Hi, everyone, and welcome back to Marketplace Masters, the podcast where we bring together marketplace experts to talk about what really works.
Paul Sonneveld
Today, we're going to dive deep into a goldmine of data that many brands, agencies, and vendors still only scratch the surface off, namely Amazon Brand Analytics. If you've ever wondered how to turn the search query performance report or the catalogue search performance report into actionable strategies, this episode is for you. I'm thrilled to be joined once again by Jimmy Barber, the strategy director at the Bluebird Group.
Jimmy has worked across dozens of top-tier brands and knows how to pull insights that make a real difference in campaign performance and retail strategy. We're going beyond dashboards today. Let's talk about what these reports actually mean, how to interpret them, and most importantly, how to act on them. Jimmy, welcome back to the show. It is absolutely great to have you here again.
Jimmy Barber
Great. Great to be here as always, Paul. Great to be here.
Paul Sonneveld
I know we will get into some real, practical stuff. I know you've got lots of screenshots and practical tips lined up, so I can't wait to get into this session and probably learn a handful of things myself.
Jimmy Barber
We'll see.
Paul Sonneveld
Let's set the scene. Jimmy, for those that are newer to brand analytics, maybe you can start by giving a quick overview of the key tools we'll be focusing on today. You know, SQP, CSP, et cetera.
Jimmy Barber
Yep. Yeah, I think there's, I would say, a couple of precursors. One, brand analytics exists for, you know, sellers and or technically can be vendors in some ways. We can unpack that have a brand registry account with a registered trademark or brand name that lives within it. Once those kind of things are checked, you get access to what's a growing actually list of reports now. I think we've got customer journey, customer loyalty, which within search analytics, we've got catalogue query and then top search terms. And then there's some consumer behaviour analytics in there as well now.
And so I think Amazon has gone from what was originally maybe two or three reports to now almost 10. And it's a really powerful data set that just for having a USPTO filed trademark, if you're a US seller, it unlocks a huge list of specifically keyword-level data that is just really hard to get otherwise. And at the end of the day, I think keywords are what drive the Amazon ecosystem. It's a search-driven platform. And so being able to unpack and understand searches actually is why I think so many things in brand analytics is pretty impactful.
Paul Sonneveld
Yeah, that's a great summary. I mean, really important point around brand registry, making sure because a lot of, there might be some people in the audience going, oh, where is this stuff? I can't see it. That's because maybe your brand is not registered. It is a prerequisite.
Jimmy Barber
Maybe a good thing to note, you can get delegated access. So even if you're a reseller, distributor, you can get delegated access from whoever owns the brand registry account to access brand analytics, which is sometimes really helpful depending on your relationship. And then even if you're a vendor, can technically still access via a seller account. Obviously, you know, you can get into that whole discussion of the hybrid model seller vendor. But it is really impactful data that you can still see because you are a brand registered, you know, brand.
Paul Sonneveld
Yeah, I think you were touching on some of the great points you were making during our last episode together around, you know, value of having a seller account as a vendor just for access to these types of reports. I remember that as a big takeaway. So look, lots of data, right? Like someone like me, I love data, right? I can easily burn a day going through it and getting myself all excited.
You know, probably sort of not seeing the forest through the trees anymore. A little bit of strategic guidance probably is helpful here. So from your perspective, I mean, the blue bird group, you guys manage a lot of big brands. When you take on a new brand and you're starting to look at this. Data set, where do you typically start when diving into these reports, waiting for a new client or product category, what's a good way into this?
Jimmy Barber
Yeah, I think maybe one of the easiest places to start is search career performance. And what that report I think can really do for you is two things. One, it'll tell you what's powering the business. And I'll actually share, let me share kind of the initial look of what you're going to see in any kind of search career performance dashboard.
When you log in, this is for an account that I own, you're going to see basically what Amazon determines is a search query score. What that is, is a combination of search query volume, your branded share of it. And we kind of distill that into what is impact to your business, basically, aka typically how much revenue is driven at the keyword level. But but first and foremost, you're gonna like immediately see if it's a really heavy branded experience, you're going to see branded search terms all over the place. If it's a really big category niche, you know, for this instance, this is filtered on a SKU that's heavily influenced in the pie weight category. And so you see, every one of these search terms is driving some form or fashion of a pie weight.
