Salesforce’s Einstein boosts search in its Commerce Cloud

Bit by bit, Salesforce is bringing its Einstein AI layer to all parts of its clouds.

This week, the company is launching Einstein Search for Commerce in its Commerce Cloud so that stores using its e-commerce platform can make it easier for visitors to find stuff on their websites. The company says its platform is used by 2,400 sites in about 50 countries, handling over 350 million online shoppers monthly.

In May, Einstein-powered enhancements were made to Commerce Cloud’s Predictive Sort, Order Management and other features. The company built its Commerce Cloud around its purchase last year of Demandware. Last month, Salesforce’s Sales Cloud got several new Einsteined features.

[Read the full article on MarTech Today.]

How Pinterest has developed computer-vision technology to power visual search for users and brands

Pinterest wants to do for visual search what Google has done for text search.

Since assembling a small team to develop computer-vision technology in 2014, the search-slash-social platform — where each month 175 million people organize, share and, increasingly, find ideas ranging from food recipes to interior design to outfits to tattoos — has been working on ways to make images searchable like text. Its work has evolved from picking out images that look alike to identifying objects within images to converting images into search queries to, most recently, targeting ads based on images.

For the complete story of how Pinterest has trained its computers to index objects within images and how it is opening up its visual search platform to brands, read the full article read the full article on MarTech Today.

Twiggle offers plug-and-play semantic search to online retailers

Twiggle is a company founded by two former Google employees. It promises to bring “semantic search” to e-commerce sites with minimal technical integration. Udi Manber, formerly head of search at Google and Amazon’s A9, is a board member.

Last week, the company released a “Semantic API,” which “gives retailers the ability to add a semantic layer to their existing search engines and interact with their online customers in a more personal and natural way.” The idea is that Twiggle will bring state-of-the-art search sophistication to companies that can’t develop the technology on their own.

I spoke with Amir Konigsberg, CEO of Twiggle. He told me that his company spent three years building out an ontology that allows Twiggle to process and deeply understand billions of products and associated attributes. Twiggle also does data structuring and normalization and enhances products with additional metadata.

Konigsberg critiques current e-commerce search capabilities as being very basic and not delivering an optimal user experience. Clicks and query matching are typically used to define relevance. He says that Twiggle doesn’t use clicks, it does semantic parsing to decipher queries and better understand them. It also allows for longer or more specialized queries.

Perhaps more significantly, Konigsberg told me that Twiggle’s solution doesn’t require online retailers to replace their enterprise search infrastructure. The API passes each query in real time for Twiggle analysis.

Konigsberg argues that Twiggle can process even the largest product catalog in a matter of hours and bring “Amazon-like search quality from day one.” Normally, it would take huge amounts of data to make comparable improvements in relevance. Konigsberg says with Twiggle’s ontology, sematic parsing and natural language processing (NLP), that’s no longer required.

Assuming this is all accurate, it becomes especially significant in a mobile context, where users are less patient and forgiving. The structured data and semantic understanding would also shine in a voice search or virtual assistant context, where queries become more conversational and specific or elaborate.

Konigsberg says, “[Our] NLP can support long and complex queries. You can be very descriptive.” He added, “We always get right [to] the product you’re looking for.” As evidence, he says that Twiggle delivers an average conversion rate lift of 18 percent to its customers.

Pinterest’s Lens app turns your phone’s camera into a search bar

Pinterest is bringing search to the real world.

On Wednesday, the social-network-slash-search-engine introduced Lens, a mobile app that uses the phone’s camera to recognize physical objects and pull related items on Pinterest. The company announced other new features that extend search beyond the query bar on Pinterest, such as a new way to find products to purchase. To read more about the news, check out the original article on Marketing Land.

Shopping campaigns: Play like every day is a holiday

Shopping campaigns are becoming a major source of website clicks and revenue during the holiday season, and the “Shopping Campaigns: Play Like Every Day Is A Holiday” panel at SMX Advanced featured tips and advice from three PPC veterans: Ann Stanley, Todd Bowman, and Mona Elesseily.

Ann Stanley: Shopping ads, buy buttons, social commerce & remarketing

Shopping ads and buy buttons are everywhere. Stanley explored those areas where ads are driven by product feeds, and clicks either lead to retailer websites or convert on host platforms. Her talk was full of data insights and provided a neat map divided into three conversion areas:

Area #1: The search giants: David Bing vs. Goliath Google

Thanks to Windows 10, Bing Shopping ads share is growing (21% US, 9% UK). With Google Shopping winning by volume, Bing nearly always shows lower CPCs. In terms of conversion and ROAS efficiency, results vary heavily by vertical. Bottom line: if you target the US or UK, give Bing a try to see how effective it could be for you.

