So What? Putting your SEO data to work

Well, hello again, happy Thursday. Uh once again, we are the Marketing Analytics and Insights Live Show. I'm joined by Chris and John. Um, you know, so the past couple of weeks, we've talked about uh SEO data. And so two weeks ago we talked about keyword research.

Last week we talked about competitive research. And so if you're following along, we assume that now you just have a big old pile of data and you're like, so what? What do I do with all this stuff? Well, you are in luck. Today we are talking about putting your SEO data to work.

We are gonna cover um how to make your keyword and competitor competitor data actionable. Um how to use the data for all for wow, my Boston accent is really trying to come out today. Uh, for all of your digital channels and how to use the data in a predictive content calendar. Um so, John, Chris, where do you guys want to start today? I think probably the place to start today would be to talk a bit about how to use this data in a in a forward-looking fashion.

Because um one of the things that we talked about recently was that you can look out to the rear view mirror of your car with like perfect clarity and know exactly what's happened, right? But it's a very limited utility uh for figuring out how to drive the car going forward. It's difficult for people to go, oh, how do I start forecasting with some of this stuff? And so I think we should probably spend a little bit of time talking about like what forecasting even is, so that we know how to make use of this data. So when we talk about predictive analytics, which is what we're really talking about here, for using this this SEO data, we're talking about uh in this case time series forecasting, saying here's a bunch of data from the past, and as long as the data has two key ingredients, um we can use to predict forward.

And those two key ingredients are uh cyclicality and seasonality. Seasonality is changes within a year, uh this time span of a year. So there's seasons to a year, there's seasons to your data, and cyclicality is repeated patterns in the data. Now that can be of any duration. So we're a B2B business, uh, and so you'll see like in our Google Analytics, there's a cyclicality.

You know, Saturday and Sunday, no traffic, you know, Monday through Friday, you know, lots of traffic. You get this cycle, but there's also cycles on weeks, on months, uh, even on years. Some years, you know, there's like if you're a business that does consulting, say in pull in politics, you there's every four years you're really busy, and then the rest of the time you like try to figure out what to do with your life. And so I know um from past talks we've given, when we talk about how to understand how time series forecasting works, uh, we often give the example of Google Maps, um, because the time series forecast is based on um an SREMA model. And so basically the way that we talk about it is if you think about you know getting from destination A to B, when Google Maps is serving up the information of how long it's going to take you to get there, it's using historical data based on that same route that other people have taken and roughly how long it takes them to get there.

So that's where it starts, but it continues to update you in real time based on weather, on traffic, uh on the speed that you're going, on the time of year, those types of things. And so what Chris is describing is taking that historical data and projecting it forward in such a way that it's updating and you can use it right then and there. Exactly. And that's why one of the reasons why you have to do some of the math on this stuff. Because unfortunately, with trend prediction of any kind, you really can't eyeball it.

Um, there's there's and there's a couple of different types of math that you need to do. One is uh determine is there even a trend, right? Is there even you know something that can be forecasted? So uh show you a couple of examples here. Um let's go to I'll put up this lovely slide, right?

Here's Chris. What that was that was not a nice looking graph. It's fine. Um so in this example, there really isn't a meaningful trend. And there's some mathematics on screen.

Maybe in another show we can actually talk through the the specific mathematics is the statistical testing, but not for this show. Um, but essentially we're saying this there is no trend here that you can use to forecast this term, and because there's no trend, you can't use it for forecasting, it's no cyclicality. Um, if you look at this uh chart here, there is a trend, it's weak, but there is a trend that is there, and thus you can forecast with this term, and then you know, this is TikTok. Again, same mathematics, but you can test it and make sure that oh, there's actually a pretty strong trend here, so you can use this. So with your search data, one of the things that ideally, if you're using software that has predictive analytics capabilities, the software has built-in testing to say, like, yep, there's a this is usable, there's a trend, or no, there's not, and then and you know not to use it for forecasting.