But I think when we audit businesses, this is one of the most impactful reports because you immediately see what are the keywords that are driving the volume on the business. And then particularly if you plot it over time, which is a little bit trickier to get the data shown at the weekly level in the UI UX, obviously the API is now available. If you start to plot some of this data over time, you can really start to see what keywords are really driving the business over time. And you can almost instantly distill how healthy is the business from a keyword perspective because sometimes you'll log into an account and it'll have 10 keywords and that's it. That's driving the business. Ideally, you're looking for a pretty long tail impact of different keywords, different searches that are going to be driving volume on the business.
Paul Sonneveld
Very interesting there. So, when you're looking at a screen like that, and I'm probably going a little bit deep for this part of the conversation, but why not? What are the, if you're looking for opportunities, like what are those gold nuggets that you're looking for. What are you hoping to see? Particularly, I guess there's a difference between what you're hoping to see after you've managed the client for two years, you're probably not hoping to see some of this. But when you take on a new client, What are those gold nuggets you're looking for?
Jimmy Barber
Yeah. I think one, maybe a couple of nuances of the report that are always helpful to know. It's a 24-hour attribution window. I think for most Amazon businesses, that's a pretty broad spectrum of your revenue. If for whatever reason, your business has a pretty long decision cycle and 24 hours isn't the purchase window, it may not be the most helpful report.
And then it includes organic and paid. And so if your sponsored product activities are going on alongside a lot of the organic activities, you can theoretically match up a direct single day attribution window in a search term report against your search query performance report and back out directly what is organic versus paid at the keyword level to know how much are you influencing impact via paid activity at the keyword level. So I think that's one kind of maybe overly granular way in which you can use the two reports together.
The other really interesting one, which there's a long debate always in the Amazon world of how do I spend on, quote unquote, branded terms. I feel like search query performance really gives you the power to make a guided decision on how and when do we invest advertising against our branded terms. One, you can really point out branded terms really easily. You can clearly see a divide.
If you were to plot all of those keywords in your purchase share at the keyword level, you'll see these little plateaus that exist. And you'll see a grouping of keywords that typically have a 60% plus purchase share. Those almost always are a brand-guided keyword. You'll typically see a middle layer of a plateau that's anywhere from 5% to 30% purchase share. Those are typically your generic or category-based keywords. Then you'll see the really little plateau, which is typically your competitor terms in some form or fashion that you're showing up in. It depends on how good your product is, but you're showing up in a little bit of a different layer. Sometimes that's 0% to 2% purchase share on a competitor search term.
But I think those are real tactical ways you can start to piece together the customer journeys that exist to find your product. And then once you've done that, you can really start to get granular in how you talk to each one of those customers. And just whether it's through ads, whether it's through your content, you can do a better job of product innovation, targeting, you name it.
Paul Sonneveld
Yeah. Actually, if you don't mind, I'd love to go a bit deeper on that. Like, let's take that branded search or branded keywords as an example, right? In the Search Query Performance Report, you know, you do get a bit of a funnel, right? Like share of impressions, share of click-through, views, add to cart, etc, etc. I mean, roughly rule of thumb type stuff, right? Can you walk us through maybe like each of those stages of funnel and from a brand point of view, like what does good look like? And if you're below, like what are some of the things that you guys will be doing at Bluebird Group to try and get you back to that benchmark?
Jimmy Barber
Yeah. So you've got first columns, impressions. You've got clicks, which is where you can really back into your click-through rate. That's one thing I'll touch on. You've got cart ads, which is a little bit of an interesting one. Amazon doesn't really disclose or historically hasn't disclosed cart ads. It's maybe one of those that's the least actionable of insight. And then you've got it rounded out with purchases so you can see your direct conversion.
One, if you plot that ratio, so your impression share, your click share, your card ad share, and your purchase share, depending on those four data points and the line that they form, you can really start to see what I was talking about of what type of keyword is this. And if the line is straight down, it's probably a generic keyword. If it's straight up, it's probably a branded keyword. But so I think whenever we look at a report like that, it's a lot of data, one.
And so I think you've got to come into it with a little bit of hypothesis or thesis of what am I trying to discover and then what am I trying to go test. But there's also, I think, some really good nuggets around, you've got data on same day shipping, which isn't really available anywhere else. You've got data available at two-day shipping, you name it. That's not something that you've really been able to access in any given point.