Area #2: Social commerce: growing ecosystem with many faces

Social commerce still carries the promise of incremental returns by opening the door for the impulse purchases in the digital world. Stanley subdivides the area into four main types, showing how the growth potential is backed up by a hugely diversified ecosystem:

    F-commerce: Retailers can offer their products within the Facebook environment. Shoppers can browse through the retailer’s product tree, liking and commenting on products as they go and putting them into a Facebook cart with checkout option.Buy buttons: Although Twitter has discontinued its buy button in favor of Dynamic Remarketing, other players like Facebook, Instagram and Pinterest still continue offering the ad format that allows brands to educate and sell at the same time.Dynamic remarketing: Facebook, Twitter and Instagram start the next generation of remarketing, where the shopper does not necessarily notice that he or she is being retargeted. In a nutshell: using a remarketing pixel, retailers inform the platform about what products users engaged with (looked at, put into cart, bought). They can retarget these users showing alternatives or complements to the products they interacted with — to cross or upsell. All of this is, of course, based on product feeds, dynamic ad creation and even works across devices.Social Shopping sites (Polyvore, Houzz, Opensky): These sites use peer-to-peer influence around different verticals (Fashion, Beauty, Jewellery, Home & Garden, etc.). Communities create content that inspires to buy. Merchants either directly offer their products or push these using paid features like sponsored promotions.

Area #3: Products and price buttons in display ads

Google starts using the Merchant Feed in many more areas: TrueView and the brand new Google Contextual Dynamic Creative carries product ads into YouTube videos and into the Display Network.

Use remarketing to tie channels together

To think omnichannel is the new, old marketing imperative. Wait, but how? With different channels managed by different departments, this is clearly a tough task. Stanley comes to the rescue by showing smart remarketing techniques:

    Retarget on Google Shopping. Try leveraging low-cost “honeypots” like educational offers (“Learn about how to…”) or open competitions (“Win these tickets / products”) to drive visitors from search, social or display to your site and create remarketing audiences in Google Analytics. Use RLSA in Shopping ads to retarget users when they express a buying intent on Google.Retarget on Facebook or Twitter. By using Facebook’s or Twitter’s retargeting tag, dynamic product ads can be shown on the social platforms, using additional audience targeting provided by the platform.Retarget on Display. Combine Google’s display campaigns for dynamic retargeting with dynamic ad units and overlay with Analytics remarketing audiences using “Target and Bid” to show display ads only to former tagged visitors.

Stanley also made it clear that a good feed management software is at the heart of being able to scale product ads across channels.

Shopping Ads, Buy Buttons, Social-Commerce & Remarketing: How to Develop an Integrated Cross-Platform Strategy By Ann Stanley from Search Marketing Expo – SMX

Todd Bowman: Strategies to take your holiday shopping campaigns to light speed

Bowman started to show how important PLAs have become, especially in terms of click share against text ads. For Non-Brand, it rose to a staggering 70% on Google, and 27% on Bing for U.S. paid retail clicks.

Collaborate to avoid a Death Star scenario

Given that importance, departments need to collaborate closely to succeed during special sales periods like Black Friday. As product data quality is at the heart of online sales, the Merchandising department needs to gather excellent product information from suppliers when buying the products. It is exactly that information that the Web Development needs to expose on the website.

Marketers’ task is to make sure the quality data is used properly, e.g. in the Google Merchant feed. Marketers also need to look at past data — when did clicks occur that drove sales? People are doing research for deals much earlier, and just pushing bids on cyber weekend will cause them to miss out on a lot of sales.

Shopping Ads: taking into account search query conversion probability is rewarded with more revenue and higher ROAS

When structuring Shopping campaigns, Bowman recommended analysing search queries and applying different bidding strategies: Generic queries like “Bluetooth Speaker” and a more targeted “JBL Bluetooth Speaker” may otherwise end up in the same campaign showing the same product and the same bid.

However, targeted traffic is more likely to convert and should thus be rewarded with a higher bid. That, According to Merkle, is a driver to affect the bottom line strongly in terms of revenue as well as ROAS. (At Crealytics, we focus on the same strategy and can confirm these findings.)

Brick-and-mortar stores benefit from strong click growth and high CTRs when using Shopping Local Inventory Ads — yet the value is hard to measure

When connecting offline and online, Local Inventory Ads (LIA) offer opportunities to traditional stores. Not only can a store be easily located on Google Maps, but also do LIAs provide a way for potential shoppers near the store to learn about products and availability.

Google is testing a lot of new Shopping features, including a “Buy online, Pick up in Store” currently in Beta. Bowman expects the majority of digital ads in the future be powered by feeds. Yet the biggest LIA challenge remains to track store sales and offline orders to measure the value of its clicks. At least measuring store visits is, as of now, no longer a vision, with Google offering a “Store visits” beta.

We at Crealytics also expect GTINs to play a major role in conversion attribution in the future. As Bowman notes: Google keeps being very serious about this number, rigorously disapproving products that fail to carry them in the feed after a 30-day grace period.

Strategies to Take Your Holiday Shopping Campaigns to Light Speed By Todd Bowman from Search Marketing Expo – SMX

Mona Elesseily: Getting Better Bang for Your PPC Shopping Bucks

Elesseily provided a healthy a mix between Shopping present and future: tactics and useful tools for today, as well as an outlook on the next Google Shopping features for tomorrow.

