EP. 60 | Smart Technologies from eTail West

This week in the Accelerator host Allison Hartsoe picks four smart and significant customer-centric marketing technologies that you should know. With over 161 vendors at the show, who all say nearly the same thing, it’s incredibly hard to find the signal in the noise. Allison identifies what nagging problems these four vendors solve for analysts and why they made the cut. From TV tracking to voice to payments and search, listen to this podcast to find out which technologies might be right for you.

Please help us spread the word about building your business’ customer equity through effective customer analytics. Rate and review the podcast on Apple Podcast, Stitcher, Google Play, Alexa’s TuneIn, iHeartRadio or Spotify. And do tell us what you think by writing Allison at info@ambitiondata.com or ambitiondata.com. Thanks for listening! Tell a friend!

Read Full Transcript

Allison Hartsoe: [00:01] This is the Customer Equity Accelerator. If you are a marketing executive who wants to deliver bottom-line impact by identifying and connecting with revenue generating customers, then this is the show for you. I’m your host, Allison Hartsoe, CEO of Ambition Data. Each week I bring you the leaders behind the customer-centric revolution who share their expert advice. Are you ready to accelerate? Then let’s go! Welcome everybody. Today’s show is a summary of interesting vendors from e-tail which just happened in February of 2019, and with over 7,000 Martech companies, the amount of noise and the vendor space is astronomical. I asked multiple retail attendees which vendors they thought were the most interesting, and by far the most common answer was, I don’t know. I’m so confused. And I agree, there were 161 vendors with booth or table space at the show, and those are just the exhibiting vendors.

Allison Hartsoe: [01:13] I know tech, and it was pretty difficult to simply understand what each company does. In fact, if you took just five minutes with each one, it would have taken you a solid 13 hours and 40 minutes in the exhibit hall to meet them all. And this is largely because many of them started as point solutions, and if since broadened their scope. So they’re all overlapping. And for example, the customer data platform space has become a hot place for companies that were formerly a channel optimization solution to reposition themselves. And I’m going to dig into that topic more on a future show because there’s a lot of richness and a lot of interesting stuff in the customer data platform space. So I really want to dedicate just one show to it. Now, companies are saying almost the same things. They’re all saying customer engagement. They’re all saying Omni channel or customer experience or increased conversion.

Allison Hartsoe: [02:10] And the new one is we help you personalize. So, uh, I was specifically looking for a technology that solves a problem that I know from the analytics perspective is difficult. And I wanted this tech to solve it in a creative way or a way that gave the retailer some advantage. Uh, perhaps more than other tools. So here are three that I found, and why I think they’re worth looking at. Now the three tools I’m going to talk about today are Alfonso, which connects TV to digital. We gotta talk about your answer.io, which is a voice solution. And I’m going to talk about Spreedly, which is a really interesting play on e-commerce. They’d been around for a while, but there’s an interesting angle to this one. Okay, let’s start with Alfonso. So Alfonso solves a really difficult problem of making TV as measurable as digital. I’ve ever been an analyst. You know how difficult it is to see this information.

Allison Hartsoe: [03:11] So what they do is they measure the direct impact between TV advertising and a website. What we traditionally see our spikes and direct traffic or maybe spikes to a landing page. And this is just inherently much better, much more measurable. Unlike Nielsen who uses the older technique of selecting households to create panel data, and then diaries where people write down what they’re watching and when, and then they extrapolate that to determine the number of in the viewership. Alfonso kind of flips this upside down. What they do is they monitor everything, and then they measure the outcome. So, they’re using AI, which I know is a horrible buzzword, but they’re using AI to recognize companies logos and content across all TV streams. And if you know anything about Ai, this is actually a pretty good use case of the technology. Uh, when we look at things to classify or to understand it, the classic example of Google was, could I hug this?

Allison Hartsoe: [04:15] A very interesting story, but back to Alfonso. So on the one hand, they’re using AI to recognize a company’s logo and the content across all TD streams. And on the other side, they’re connecting to smart TVs and device fingerprints to understand what’s actually being watched, where, and for how long. So that’s pretty cool. They get real-time viewership data from about 34 million households, which represents a quarter of US households. So views come from mobile devices as well as enabled smart TVs. Now we’re not talking about 100% of the US households, but the data is significantly better and more insightful than what we had before. So, for example, you could use their data to extend your digital reach on certain campaigns where you could see that someone saw your ad, then they click through to the site, and then you could remarket to them through search, display and social.