But at a fundamental basis, calculating your purchase share at the keyword level, and then you compare that to keywords that you're advertising on, I think is a really baseline way for you to quickly see, okay, what keywords do I have a really strong impression share, click share, purchase share on?
May not have a ton of search volume, but if that relationship is pretty positive, it's a great keyword to add to a manual campaign and kind of go after from a harvesting standpoint. Because you've already seen that the search data that exists for that search term is in the same trend of what your shoppers would likely want. And it's a great way to get, you know, obviously, hopefully, a keyword into your campaign structure that, you know, gives you a pretty high ROAS or whatever you're looking for from a KPI perspective.
Paul Sonneveld
Yeah, very interesting. I mean, I think there's so many different use cases and examples you can pull out of here. The link between the search query performance report and the advertising, or certainly the keyword advertising piece, is super valuable. I haven't seen Amazon, and it's been a while since I've been in the interface, but right now, I believe Amazon doesn't combine those two data sets together in the search query performance report. Have I got that right, or have I missed an update there?
Jimmy Barber
So search query performance will show all search data, which they define as both combined organic and paid. What you theoretically though can do is you can pull an advertising report from the advertising console and kind of back into what your breakouts are. Does that make sense?
Paul Sonneveld
Yeah. No, no, I understand. So you could look at something like, well, let's say you've got really low impression share or click through, probably yeah, impression share on a particular keyword that you do want to rank. I guess you've got two levers. Either you can try and start bidding on that keyword in a certain way and drive it through paid or probably play the longer game around organic SEO and ranking and those sorts of things. But paid probably is quite a nice short-term lever you can pull and be very targeted and specific around what is this keyword or what is the ASIN and all of that. Absolutely makes sense.
Jimmy Barber
It's also good to see like, you know, impressions is when the search occurred, did the product show up? Not was the product seen, but did the product show up? And so it's a good way to kind of analyse, especially if you look at the relationship of impressions to clicks, it's a good way for you to understand, you know, in some degree, like, where am I ranking?
Because there's a good instance if your click share is really low, but your impression share is actually decent. It's like, okay, you're indexing on the keyword pretty well, but you're on row 45 or wherever it may be. And that's why your click share is so low. And so there's some good ways to kind of look at it to see quickly, okay, where does this keyword fall? And then what lever can I kind of pull? Is it how I need to rewrite my listings? Is it how I need to leverage my paid strategy?
Paul Sonneveld
Got it. I didn't realise that. So what you're saying, you could be well below the default, talking like old school SEO language. You could be well below the fold and therefore unlikely to be actually seen. You're sort of generated as part of the search result return set. Your click-throughs are low. Actually, yeah. Sorry. I always immediately go to, oh, your hero image is off. You know, your click-through is wrong. You must fix up your hero image. But you're saying, well, actually, maybe you should think sort of reverse engineer and think about how many people do you think saw your image in the first place? How many were actually given the decision to click or not?
Jimmy Barber
Yeah. Which is very interesting. I mean, campaign boosts in the ad console is obviously top of search boost is a great way to test the theory. Because I'm convinced a lot of products actually have pretty good products, but for lack of a lot of reasons, they just never get the time of day on the front page. And I mean, it's really clear. A good example, let me share this one, is you can see, and I don't know how many folks, This is a little bit of a weirder part of search query performance and maybe too granular for some folks.
But if you filter actually within search query performance, it defaults to the branded view, which is your total catalogue rolled up. If you'll actually pivot to the ASIN level view and then filter down to a specific ASIN, you can click into the search term. So this is looking at, for your brand, search term level. At the keyword level for the ASIN, you can see then the top 10 of ASINs that show up for that search term.
So search queries, pyweights, this was looking at for a specific ASIN, but Amazon then shows you the top impression shares, the top 10 impression shares for that actual query. And what you can see is impression share is pretty much like a linear relationship, right? Like someone searches pyweights and they gain impressions pretty linearly. But clicks, you can see there's like two big anomalies. And guess where these two are in organic ranking? Number one and number two.
And so you can almost immediately back into like, okay, what is organic? Forget impression share, but like what is organic ranking do? And everyone always like theorizes the impact of being organic rank number one or organic rank number two. But like the data is right here, it's telling you 19% of the time you're getting the click if you're organic rank number one. And so, I mean, it's become a pay-to-play game in a lot of ways, but it's also like there is real value in organic rank and you just can't argue that still.