So that would be the first and most important thing is the data you're using, does it have cyclicality? Now, some data, and like SEO data, thankfully, um, is something that it has tends to have a lot of seasonality and a lot of cyclicality, and and particularly for generic terms, when we look at everything from like holiday gift guides to consulting firm to do whatever, those are terms that because they're not really unique and they're not really uh you know a small niche, they have a lot of seasonality and a lot of cyclicality. And one of the sort of the axioms of forecasting in SEO is the more difficult the keyword term in terms of ranking for it, the easier it is to forecast, right? You know, because there's more volume and it's been around longer. So the that's one of the reasons why we say when you're putting together your keyword strategy, we talked about this two weeks ago.

You have that core term, like analytics consulting or change management, and then you use tools like answer the public to build you know your question content, your your your corpus of content around that core term, but that core term, because it's so competitive, it also makes it really easy to forecast. That makes sense. Um, obviously, more data equals a better forecast. And so I know you're gonna walk through putting that predictive forecast together using your SEO data. Um, I do want to sort of mention that obviously we're gonna be using software that we've created, but if you had like a small handful, like less than 10 terms that you really cared about, you could plug each one individually into Google Trends to see if there is any kind of seasonality so that you could sort of have in the back of your mind, okay, when I want to be talking about change management, it looks like for the past five years, people tend to talk about this in December, which makes sense because they're putting together their annual plans.

You know, so there is definitely ways to use uh these free tools without having to write any software. Will it be as efficient? No, but will it be effective? Definitely. Exactly.

And in some cases, depending on the search term and whether Google Trends supports it or not, uh, as an example, because Google Trends may actually offer a forecast. It does that, it doesn't do that consistently. So if I pull up uh Google Trends here, sometimes you will see a little dashed line at the end. Um and Google will say, like, yeah, this is our our our attempted forecast. So let's look at the the I guess the the non-software route, the non-build-it-yourself software route.

To what you're saying, Kate, if we look at this search term for change management and we'll look at the last five years, there is a clear cyclicality and a seasonality to it, right? There's inter-year season, interyear season, and there's year over year season. Those dips there, of course, are the holidays, right? Every year you know that whatever Christmas and New Year's is, people aren't searching for this term. They're they're doing anything but.

And so roughly speaking, um, if you wanted to to, you could actually take this graph of this data out, export it, put it in something like in Excel, and just line up, you know, week over week for every 52 weeks and see, okay, what weeks of the year in the past has this more or less uh increased. And so that would be, I guess, the the least onerous uh way to handle that. If you have access to um tools like IBM Watson Studio, for example, you can plug the raw data into Watson Studio and have it build you a forecast. Um we use uh the programming language R, which does essentially the same thing with um uh a type of model called profit, uh, which is made by Facebook Labs, and exactly the same thing. Takes five years worth of data back and then projects it forward.

And so I think you know the past couple of weeks we talked about you know getting that keyword research out of those keyword tools, but then you also want to use a tool like Google Trends for exactly that reason because it's literally what it says. It's the trend of search terms. And so, how is that specific term going to be trending? Exactly. Um, do we want to go into like trying to assemble one of these models um with uh Watson Studio Katie, or do we want to go straight to to processing the data itself?

Let's just do a quick uh walkthrough of Watson Studio. No, so let me ask you this question Is is there a free version of Watson Studio that people could play around with that you know they don't have to sign up for an enterprise license? There is. So Watson Studio um offers uh I believe it's 50 hours of compute time. It's like 40 hours of compute time uh a month for uh for free at the free tier.