Allison Hartsoe: [05:14] So it would be a much more precise digital retargeting. You could also use it to understand the share of voice from a competitor. This is an area where we’re oftentimes blind because we don’t understand exactly how much volume our competitors are putting in the market. And not that we want to imitate that volume, but it’d be certainly helpful to understand while competitor A is marketing really heavily, in this particular demographic zone. And then maybe could we understand the impact on our ads and stores at that same time. So it would help us understand where to put the gas pedal down, and maybe where we need to pull back if we’re over marketing in certain areas. So I think those two use cases are pretty interesting, but those are not the only two. You can also do it for something like pre-exposure. So, for example, an insurance company has a list of every household who is rolling off their policies in the next 90 days.

Allison Hartsoe: [06:16] And they cross that with WHO’s watching their ads or the insurance. So they can then start to selectively retarget people who are exposed to the ads, and prevent any slippage or try to prevent any slippage of people who might not be renewing. So I think that kind of pre-exposure is a really interesting use case. Now they’re not using PII data because you know, naturally, you might think about this and go, uh, it’s just a little bit creepy, but they’re using household level data. So I think it’s a very close line between, is it PII, you know, is it our name, address and phone number or is it our behavior? Technically our behavior, what we watch on TV is not be a Pii, but it is incredibly predictive. And I think their data works really well for things that are household level decisions like insurance or vacations or maybe even automobiles or pizza.

Allison Hartsoe: [07:20] But it might not work so well for things that are individually purchased, uh, like my fashion choices might be harder to, to be precise about. So, uh, so they’re looking at the exposure share, did I see the ad versus the session share, did I go to the site? How much was spent and where to optimize? They can also connect to in APP activity. So anything that you can see on the digital space, they can just basically bridge that connection so they can see that you got a session or reactivation or a registration or someone made an appointment, which is a classic problem, right? We oftentimes use TV to advertise to get people through the door. And then the fact of did they actually come through the door is a big mystery. So I think this is an interesting technology that one final use case, which was about make goods that happen at the end of the year.

Allison Hartsoe: [08:15] So networks are moving a little bit closer to results, spaced advertising. They’re not quite there yet. But if you can really understand how much was exposed and look at the lift between the control and the exposure, look at it by day part, you might be able to not just refine your campaign based on spend, but you might be able to go back to certain networks and say, Hey, you know, your performance is really not so great compared to other networks, and see if there’s a way that you can negotiate more effectively. So I think overall, this tool, Alphonso.tv, which is spelled a l P H O, n s o. Dot. TV is a very interesting tool to look at when you’re trying to bridge the connection between your media buys, and your digital exposure, and making media more precise overall. Next up is your answer.io

Allison Hartsoe: [09:13] Now, it’s no secret that Mobol has become a dominant way to interact online, especially for eCommerce, but companies still struggle with what this actually means. Does that mean I need a downloadable app? Is a Mobol responsive site copying the main website enough or should we be doing something different altogether? Well, your answer takes the third approach by inserting voice search capability for your Mobol visitors. Now, they wouldn’t necessarily tell you that it’s search, and I’ll talk about that in a minute. And it’s also not just mobile. It’s desktop. But let’s be clear, the real upside here is mobile. So for companies with large product sets, this is a surprisingly difficult problem. It’s not just a search engine, but what they call a real intelligence engine, which I think uses NLP natural language processing, not keyword matching, not click the box filtering to get people to what they want.

Allison Hartsoe: [10:15] I actually think they’re using underneath vector-based analysis, which is the mathematical space between the words in a given language. But I’m just guessing on that. I’ve heard that that’s a very powerful search technique. It’s what underpins Google. It’s what underpins, um, certain really cool startups like Luminoso as well. So when you say, show me all your red dresses, size six under $50, your answer understands the color red that you want to dress, and you want it in size six, and that the price must be under $50. And so the results that are returned, it match exactly what you’re looking for. And it’s the goal is to save time and help your customers get to check out faster and more efficiently. It also understands general questions about the shopping experience, like what are your shipping costs, and what is your return policy. So that’s good overall. But, in order to check this out, I actually went over to Amazon, and I did this exact same example.