Paul Sonneveld
Yeah. And the reason they rank well organically? Is this just something you've spot checked yourself or is there another metric?
Jimmy Barber
Yeah. We use separate tracking to track organic rank at the ASIN level and it validates it to you.
Paul Sonneveld
Yeah. I mean, look, I'm assuming that the ones here are, they're sort of talking click share of, I don't know, 5%. It's a bit harder to read, but, or low single digits. You know, the difference is five to seven X between sort of being ranked and being visible versus being hidden. And I'm assuming these are still within the top 20 listings that are being shown, right? So they're not that far down. They're not all the way at the bottom. They're probably just slightly hidden.
Jimmy Barber
Just slightly hidden. One of the ones as we were going through this is one of the more interesting stats that Amazon gives, but it was one of these I was like, I'll put this back to Paul to see. I haven't quite figured out how to put this to action because they give you at the search query level, what's the actual click rate on the query? And so they're saying there's 22,000 clicks whenever there's 37,000 searches, which gives it a 59% click rate. And that's in total.
We went through and we looked at a bunch of different search terms. Some search terms like Nike or whatever it may be, have a really different click rate just in general. Then when you get into hyper-niche terms, that click rate can be as low as 30% to 40%. They give you the data point, but I have not figured out a good how do we use the average click rate on a search query as a guiding principle or advertising strategy, SEO strategy? It's one data point that's kind of confounded me and the team of like, how do we put that data to action? Because it's kind of an interesting data point to know if it's a click-heavy search term or not, basically. But I haven't quite figured a solve for that one.
Paul Sonneveld
Yeah, it's a really interesting question. I suspect it may have something to do with the intentionality of the purchase, right? So if I'm looking for Nike or something, maybe I don't know exactly what Nike shoe I'm looking for, what size, and I'm just going to check Amazon to see if they've got a better price or faster delivery.
So, you know, my intent to purchase is very, very high, but maybe on the Pio Edge, maybe it's more like I'm just trying to wrap my head around, what do these things do? I'm sort of exploring a little bit. And maybe I'm not completely sure what I'm after. So I suspect it's a little bit around that, but maybe in that letter scenario, then there's more opportunity for, you know, kind of retargeting and following up and trying to stay connected with that, with that particular customer, right. Or, Anyway, I'm speculating.
Jimmy Barber
Yeah. Yeah. I thought about one of the things we always struggle with, I think, is when we get into like, you know, take Prime Day, right? It's like, Hey, if we had an extra $10,000, what could we go generate? And I'm like, I, you know, maybe we could run some better modelling to predict click-through rate, your average click-through rate, what we know to be okay.
We've never gone after this keyword, but you know, we could back into some of what we think we could do. But yeah, I thought that was a fun one that we noticed. And it's like, no one ever talks about the click rate on a given term and just what to expect, how many people actually tap on the screen when they type in the search bar. And that varies pretty wildly depending on the search term.
Paul Sonneveld
Yeah, yeah. Well, look, I think the other argument you could spin is maybe certain niche categories are not as well run. Maybe the results that appear are not as intuitive or engaging. Or it's just like a mismatch between the intent of the search and the actual results, right? But you know, if that's the case, that's fantastic because that creates a lot of opportunity to improve the game. So for sure, yeah. So one of the, I just want to sort of clarify one thing. We talked about the search query performance report and then we talked about the catalogue, was it the catalogue performance report? Is that right?
Jimmy Barber
Yeah, there's a catalogue search report. Search report, yeah. They call it search catalogue performance, and then search query performance is kind of the keyword level.
Paul Sonneveld
Just another two, three-letter acronyms to add to your 3,000 acronym glossary that you've already built, right?
Jimmy Barber
We've got ChatGPT now. We've got all the acronyms all buttoned up. So, it’s all easy.
Paul Sonneveld
It's easy. Yeah, easy. So what's the difference between the two?
Jimmy Barber
Yeah, so search catalogue rolls up your total brand level of basically how many sales come from search. It actually, when they first rolled out the dashboard, so it doesn't give you keyword level, which is sometimes actually nicer because you can kind of quickly see how much is search driving, but you also get some good data points around click-through and then conversion rate in particular at your brand level.
