Create and configure freeform tables in Analysis Workspace in order to analyze internal search functionality on your site.
Hey, everybody, this is Doug. In this video, I want to talk to you about configuring internal site search reports in Analysis workspace. I thought I’d actually start here in the business requirements document, this BRD, to kind of show some of the things that we implemented in this use case example. So, when you’re talking about internal search, of course, the search terms are the most important. So, we’ve got that. And we’ve put that into an eVar. And we also have the number of searches. So, whenever there’s a search, we are triggering event4. And we are also capturing the number of search results returned in eVar11 so that we can see if there were, for example, searches that happened with zero results. That can be very important to know, of course, so that we can make sure that people can find the things that they’re looking for, and that we can provide things people are looking for. Actually, jumping down two spots here, we have track search filters. So, we are tracking our search filters in a List Var, because multiple filters are being added at once. And then we also decided to do unsuccessful searches as well and put that in event5. So, whenever people do have a search that there were no results for, we trigger event5. So, let’s take a look at a project that I started with some of this stuff in it. So here we are in Analysis workspace. You can see that I’m using this Luma Web SDK Demo. So, this data was brought in through a Web SDK implementation. But in reality, it doesn’t matter as far as, you know, once you get here, it doesn’t matter whether you’re using an app measurement or a Web SDK implementation, you would still have the same data in here. So, for this video, I’ve really focused on freeform tables. As you can see, I’ve got a bunch of tables here. And then from here, you know, you could add additional visualizations. For example, if you want some line graphs, or if you want some, you know, donut visualizations, which sound delicious, or any of these others that can be helpful to you to be able to visualize this data. So, I go over here to the components. And of course, I can search for internal search.
And I’ve got my internal search filters there. But I’m actually looking for a search term right there, so let me just go search term instead, search term. And you can see that’s in eVar10, just like we put it in there. And then I just dragged that into this freeform table. Now I also then dragged over site searches. Let me just kind of get rid of this. You can see that that’s actually right here. And I dragged that over instead of the automatic instances that are put in there, and I replaced that with site searches. So now we can see the things that people are searching on on our site. And of course, what people are looking for the most. Of course, I just ran this a few times for this demo. Now if we scroll down, I also then dragged in search results. Since that was in eVar11, I dragged that into an additional table here. And now we can see, you know, how many times people get zero results, four results, five results, etc. How many results people are getting back, and how many times they get back that result. Now maybe it doesn’t really matter if there’s five or six or whether there’s 10 or 15, but zero is going to matter a lot, right? We’re going to want to know if there are searches that happen that return zero results. And so that’s one of the main reasons why we captured the search results. And now I can take this search term, and I can go, well, what are they actually looking for when they get zero results? And I can drag it down here and then just do a breakdown. And the ever popular asdf, unfortunately, is not to be found on our site as well as apparently, apparently, I’m all against socks on this site as well, because I don’t sell socks. So, people looking for tube socks or sweat socks are out of luck. So anyway, just an easy way to do that. Also, even maybe if they only get one result, maybe you’re going to want to know what they’re looking for when they get only a few results. So just something you can do there. Now in addition to viewing this data this way, we can kind of turn it around and go up here and say, here are the top searches, let me actually break those down by number of search results. So maybe I take the top five searches, and then I break those down by the number of search results.
And then I can see how many search results for the things that people are searching for the most on my site. So that can be very helpful way to do this as well. Really just two ways to look at the same data between those two tables there. And then last but not least, I have a search filters table down here. And so, I have dragged that in as well. That again is in a list variable. Now when this comes in, it’s actually things like, you know, a, b, c, d. But because it’s in a list variable, it breaks it out into individual lines. And of course, if one hit came in with b, c, it would add one to the b column, and it would add one to the c column. And so again, this is why List Vars are so powerful, is because you can have a delimited list of values with multiple values come in, and it can split them out to different lines like this so that you can see which ones were chosen the most, again, even when there’s multiple items coming in. So, we’ve had searches, of course, with no filters selected here. And then of course, we can see these other filters that have been put there. And so, I just put, you know, the letters there. But you get the idea. You could have any kind of filter there, whether it’s a price range, whether it is categories or brands or locations or whatever kind of search filters you might have on your site. Then just like above, we can also break down the search terms by those filters. So, when people add filter b, what are they actually searching for again with the socks? So in any case, this hopefully gives you an idea of some of the cool data that is available to you when you’re doing internal search tracking on your site. Good luck. -