And then you obviously have to to go on and and and pay a whole bunch of money for for it. Uh it gets uh reassuringly expensive uh so I'm guessing that most people aren't going to be using you know if um like people like us who are just sort of experimenting and seeing what they can do aren't going to be using you know those 40 hours of compute time exactly yeah you're probably not gonna do that right into that so this is what Watson Studio looks like I'm gonna go ahead and take my timeline uh my export so I just hit export on that Google Trends thing and I'm gonna just take that CSV file it's a real simple CSV file um just show it briefly on screen here right there's week and then there's the change management term real simple uh not too not too terribly complex I'm gonna start a new auto AI experiment we're gonna call this change management consume 20 hours uh 20 capacity units per month is uh per hours what the the cost of it is and you get 50 uh units to play with okay um and that's per hour so if you like turn it off then you're you're fine I'm gonna choose my uh multi-timeline like that and says hey do you want to create a time series forecast well yes yes I do uh my column is change management my date column is the week and I because this is weekly data I'm gonna crank this up to let's do 26 weeks so that's half a year ahead I hit go and now it's going to go and try and and look at this data, analyze it, pick the best algorithm that um that it can from its portfolio of things it knows how to do, and then ultimately uh spit out an answer. Now, this could take anywhere from 30 seconds to a couple of hours, depending on the complexity. This is a relatively very simple data set. So it shouldn't take more than uh a couple of minutes to run.

But what it's doing now is it's saying, okay, well, take this five years worth of data that you've put in here, and I'm gonna do that math and look at look for patterns, look for trends, look for ups and downs, um, and try and guess based on all the uh tools it has what the best fitting algorithm is for it, and then produce a forecast. So, John, you uh you interview a lot of people for marketing over coffee, and you know, you're pretty active in the marketing events uh year over year. Do you hear a lot of conversation about predictive forecasting? Are people talking about it? Is it is it something that you think marketers are using?

You know, I think it everybody wants to get there. It's definitely one of those shiny objects. And you know, the unfortunately the truth is if you don't have your data in order, you can't get to it, you know. And so many people just struggle so often with getting their data in order to get there that they're you know, they don't even know where to start with this stuff. The the thing that gets me the most is pretty much every business that you know, both for marketing over coffee or even for trust insights, you go into and management is always convinced that they already have the seasonal cycle, that they know exactly when things go up and down.

And there's been many cases where we actually run the data and well, let you know, yeah, you're right about those two, but actually here's three others that are far easier and more important that you're paying no attention to. So it's um yeah, don't leave it to your gut. It's too important for that. I think you know, you're speaking to a larger issue uh in terms of using data to make decisions, but we can certainly cover that in a future episode as well. Exactly.

Uh while we're waiting for this, Brian had a question here uh asking how much SEO work do we do around branded keywords? Um a lot, actually. Uh we use uh branded keywords and branded search, we use actually use um Google Search Console data for branded search terms to measure the impact of brand, right? To measure you know brand strength. It's one of the best measures available for doing brand equity and brand strength because the more people who are searching for you by name, the more likely it is that um you know you you are doing a good job with building awareness and trust.

Okay, it looks like so. What's happening under the hood here is that Watson has gone through, and you've if you've been watching the uh the the very cool looking graphics, uh, which hats off to them, I think you know you could see this in an episode of Star Trek Discovery at some point. Uh it's going through and testing all of the data. And it's doing three things. One, it's doing what algorithm is it, what series of steps are you going to take at the pipeline, and then uh what transformations are you going to do on the data.

So it looks like it has chosen uh a random forest model here uh as the the winner uh based on the the error metrics chosen plus uh feature engineering. And what we can see here is no surprise, you know, as as it does its forecast, there's the Christmas dip, and then so on and so forth. So we can take this uh forecast result uh and then export it and then essentially start building a calendar from it. You can see uh its predictions and its forecast values. Now, the one thing I will note here if we look at the back testing performance on this um and the prediction performance, this is actually not a great fit, right?

This is not a a super good forecast. It's it's okay, but it's not great. And the reason that's not great is because some of the the more cutting edge predictive algorithms are not things that find themselves in into enterprise tools like Watson Studio because they're unstable and IBM has a reputation deservedly of not selling unstable things to its customers. Um when you build your own code, uh you can do that. You can you can play a little closer to the edge, um, occasionally falling off.