One of the nice reports though that they just rolled out that pairs actually, when they first rolled out the search catalogue performance, I think a lot of people glazed over it and were like, this is interesting to know how many dollars come from search because theoretically, you can look at your total POS or revenue on any given week and then back out the search catalogue performance for you to then back into how much of my revenue came from the search bar and how much my revenue came from more or less all else.
So direct links in between word of mouth, text messages, wherever it may be. If someone didn't have to go to the search bar, then it's not going to get captured in the report. They did then just, I guess almost a year ago, I'm not sure, I don't remember the timelines. They rolled out the customer journey analytics, which is a little bit of a pivot of the search catalogue performance report, which shows you then the individual journeys customers go on as they unpack your catalogue.
And I think that report is really then where you can really put to life like, okay, how much of my revenue comes from the search bar? For some brands, we've seen that be staggeringly low. And it's like most of their traffic doesn't even come from search. They are just found through direct links, affiliate links, random blog posts, whatever it may be, through their D2C site, it's all direct traffic to their PDPs. For other brands, it's totally flipped. And you see they rely 100% for their sales on the Amazon search bar. I think there's pros and cons to each scenario.
I would say it's always nice to have a little bit of a balance. The customer journey then kind of gives you the double click into the actual search catalogue performance to know, okay, where are they coming from? Where are they hitting in my catalogue? Is it my PDP? Is it my brand store? And it actually kind of gives you the ability to actually draw maybe a little bit more action out of it. Whereas before it was like a good, good to know, but I'm not really sure how I influence anything.
Paul Sonneveld
It certainly sounds like the search query performance report that has that keyword performance at a product level, that gives you some real tangible, you know, I guess, opportunities or strategies that you can execute on. Whereas the catalogue search performance, search catalogue report, so I'm getting, getting my head up here, but that's sort of like, yeah, that's nice. But, you know, it's not granting enough to actually action, whereas that customer journey sounds like it's trying to bridge that gap somewhat.
Jimmy Barber
Yeah, I'll share this, this was a, I think a good kind of example of how you know we've seen this kind of come to life of like you've got. This is the customer journey analytics and it shows you basically count at each stage. But so if theoretically, we're looking at total search performance, one, for this brand in particular, there's literally no branded search.
And so when people discover, they discover the brand entirely through category search terms. What's really interesting though, is the detail page views are insane compared to the category searches. So what this immediately tells you is category search doesn't really fuel your view detail pages. Your view detail pages are existing outside of the search bar. It just so happens that this brand is really heavily affiliate and blog post-based.
All the traffic that comes on to this brand comes from direct links somewhere outside on the Internet. It doesn't really come from the search bar. That then gives you, how do we develop strategies around? We don't really have a lot of placement in search and we definitely don't have a lot of branded search. How do we better, one, maximise the fact that people are coming on to a PDP? How do we get them to explore our catalogue? Because it's a PDP, it's not a catalogue.
But then two, specifically branded search versus category search, how do we start to benchmark this as a KPI and drive branded searches or drive our percent of share of category search to detail page views? And that's a great way where you can really evaluate your overall brand awareness on Amazon, which is how much traffic is actually coming from brand searches versus direct links or wherever else it may come from.
Paul Sonneveld
Yeah, that is super interesting. As I'm even looking at this, I'm going, well, that seems like a little bit of an outlier there, but very interesting nonetheless. I mean, obviously, there's driving a lot through affiliate. I mean, the drop off is really, really high, which you'd expect because a lot of these visitors probably coming in really cold without necessarily like having that, you know, in-depth purchase intent. But you know, a great strategy, actually, and a great example that you can build an Amazon business that relies on a completely different funnel than what you're used to.
And I think the idea is to have multiple of these on the go. Yeah, super interesting. We're just looking at the clock here, as per usual. I'm always over time, so we need to sort of steer towards wrapping up here. What are the biggest pitfalls that you see brands or other agencies or what are the pitfalls you guys have made at the start using these reports? Any must avoids? What would be on the list?
Jimmy Barber
Data just as a whole inundation, I think it's really easier to get super, super granular in these reports. And sometimes that's beneficial. Other times you end up spinning wheels and not actually driving a lot of action. I think the other pitfall is that I see a lot of correlation into paid. And it's like, okay, great. Here's a bunch of search terms that showed up in search query performance. Are they in a paid strategy campaign? No. Okay, add them. And I think there's good use cases. But I am always thinking of how do I do things for free in the Amazon ecosystem versus having to pay for it.
