in the market right now most social media listening tools are looking at your own personal kind of engagements as in like, you're only finding out what a post of your own are working. But as you mentioned before I think it's important, or I think the industry needs to move in a direction where you're listening to all of the trends as a whole, like you're listening to all the different kinds of things that exist in the spaces, so that you're catering your content just to like what you find individually works but rather what works in the context of like the social media space as a whole. So, like, again, like listening, going back to what listening is in general when you're a good listener you're taking the information that you're learning from another person and applying it to your response. So if you're not if you're just building off of what you said previously you're not going to be seen as like you know, an emerging kind of speaker or that sort of deal. So by just applying that knowledge from seeing what other people are saying seeing what works. I think that's going to really shift the way that organizations and also people are dealing with their social media. And I guess to just add on to that I think the one thing that could be improved I think is that a lot of this technology and a lot of this work is is described as highly technical because you are using those algorithms you do have to like run those tools. And that typically requires someone that has a bit of a technical background to be able to do that. But when it comes down to it it's it's sort of like, as the neighbor already said like it's it's listening it's good listening. And these are tools that like make that good listening easier and faster, and you can get a much higher volume of information, but it's good listening. And it really starts with like social media literacy within your organization like does your organization care about it. And does your organization sort of care about like being better listeners and having better literacy about your industry about your peers about your audiences. So, yeah, I think that's something that comes to mind of what can be sort of done differently is like shifting how we view it as something like, doesn't have to be an expensive tool or any very technical like, like thing that you have to adopt on the go. And as like the world is shifting feels like everything's moving so quickly but it just requires like, do you care about social media as a communications channel. And if so, are you listening to your audiences to make good social media like communication practices in your organization. And then the tools come with, with that developing sort of procedure, those developing practices. That's amazing. Thank you so much to both of you for all of your amazing insights Chelsea and Zanae I really appreciate and also thank you so much to all of our attendees for joining us today we're so happy that you're able to come. And we also would love to stay in touch we have our LinkedIn page and group links up here for you to join. It would also love to talk with you one on one about how social media can help your organization and how to adapt it and adopt all of those complex technologies that Chelsea and Zanae