Allison Hartsoe: [11:16] You know, I want to see red dresses size six under $50 just to see what I would get, and what came back from Amazon in the top 10 results where one red dress for a woman, a children’s red dress, a frilly underskirt and red, like a tutu, children’s leggings and a blue lace dress. All of that by asking four red dresses, size six under $50. Now I think that’s incredibly imprecise, especially for coming from Amazon. Can you imagine what it’s like with the rest of the retail space? So I did the search again, and I added the word woman in front of my search just to see what would happen and see if that would improve the results, and sure enough that did take out the child’s stress and the legging, but it increased the number of Tutus and added a blue watch in addition to the blue dress.

Allison Hartsoe: [12:11] So clearly this ability to get to the point when you have a large Dataset for products is a problem. They have a testimonial on their site, which I thought was worth sharing, and it said, without a doubt the most impactful capability we have ever added to our site. We did almost nothing on our end. We added a microphone link, and we were good to go. What I think is so interesting about this, quote, is it’s from a leading apparel site, which I sometimes find on very effective technologies. The people who are using them refused to give their names because they don’t want the competition to know what they’re using because they’re getting a real advantage out of it. Now is not the case. Yeah, maybe, maybe not, but I think this technology is pretty damn cool, and I really like their solution. Mobol is oftentimes what we see coming in from advertising.

Allison Hartsoe: [13:04] And it is a fantastic use case as people get more and more comfortable for using voice for search. So check out your answer.io and be sure to read their limitations of AI as well as the future of voice sections in the DigiNo part on the top Nav. Very interesting tool, very interesting technology. Uh, finally I’m going to cover spreedly.com, and this is spelled Spreedly, s p r e e d l y.com. This tool has been out there for a while, and it’s was originally offering consumers a credit card vault in the cloud. When that adoption didn’t take off, they pivoted to support transactional businesses like e-commerce. So they’ve been kind of shifting around a little bit. I like to think about them like a router on the Internet for payments. They have access to 120 different payment gateways worldwide and 220 payment services. So you plug in once, and you get access to all these places, you plug in, or your developer does.

Allison Hartsoe: [14:09] When you pull up the site, and you look at the tool, it tilts very technical, but that’s not the part that I want to talk about. The reason I like this tool is because it gets back to customer equity and analytics, which Spreedly doesn’t talk about a whole lot, but I did gain from my conversation with them at e-tail. It makes no sense to run customers into a wall, and this is often what happens when a customer goes to purchase again, and they have to reenter all their data or worse, their credit card is expired, and this actually happened to me at retail while I was using a ride-sharing service, which I won’t name, but the interface was so arcane, so awful that to update my business credit card, it took two of us 10 minutes to figure it out. You actually had to call the car and have a live request before you could update the card that was going to be charged, which is completely upside down in my mind. Anyway,

Allison Hartsoe: [15:06] Spreedly contains not only the secure storage for your customer credit card data, which obviously reduces your risk but they update the credit cards automatically, and they’ve done some research on this which causes the decline rates to move from 21 to 28% to about five or 6%. I think that’s really great. That’s basically giving your customers, you stop running them right into a wall because, oh my credit card’s not good anymore, and instead you pass them right through in a very expedited fashion by having the technology do more heavy lifting, but that’s not the only aspect that’s interesting about the tool. The other aspect is related to a huge blind spot in e-commerce, which is which payment should I offer my audience and how well do they work? I was once privy to a PayPal analysis at a very large retailer, which took several weeks just to figure out what was happening.

Allison Hartsoe: [16:02] The end diagram, which was this giant map looked exactly like a rat’s nest. It was a horrible customer experience, not to mention lost revenue. People were going in and out of the other service, and getting turned around, and coming back in, and sessions breaking. It was a mess, but when you have Spreedly as the intermediary, and they don’t touch the cash, they’re just passing things along. They then know who works really well and who doesn’t, so you can set business rules to reroute payments if a gateway is too slow, or you can also see the metrics associated with various gateways on the platform. Things like the success rate, the volume, the average ticket size, so that helps you understand or at least diagnose which payment technologies are helping or are supporting your larger transactions and which ones aren’t. So you didn’t have to be in the dark about what payments you’re best customers are using and which ones are worth keeping.