Well, that said, it's still something that I can use because was it what it does is for me is it at least validates that okay, during the holiday season, nobody's searching for change management. Therefore, that's the time when I should be doing my planning for the rest of the year's worth of content. And then, you know, it's interesting, the you know, peaks and valleys of the trends are not so significant that I shouldn't be putting out the content year-round. So I should be planning for the next you know 50 some odd weeks leading up to the next holiday season where I can you know restart my planning. Exactly.

Now, so this was a step through of just one keyword. When you do run this through um, you know, software either you've built or larger scale systems, because obviously one of the things that uh is a benefit of a system like Watson Studio is you can deploy this model and then run additional keywords and data through it programmatically if you have the a developer that can feed the data to it through the API it creates. Um we do essentially very similar with uh tableau software and um uh and some software we wrote in R. Uh let's go ahead and look at this from like a calendaring perspective because no one wants to look at that graph. And what we see here is searchable week by week.

Uh if we just sort by column, what are the uh of these terms? So remember this is these are the terms that McKinsey, Bain, and BCG are all fighting each other over for. So these are terms that they all three find valuable. These are terms that all three, uh again, very high keyword difficulty for a lot of these, but we're gonna use that to as at the as our starting points to understand uh these terms. So this week, the week of October 10th, digital marketing is uh the top search term, following business company strategy, best in class and management companies.

Now, if we look out two weeks or three weeks, let's go to Halloween week, right? Digital marketing change management actually starts to pop up there as the number three term uh in a couple of weeks, uh, followed by strategy and digital transformation. Now, if you want to get well, the the the first and most obvious thing is you know, to what you're saying, Katie. If you are if you want to create content for change management, then we know that this is a term that you need to plan for, right? This is a term that you need to focus on and look at week over week and figure out when in the next 52 weeks or so uh is this term gonna be most popular so that we can build content around it.

And so, you know, it's interesting too because it's like, oh, well, you can roll the topic of change management into a lot of these other uh keywords that are gonna be spiking in addition. So digital marketing. I can talk about change management in digital marketing, uh, strategy, digital uh transformation, all of those things, I can talk about change management in that context. So I'm not solely focusing on line three of change management. I'm looking if if my core term is change management, I need to be looking at all of these terms and when they're going to be spiking so that I can talk about how change management plays into each one of those things.

You are getting towards one of the secrets that I was not planning on talking about in today's show. Well then, well, then let's just stop it right there. Um, there's this concept of intersectional marketing, uh, which is essentially what it sounds like is the Venn diagram of different terms, right? Of different terms or topics or ideas. Um I was watching uh a marketer who uh works in the adult industry talking about intersectional stuff and and how you can take multiple niches, multiple different um areas of focus and combine them in ways that you wouldn't necessarily think would be great, but as long as there's enough interest in each of the niches, then those intersections are where you find the die hard fans of both things.

So to your point, change management and digital transformation. Now, what you would want to do is not just look at the top list here, but actually run a large-scale cross-correlation to see what term, if we choose change management mathematically, what term looks the most like change management, has the exact same ebb and flow throughout the year because those are the terms that those are the topics that cluster together. One of the things that people tend to do wrong with SEO tools is they tend to look for semantically similar terms and group them together. And it's not, I shouldn't say that that's wrong. Um like change management, organizational change, uh managing change, uh, change management for companies.

You can say they're semantically related. But if you statistically compute the ebbs and flows of the terms themselves, you may find there are things that like change management and diversity and equity and inclusion may march together, even though they're they're different topics, they're temporally concurrent, and so you can find interesting intersections to create content that hits both interests at the same time. Which is funny that you gave that example because as you're saying that they're two different things, I'm like, those are absolutely a hundred percent related. Like that's they are conceptually related, yes, but not linguistically. No, not linguistically, but you know, conceptually those two belong together, and that's a topic that I absolutely it's actually on my list of things to cover over the next couple of months.

