[Music] hi everyone welcome to my session on ai powered recommenders i am garrett schweigler the program manager for digital commerce here at lucid works and i really wish we could all be doing this in person together but considering the circumstances this will do i there's there's really nothing more that i enjoy and love than uh you know meeting all of you and having these conversations uh in person so that's it's different times but like many of you my wife and i are now full-time work from home and so over the last year we haven't really you know gone on any trips it's uh it's been mostly on lockdown and so with that we've had some opportunities to do some pretty cool things one of which this office that i'm in we built this so that was quite a fun project and so this is the one that i work out of and then my wife she's in one of the rooms in the house and so we spent some time decorating that and so when i say decorate what i really mean is we put a dog yoga calendar up in her office and so with shortbread and tinsel you know looking at these guys every day you can only imagine after a few months of sheltering in place what happened next so we got a dog and and and really so much more than a dog we got colin the dog so he's uh maybe around five five or six years old he's an airedale terrier mix and uh those dogs tend to be pretty hearty they're fun and uh he's been quite quite an addition to the family fits right in and really fits in so well because not only am i do i have a sensitivity to gluten he does as well which is quite bizarre so we knew it was a match made so like any new pet owner and first dog owner i went to where i grew up going for goldfish right so i created an account and and just started started on the shopping we had to get food we had to get the toys all the good stuff so we go to the dog food and uh luckily there's a filter that we can filter this by grain free i scroll down and i ended up clicking on this one and after reading through the details on the sides of the package it looks good i was gluten free a little pricey this is probably more expensive than some of the food i eat um and so wanted to explore a little bit more uh what other options there are and being in commerce i feel like i know how to get around sites pretty efficiently and so i head down to the recommendations which are you know generally below some of the long description and um and started to take a look at those to see what my options are right that should be able to help me get around pretty quick and so i started reviewing these and i could tell that so the first one uh is totally irrelevant the melting ice that's not necessary in the part of the world that i live in and then there's a couple options that are all from the same manufacturer that i'm looking at however there's an option there that has whole grain and that is not what i'm looking for so also totally irrelevant so two of the five products here uh are just not not giving this company an opportunity uh to have me convert and add to cart with these so at the end of the day i ended up buying uh the dog food that for the pdp that i landed on and and all was good after i bought it i go to facebook right i want to see what my wife and what she's posted about colin and i lately colin and i are the models and she's a photographer and uh right right away like after a couple scrolls uh petco they're there again so great stitching they know who i am and where i'm going also pulling in the honest kitchen knowing what i was shopping for so that was a brand that i purchased and this time there are two of the four products listed are whole grain so uh you know now we're down to 50 of relevant products in the recommender lastly i get the email thanking me for joining um and at the bottom of this there's another series of recommenders of which many of these are you know or both of these rather are off-base too right colin's not a large dog and he doesn't have dry skin so i'm just highlighting here the opportunity uh that you know this occurs across so many websites and we have today the technology the sophistication the machine learning and the ai and the human intel to build better experiences quite easily so let's jump in for one more example many of you probably recognize this brand just by this statement and um i i love these guys lots of shoes from them especially through this work from home era which is why i came back to the site was to get some fresh slippers so search for slippers filter by men's and i see here the berkeley ones pretty stylish for me and click into those to go to the pdp and as you can see here when you look at the recommenders at the bottom of this page all four of these are women's shoes right and i explicitly provided the filter of men's and i'm on a manchu pdp and so just you know another reminder of the indicators that we should be picking up on the signals that shoppers are expressing and how we should be able to leverage those so aside from being frustrating and introducing friction into the experience what's what's the big deal why are we here for this so you don't want to raise eyebrows with your recommenders you want to raise aov and so this is a study in which they took a look at the engagement of recommenders and the impact that they had with average order value with uh no engagement in the recommender starting around 45 and then if you go to just one engagement one click with a recommender the lift is almost 370 right so almost 4x and then it continues to climb until about five clicks for this particular study the significance here is not the nominal like the dollar value specifically so going from forty dollars to two hundred dollars right so it's it's really the relative increase and so plug in your aov number is there and you'll see it as well so what else how about conversion nearly the same story conversion can nearly triple once shoppers start engaging with recommenders so what 200 yes almost three times uh for conversion and you know i guess what's the big deal right well recommenders can not only reduce friction but they can make you boat loads of money when it's done well and so that's what i want to focus on today right everybody has recommenders but clearly there's an opportunity to do this better and clearly there's an opportunity to make more money so let's dive in and nerd out