Allison Hartsoe: [17:06] In steadily, I think that could also give you leverage to refuse to work with a gateway or at least pay less based on their performance. God knows I wish I could do that with my local Internet provider. We use a local cable provider where the cable is sometimes up super fast connection, and then you don’t get it. I wish I had that kind of leverage. So the ability to give the retailer more leverage and a view into the unique stream of customer knowledge on payment preference and performance is a big reason why I picked this tool. The checkout Spreedly, s p r e e d l y.com. Now I know I said three vendors but I’m actually going to add in one more, and that is constructor dot i o. Constructor io is an AI optimized site search tool. Again, buzzworthy I know, and we’ll come back to that, but the company’s founders worked together at Shutterstock where customers would often enter generic search terms like business to find photos, and then come away disappointed, not unlike what most retail sites must go through.

Allison Hartsoe: [18:14] And they found that by adding modifiers like a business place or business meeting, they can help a customer find what they were looking for a little bit faster. Now, of course, that made them more likely to purchase. And this is a very simple autocomplete function, but it’s significantly increased the site’s revenue. So then I thought why couldn’t every site use this? Now if you remember, Google used to have a site search tool that you could actually install on your website, but it has since been deprecated, and that has left a bit of a gap in the market. And so most sites we’re building their own tools from open source options, but that still results in a lot of custom code and maintenance. It’s kind of a pain. Moreover, the site searches treat every search as if they’ve never seen you before, so they don’t learn and improve on the results.

Allison Hartsoe: [19:05] So what constructor does is it creates a virtual loop by feeding engagement data into the site search, which makes users more likely to complete a conversion activity, like watch a video, read an article, or of course purchase. Now Constructor says that it has helped e-commerce customers achieve up to a 23% revenue growth, which they further quantify they say that for their larger e-commerce customers, that’s actually eight figures. Now back to the AI point, this technology is using a combination of pieces together. They use natural language processing, which might be obvious, and they use machine learning, enhanced re-ranking of results. And then finally they load in collaborative personalization of their results to get another 6% left. Now that’s a lot of terminologies. So here’s an example from their site, particularly around that last piece. Consider a customer who regularly purchases organic products on your site like milk, bananas, and meat. If the customer searches for Apple,

Allison Hartsoe: [20:16] Will they see organic apple at the top of the results? Well, ideally with constructor the answer would be yes. So constructor search makes individualized recommendations by using product affinity data from these customers and many others. They call it collaborative personalization. And by filtering the histories, the purchase histories of other apple buyers, and analyzing them against their customers, organic, heavy purchased history, they can use an algorithm which is likely to find and rank organic apples above other varieties. So the next time that person comes in and searches the word apples, then organic automatically flows to the top without it being in the original query. Now here’s a subtlety you may not have caught, but I suspect is what’s happening. I think they’re looking across their network beyond your site to really understand more about the customer, almost like a mini Google or Facebook. That’s why I think this technology has the potential to be really interesting.

Allison Hartsoe: [21:21] So now again, back to just your site. From an analytics perspective, the use of site search can be a signature of confusion and I think it’s very valuable to find and rescue these customers, because essentially you’re showing them that you value their time because you’re enabling more convenience, and frankly folks it’s 2019, and nobody has time to wade through poor search results anymore. Constructor Ios clever approach is definitely steeped in customer knowledge, and I think it’s well worth checking out. You can check them out at obviously constructor.io. Now if you want to talk about what technology is right for you or perhaps build a roadmap to find your way, you can always reach me at Allison at ambition data or at a Hartsoe on Twitter or Allison Hartsoe on Linkedin. Be sure to like, share, comment on this podcast with other vendors who have impressed you. I’m very interested in hearing you know where the signal is in all of that noise of Martech vendors and as always everything we discussed including direct links to the companies I mentioned. We’ll be at ambitiondata.com/podcast. Thank you for joining me today. Remember when you use your data effectively, you can build customer equity. It’s not magic. It’s just a very specific journey that you can follow to get results.

Allison Hartsoe: [22:49] Thank you for joining today’s show. This is your host, Allison Hartsoe, and I have two gifts for you. First, I’ve written a guide for the customer centric Cmo, which contains some of the best ideas from this podcast, and you can receive it right now. Simply text, ambitiondata, one word to, three, one, nine, nine, six, (31996) and after you get that white paper, you’ll have the option for the second gift, which is to receive The Signal. Once a month. I put together a list of three to five things I’ve seen that represent customer equity signal not noise, and believe me, there’s a lot of noise out there. Things I include could be smart tools. I’ve run across, articles I’ve shared cool statistics, or people and companies I think are making amazing progress as they build customer equity. I hope you enjoy the CMO guide and The Signal. See you next week on the Customer Equity Accelerator.