Exactly. If you look at the graph, you look at the graph of what this look like, though, you can see, for example, here strategy and business company, these two terms march together pretty well, right? So let's do makes sense. Uh business company. So, John, um, not to leave you out, I'm gonna give you an assignment because we have to assign you something during every live stream, is you need to go through all of these different colors and match up the corresponding colors and group them, you know, using the regular rainbow palette.

Right, finding the matching curves in this wear as well though, graph. Exactly. You don't actually have to do that, but I feel I don't feel right if I don't give you something to do with that. Yeah, if there's not something on the test list. So you can see there's there's change management, which is sort of that brownish color.

Then there's there's business company and strategy. Those things actually do march together pretty well. Um you can see that from a trend perspective, they are fairly well aligned. So those would be three terms that you would say not are they're not a topic cluster, which is again something that you see a lot in in SEO tools, but they are a time cluster. And from a from a forecasting perspective, that time cluster is important because if you whatever's going on, under there's something underlying going on.

Just like we know in September of every year, for B2B, that's budgeting season, right? So all the cont things that that happen, like okay, we need to do an attribution model, we need to do ROI and ROAS, we need to uh do our incrementality measures and and you know, do an uplift model. We know all those things come together even though they're not word related, because people are like, my butt's on the line, I gotta figure out my budget for the next year. What are we gonna do? And so if you can create content that detects the concurrence in time, you're gonna do really well.

So noted. Um we had wanted to also talk about how to use. So the past couple of weeks we've done a lot of SEO keyword research. How do we use that data across all of our digital channels? Because, you know, obviously, you know, when we talk about keyword research, there's this assumption that it's just for writing content.

But how do we use that data across other channels? Depends on the channel. What pick a channel, Katie. Let's start with email, your favorite digital channel. It is my favorite digital channel.

Very straightforward. Think about email is just sending content to somebody in their inbox as opposed to sending the content to that person through Google or sending that content to a person through TikTok. So when you look at this calendar, well, actually, let's switch calendar this year. When we look at at this uh thing, if we're to if we're putting together the email, the Halloween week email for for the Trust Insights newsletter, uh, digital marketing and change management are sort of you know, in the top three terms, there's strategy and digital transformation. That week I might suggest that the cold open you write be about change management and digital transformation, right?

I think that would be a sensible thing for that newsletter to be about when we go out to the first week in December, right? Uh we have transformation, um, yeah, and we have business strategy, there's some of the diversity and stuff. So you can see, you know, if you as you start to pull together your keyword lists and refine them, you can figure out okay, what are the topics from week to week that I might want to be creating content about? Uh when we get towards the end of the year, uh looking again, diversity, equity, and inclusion shoots up to the top at the at the very uh end of the year. I don't know why.

I you know that that I don't know why that would be doing that, but it'd be something that we could investigate and then put that in email. We could um it would require building a system rebuilding the system that does it. Um, but we could experiment in the curated content we share on social media and in the newsletter. Is there a way to identify specific topics that are in the database and say, okay, well, the top five topics that are showing up in you know this following week? Can we find news articles about those topics so that the newsletter is time relevant uh to what people are searching for that week?

That'd be a a fairly heavy lift, but it is theoretically possible. There's a point in there too. One thing that can get missed is a lot of times you'll see cyclical drops and you don't know why. And digging in to find out what's going on during those drops can give you insight into the space and into your customers that you didn't have before. And that's just kind of knowing what goes on.

Because most of the time it's like, oh, yeah, there's the you know, you can see Christmas and you can see the weekends and B2B and stuff like that. But it yeah, you can stumble upon gold if you find you know peaks or valleys that you right now don't understand why they're there. Exactly. Um the other thing is to look at these terms, and then again, if you're doing content curation of any kind, um do you have enough sources that are being pulled in that are covering those topics, right? I know for sure that right now we have the major consulting firms in our content curation uh system, but we don't actively look for change management um in the weekly you know download of articles.