And so I think there's a ton of strategy implementation that can happen from search query and overall brand analytics that it's like, this can really help you guide and understand your customer and who your shopper is. And then specifically, what shoppers can we go reach into. And I think one of the biggest pitfalls is, I think as the report has kind of become a little bit more seasoned, it's influencing a lot of our paid strategies, which is good, but it's not always influencing like our broader brand strategies. And I mean, you can really like literally set an index on how big or how valuable is your brand based off of this data set. And I think that that is a metric that you should be tracking if you're at all a brand builder in the space, is what is the brand efficacy of my brand on Amazon? And I think those are areas where people are using the report to do one thing, but it can do a lot more than just that.
Paul Sonneveld
Very good point. Very good point. We do have one question here. I usually go to a Q&A and I don't have much time for it, but I do want to do the right thing by Ken here. Ken's asked a really interesting question. So before we wrap up, I'm going to see if we can help him out here. So Ken's asking, and thanks for your question, Ken, really appreciate it. How accurate do you think Amazon's category versus brand search metric is? Any way to audit what they think is brand versus category? I mean, does brand just include the brand name, not a popular product's name? You know, how do they distinguish there? And what's your view on that?
Jimmy Barber
Yeah. It's a good nuanced question, I think, Ken. Technically, we dug into this because we have a couple different products that have multifaceted brand names that have three or four words in it. In a lot of the reports when they're talking, if Amazon is referencing branded in any of their reports, they're literally looking at direct text match. They don't always catch misspellings. Obviously, your brand name, everyone has maybe a unique brand name and so misspellings can be one thing. You can end up with brand name on the back end of the search wherever it may be.
Typically, what we've seen is they're almost always matching your direct text in brand registry. And so if that doesn't match, it's not going to show up as branded, quote unquote, for Amazon's lingo. I think it's one of the nuances of the reports. It's why search query performance, I think, is really helpful because you get to see the actual queries. I'm always a little bit weary with the customer loyalty dashboard. It's like, that's a great thing to help influence my opinion and decisions, but it shouldn't dictate them. You've got to bring in your knowledge of the catalogue, the business, the category, and then pair that, I think, with what that report's telling you to actually get to the truth.
But from what we've seen is it's not a super sophisticated way in which Amazon's identifying branded terms. I almost encourage folks to keep a list of look at your search query performance report at a bare minimum filter for the queries that are greater than 50 percent purchase shares and look at them and understand those are likely your branded searches. And then use that to kind of start to build some of your data sets off of. That's a good question though. I'm always weary of having Amazon's data dictate my plans. So it's like, I'll use it as a data point, but I won't let it be the only data point.
Paul Sonneveld
Yeah. There's always, sometimes you have to consider Amazon's agenda in a lot of this data as well. You know, if I was Amazon, what would I like to tell you? And then sort of calibrate it with a few other things.
Jimmy Barber
Yeah. I mean, it's also, you know, they're building a machine for millions of sellers. It's like, it can only be so sophisticated at the customisation level.
Paul Sonneveld
Yeah. Really good point. Thanks for your question, Ken. That was great. All right. We are out of time, Jimmy. We're going to have to wrap it up today.
Jimmy Barber
Love it.
Paul Sonneveld
But huge thank you to you. I love every time you come on, you bring your practical wisdom. We get into the numbers. Really, really great. Thank you so much. If there's anyone listening who wants to get in touch with you, maybe continue the conversation. What's the best way to get ahold of you?
Jimmy Barber
Yeah, shoot me a message on LinkedIn. I think I'm always active, always there. If there's something I can help with, happy to chat.
Paul Sonneveld
Excellent. Thanks, Jimmy. Take care.
Jimmy Barber
Awesome. See you, Paul.
Paul Sonneveld
All right, everyone, that is it for today's episode of Marketplace Master. If you're looking to elevate how your agency drives strategy from data, today's conversation should give you plenty to run with. Don't forget to subscribe, share this with your team, and join us next time for more ideas on how you can master the Amazon Marketplace. I'm Paul Sonneveld. Thank you so much for listening. Take care.