a little bit about these recommenders now when we take a look at this is this is kind of going to be the cadence that we go through here the there's not really one model or approach or application that's that's going to be the answer and so if i look at these and i see that there's uh training and post-processing under human and ai synergies that can really also apply to we can post process the popular and trending models or the product content based models and and we can we can merchandise all of these right but for the context and for our conversation what i want to do is just kind of compartmentalize these and we can run through them and then show a couple of use cases as well so self-driving mode this is i can't i thought this was relatively appropriate for uh how we're going to talk about recommenders right this is this approach is really it's taking into account all kinds of pieces of content and context as well right not only is the the algorithm looking forward and down the road it's looking in the rear view mirrors looking all around it understands how fast you're going right and all these different components get plugged in to to really provide an autonomous experience where it's learning and the output of it is going to be predictable all right so let's dive into a couple of these self-driving mode approaches so we have content-based recommenders which is which is really a neat way to train a model without signals so we consider this a good cold start option in which you're essentially feeding the model your catalog and all of the uh what we're referring to is content are like the attributes about the products and from that it'll it'll determine the relationships amongst the products for a good item to item recommender to start and so this is a good way to get up and running pretty quickly again if you don't have any signals the next one here we can power a trending now popular items popular queries with our trending algorithms approach and these can also be for and all of these really uh can be used for both products and content and in this context when i refer to content what i mean is branded content and other documents of interest that you might want to surface to the customer in their product discovery experience many of these uh can really be applied deployed without any data science background right so we've done all the leg work to get these into a position where most teams can just jump in and train the model and and be up and running pretty quickly however if you have a data science team or resources capable of building their own models we have a data science toolkit and we frame this as bring your own model byom where teams can can build and then train models and integrate them into index and query pipelines with with the machine learning from your own team and so if you are saying well yeah there's not really a solution here or we have resources that we really want to get creative with and build models and experiences that are specific for our business this is the perfect way the perfect tool to do that so next up uh we have human and ai synergies and you know this uh i love this picture for a couple of reasons one the girl on the right that's probably the face many of us make when we're scrolling down down the page and preparing for what may or may not be a relevant recommender on our own site uh but also these toys if we if we think back to uh playing around with them and how uh they can walk uh for a certain distance and then you know they need to be picked up repositioned and there's this human interaction with it and it's it's a pretty good synergy so um let's dive into into this next topic so when we think about uh some of the algorithms that we have listed here i see right als and bpr those tend to be pretty common and i think most people are familiar and aware of how those ones work but the one i want to focus on briefly is the semantic vector and this um if you if you really want to geek out i would go see eric redmond's talk and he will dive into this really deep so for for our purposes here the way that i would i would convey this is uh that graphic above it is a high density vector space and that probably makes everybody scratch their heads but what it does this model is it's looking at the relationship of products between each other and across the catalog and grouping them so these concentrations of colors and dots or similar products and the relationship or distance between the dots indicates how similar to the other ones they are and so for a quick analogy if we think about maybe a grocery store and if we're at the fish counter we we may have uh the fish laid out there like the salmon fillets and and then some of the rock fish and and some of the shellfish and each of those uh groups of seafood are compartmentalized and then also you know you might notice that there's lemons or ingredients for tartar sauce or some old bay and so those generally would be further across the grocery store but are bringing brought closer because there's a relationship there so um the lemons have relationships other produce but also in this in this instance so uh in the grand scheme of things we're using models like this to to understand uh not only like a semantic similarity between products right just really beyond their their lexical definition but really the goal of the shopper so between queries and products as well so that we can make sure we're generating products that can meet the goal of the shopper that may not be just on a keyword match or a lexical relationship so it tends to get pretty deep uh into the recommendations and and very valuable so that's a very sophisticated approach that we've again packaged up so that it can be picked up trained deployed relatively easily so talking about training what do you need to train these models a lot of times it just comes down to your product catalog signals which you're likely collecting already so these are the clicks add to carts purchases these can be signals from the store these can be signals from customer service anything that can help train uh more relevancy more sophistication in these models can be can be captured if they're not already and brought in and then business intel and what i mean by this is oftentimes because of the challenges that are faced and some of the legacy technologies for