And that's something that we should probably consider adding in is having something that is consulting focus, change management, digital transformation, pull those articles as a separate thing in to our content system so that we can start recommending you know the best articles of each week in those topics, maybe even start a new section in our newsletter. We have right now it's SEO social content marketing and data science. We might have a uh an organizational or a consulting uh I don't know what we would call it, but a section just for that. So again, that would be using we're taking this data from from SEO and saying search is just behavior, search is just people telling us what's on their minds. So if we have the information, we should we can probably use it to say, okay, well, we know this is going to be on your mind.

Let's see what we can do to feed it. I think it makes sense. I'd probably lean towards calling it management consulting. Um, because that kind of covers all of those things like business consulting, change management, all those things. Um, you know, but it's it's interesting to think about it that way, um, because it's not just you know, writing a blog post and sort of using this as the SEO keyword.

Everything that you're doing, in a sense, is content creation. So if you're posting videos on social, that's just creating content, it's just a different kind of content. Even if you're writing just a social post, that's a different kind of content. Your email newsletter is a different kind of content. And so thinking about it in those terms, then this kind of analysis is easier for you to wrap your head around and how you can use it for those other channels because it's all just content creation.

The ads that you're creating for your ad networks, it's just another kind of content. Um, I feel like I'm you know forgetting, you know, a channel or two, but basically just think about everything you're doing as some kind of content creation, whether it's written or audio or it's a video, it's a piece of content that somebody is going to consume. So therefore, you want to be creating content that's timely and topical. Exactly. Um Ashley Zeckman at her uh market props B2B form uh presentation uh at the most recent event was talking about using search data to for influencers to identify the topics you want to approach influencers about.

And um, one of the things I pointed out in the QA was do you do that with a forecast or you just do that with you know uh existing SEO data? She's like, just do it with the existing SEO data. Well, if you have a forecast and you know like when an influencer's contract is supposed to start, this is the tool that you give them to say, okay, uh in Jan in the in January, in the beginning of Q1, uh John Wall, we want you to be talking about our j our diversity, equity, and inclusion efforts, because it's gonna be a popular term that. And so your editorial calendar as an influencer will be influenced by the predictive forecast to say, like these are the topics that we know our audience is going to be caring about. Let's make sure that you are on board with them.

And by the way, if you have nothing to offer on that topic, let's say, okay, great, let's defer your content to like Q2 and bring in somebody else and to patch that hole in Q1. Congratulations, John. You have been promoted to the trust insights influencer. Yeah. Diversity and inclusion.

Uh right up my uh alley. That's the number one. I mean, I wasn't gonna point it out, but you did, so it's okay. Jobs not hiring old white men for is uh the key word campaigns. We are nothing if not self-aware.

Look at April, the first week of April, cost reduction springs to the the number two slot. Why? Well, obviously, you know, that's budgeting season. Uh that's uh the a new fiscal year for a lot of folks and things. So in the management consulting world, that would be a topic that we'd want to have an influencer on board for to say, okay, what can we do about you know zero-based budgeting and things?

There's some other things, Katie, that I thought were very interesting here that could be super compelling for creating dynamic content around. So one of the topics that shows up a lot is Generation Z or Gen Z age range. Well, that content's great, except that it obviously has a tendency to age out um pretty quickly. But if you were to write some code within a web page, uh you can actually make that web page stay fresh and current um programmatically and have an ongoing useful tool to people, like you know, as the calendar changes over and it becomes 2022, you know, these are the the new age ranges for for Gen Z and the generation after them. So even things like that, even like the the software that you create can be influenced by search data.

Hmm. That's an interesting interesting way to think about it. Yeah, I like that. So Chris, you know, you're talking about using the SEO data that we uh researched the past couple of weeks in the context of a predictive calendar. What if people don't, you know, so we talked through like the IBM option, the Google Trends option, you know, can they still use that data if they don't have a way to forecast it forward?

You can again what I was saying is if you download the last five years worth of data and you line them up column by column and excel, you can you know see you do a like a weighted average of you know week by week and figure, okay, on average this week tends to be the highest of the last five years. That'd be a uh a simplified version of this. It's uh not going to be as accurate because there is there's no mathematics of of the forecasting, but that'd be good enough. Um it would be very labor-intensive to do it that way. So you'd probably only want to do it like a handful of keywords because that that is not fun to copy and paste.

Um but that would be an option as well. Um so you don't have to use the forecasting tools. It's better if you do, though. Well, so it sounds like you're saying if you're going to be using that uh SEO research that we've done, you want to have some mechanism of forecasting. You don't just want to look at it and go, okay, I pick this one and I pick this one and I pick this one because I like them, or because they're high difficulty, or no, they're low difficulty, you know, high volume.

I'm just gonna pick that one. Timing of these things also is just as important. Timing time is literally the fourth dimension. Um, and it's the thing that a lot of these tools just don't do. Uh time is not taken into account in a lot of the software that's out there.

A, because it's hard to get a hold of the data, and B because depending on where you are in your content marketing maturity, it's a nice to have, right? You need basic pillar content, like we need to have a change management blog post of some kind, any kind, right? It doesn't and it's one of those things like okay, we know that's a core keyword. Katie, you gotta write this blog post and just put it up. We don't care when it goes up, it just has to exist, right?

Uh right now we have none of that. And so from a maturity perspective on for that, it doesn't exist yet. I'm just using this as an example. No, I know, but I was gonna say actually, we do have one post a week going out right now, right? Um on change management, yep.

Cool. Um and so that term, that keyword would be uh locked in there, and then once you've taken care of the basics, you've got the top 20 or 30 you know keywords that are on the list all in play and there's content out of them, then you take it to the next level and say, okay, now our promotion calendar or our new content creation calendar should be time so that we have this collinearity of time um cooccurrence with the marketing. So that that's why I would say it's not an immediate thing if you just don't have the stuff, period. Right. But it is one of those things from a promotions perspective that will help make your promotions work better.

Well, and that's when you start to dip into uh things like the transmedia framework, where if you're low on resources, uh, but you know this is important for you to get out there, then you start to think about how you can scale your content. And so you do something like this where you talk about it like on a live stream where you're recording the video and the audio, and then you start to pull apart the audio, transcribe it, that's your blog content, your video, smaller snippets, those go on social, et cetera, et cetera. Exactly. So that's um that's how you would make use of the data, even if you don't necessarily have uh you know the forecast and stuff again. The the correlation, the you know, the time-based uh cross-correlation functions, that's something that's built into Excel.

So if you have the data, you can do those time-based cross-correlation functions and and understand what words and phrases co-occur in time, and that alone may be enough to give you uh a content calendar, right? If you've done your five-year average and you've got these terms, you could get a a some interesting insights that way. Well, John, I assigned that to you this week. So I expect to see that on my desk tomorrow morning. All right, yes, all the insights we have.

So I think for this week, that's it. I mean, there's always more that we can talk about. There's lots more things that you can do with that information. But you know, to get started, you know, we've again two weeks ago we talked about doing the basics of uh keyword research and how to approach that. Last week we talked about doing that research on your competitors.

And now this week we've talked about how to use that data, but really the most important takeaway, if nothing else, is that timing of using that data matters because if you're creating content and it just kind of goes into the ether, nobody's gonna see it. But if you're creating it on that, even if it's you know a very lightweight forecast, then you're creating content when people care about it. So that is the most important way to use that information. Exactly. Um so yeah, I'd say that does it for I guess this three-part mini series.

Uh, if there's uh a new mini series that you want us to tackle, let us know. You can uh pop by the Slack group. We'll have a link for that in in just a bit. Um, but uh I think that's it for this week. That's it.

All right, thanks for tuning in, folks. We'll talk to you next time. Thanks for watching today. Be sure to subscribe to our show wherever you're watching it. For more resources and to learn more, check out the Trust Insights Podcast at TrustInsights.ai slash TI podcast and our weekly email newsletter at trustinsights.ai slash newsletter.

Got questions about what you saw in today's episode? Join our free analytics for marketers Slack group at trustinsights.ai slash analytics for marketers. See you next time.