product discovery there may be a lot of human intervention by way of uh creating relationships between either words so like synonyms but also like project oriented things maybe so if you're going to repaint a rusty railing you may have built manually built a project for that and have a relationship of products that you've put onto a product detailed page so things like that can be taken and put into models to help train them as well and then lastly post-processing right which is which is really the bulk of this human synergy with ai where we can block boost we can put rules in we can if you think back to those examples from the slippers we can make sure that there's gender rules that are put in to only recommend men's products to men's or grain free dog food if you're shopping for grain free dog food so there's different ways that we can make these models stronger by by post-processing them so i want to show an example of what that can look like and so here we are this is just a demo site that we've that we've assembled but on this pdp so i can scroll down to find the recommenders and you can see that they're all relevant right these are other jackets that are women so we've done a good job of delivering that and then when we think about post-processing there are different ways that we can do this right so we can boost with signals we can apply rules there's all kinds of different tweaks that you can do right in fusion to deliver a more enriched experience right so maybe taking a color boost of the the color of the product on the pdp and and making all of the products in the recommendation carousel show that color as well so different different tweaks and things that you can do so lastly i want to take a look at uh right human intervention so what are the different tools that we have to really get the human involved into these experiences and and this is really what i love having i've done some of the merchandising work in the past on both the business side i've helped out on the technical side here and uh this you know merchandisers know best there's there's always times where it just makes sense to have the merchandiser put their touch on it and whether you know it's for campaigns or emails product detail page or the home page there's always going to be a need for uh the flexibility to to curate these and so with our predictive merchandiser tool we can not only block but also pin and there's also a breadth of functionality on there that expands beyond recommenders and i think if you go over to tom and katie's talk they'll dive more into that as well so um so that's the human intervention side of it now when i thought about what use cases i wanted to show here you know i don't want this to be so much of a recommenders 101 and as you know this is where the recommenders go but i want everybody to come out of this thinking okay we can try something new we're in these unprecedented times where the evolution of product discovery is accelerated drastically and at this point you know to me it seems like this is this just seems like the natural next step so let's take a look the this virtual uh shopping experience right i've seen these popping up across all kinds of brands and the value that you can provide is extraordinary by having a human individual participate in the shopping journey when you know we can't do that at the stores anymore and so what if uh there was a human on the other end but the the experience was still being driven by machine learning and ai so it's kind of the best of both worlds and so when you click into this you know the first thing you see product recommendation so to me this sounds like a tremendous use case gifting also that sounds great and if you go a little bit deeper to schedule an appointment they will ask for a bunch of different information and at this point they know everything about me they know what i've bought in the past what size and colors fit me what my order history is from both in-store and online and what my interests are for this conversation so there's a ton of rich data and this i can really drive uh at the end of the day a lucrative experience and so maybe we start out with okay what i came here for i you know i got my slippers now now i need some joggers so this model could recommend the joggers and then how about maybe a propensity type model to say hey garrett we know that you've bought all these things in the past and when we we look at that against all the other data we think that you might be a good candidate to you know start getting into our polo short line of business and to make a recommendation there um but there's really endless opportunities here and it's pretty exciting lastly uh from uh from use case perspective uh the in-store experience right so we're gonna we're gonna go back into stores and and when we go back into stores it's probably gonna feel a little bit different and i'm hoping what's been happening over this last year is we're figuring out creative and innovative ways to enhance that experience to get people to come back and when you look at the situation here with the tablet and recommendations on it you know whether it's products that are in stock or out of stock or again knowing who that shopper is and being able to recommend pro products based on their past purchases or their painting what are what are other products that should be recommended for that in the event that that associate doesn't have that that knowledge so again another huge opportunity here and as far as our recommenders go uh you know this this i think i showed a fair amount of apparel but between the dog food and the paint i just want to make it clear that these go across b to c across b to b d to c life sciences apparel grocery these models are ready to roll and we can cover the spectrum so before i wrap up i just want to let you all know that you know this is lucid works and we're not just a search company and i hope that this was a compelling and convincing uh demonstration of where we're at with recommenders and that we can compete in the space and again you don't need an army of data scientists to be successful with these tools they're they're ready to roll so thank you appreciate your time and your attention please join me and actually a couple of my team members as well for a little bit of q a right after this thank you
2021-03-24