Today: Combine Shopping with other ad formats, leverage Similar Audiences, and optimize for mobile

Combining Google Shopping & Dynamic Search Ads is a promising way to get additional shopping reach. Elesseily highly recommends to ask Google for help to set up additional DSA campaigns from the Merchant Center feed with their internal tools. Mix these with regular “backup” DSA campaigns to cover any product currently not in your feed. Bid low on the backup campaigns to avoid cannibalization, which can go up to a 25% in her experience.

As an estimated 74% of internet traffic will be video by 2017, YouTube should not be underestimated to leverage more reach. Combining Trueview for Shopping, layered with dynamic remarketing yield amazing results according to Elesseily.

In the past, “Similar Audiences” were only available for Display. But since the Google I/O Summit in May 2016, these are also available for Shopping, Search and DSA campaigns.

Intended to reach new customers, Similar Audiences are incremental and recent; audiences do not overlap with existing RLSA lists und remain 24 hours on the list only. Her bet is that you’ll have to include higher funnel audiences (e.g. cart abandoners) to get traction from Similar Audiences as lower funnel audiences (e.g. purchases in the last 30 days) will be too small to provide significant traction.

For Mobile optimization, Elesseily considers site speed a crucial factor. To get started, she recommends the official Think with Google tool, testmysite, which provides actionable recommendations customized to your site.

Tomorrow: More ad inventory, better offline attribution, conversational shopping and…

In addition to more Google inventory opening up for Shopping ads (like the image search) Driven by mobile and local initiatives, Elesseily sees online and offline growing together heavily. Beyond informing the shopper what he can expected to find in a store, better in-store attribution remains a core area of effort.

Siri Voice Search, Amazon Echo and Chat Bots will become more and more popular. This changes the way people shop today – towards something more of a conversation. Advertisers will have to find new ways on how to respond to shoppers’ questions.

On the long term, we can also expect Google Tango, an augmented reality framework, to bring new digital marketing opportunities into the game. It can project virtual products into images of the physical space in nearly real time – just think of examples like placing virtual furniture into a room to see if they fit before purchasing them.

Getting Better Bang For Your PPC Shopping Bucks By Mona Elesseily from Search Marketing Expo – SMX

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

Google Product Listing Ads officially launch in Image search, among announcements for retailers

Purchases on Google testing continues, Product Listing Ads are getting access to more inventory and brick-and-mortar retailers running local inventory ads see some new features. These are among the announcements from Google at Shoptalk in Las Vegas on Monday.

Product Listing Ads in image search

Ads in Google image search first started appearing in the fourth quarter of last year. First noticed by Merkle, the ads are officially launching Monday. On mobile, they display in a carousel format above the organic images. Image search is considered part of the Google Search Network. If your Shopping campaign is opted in to Search partners, Product Listing Ads (PLAs) will automatically be eligible to show in image search results.

Store pickup promotion in Local Inventory Ads

Retailers using Local Inventory Ads (LIAs) can now include a “store pickup” link for shoppers who want to buy online and pick up their orders in store. The option appears on the local product landing page hosted by Google after a user clicks on an LIA.

Inventory search from local Knowledge Panel

Also new for advertisers running local inventory ads, users will be able to click a “Search items at this store” link in the retailer’s local Knowledge Panel to see whether particular items are in stock at their local stores.

Update on Purchases on Google

There isn’t any news here other than to say the pilot is still alive. Google launched the test for Purchases on Google — the “buy button” feature that allows users to buy products right from a product listing ad on their phones through a commerce experience hosted by Google — last July. Google said on Monday that Ralph Lauren, Ugg, Staples and others are continuing to test Purchases on Google

Faisal Masud, EVP of Global E-Commerce at Staples, says 95 percent of Staples products are now included in Purchases on Google.

John Kalinich, senior vice president, global digital commerce at Deckers Brands, told Google that UGG is seeing a nearly 50-percent increase in conversion rates with Purchases on Google and a 25-percent lower cost per conversion compared to mobile PLAs.

Etsy Boosts Search For Better Content Discovery, User Engagement

Etsy has more than 30 million items for sale from more than one million sellers globally. There are no SKUs, and most of the data is unstructured, creating a messy and massive discovery challenge for both Etsy and its users.

Accordingly the company is today rolling out more sophisticated search functionality, after a month-long beta test in which it saw increased engagement from both desktop and mobile users.

Etsy has always had search but at a basic level — it was described to me as a “one-size-fits-all approach” — that didn’t do a good job of showcasing the site’s products. The company is now doing a better job of recognizing user intent and delivering more tailored results.

The top image below is a “before” screen, and the second image below is the “after” screen. These screens may look similar but there’s now a great deal more going on “under the hood” to make results more relevant and to expose more “long tail” content to users, which is better for sellers as well.

Etsy told me that roughly 30 percent of the site’s queries historically have been very broad, “low intent” search terms. Rather than just exposing items in a crude ranking hierarchy based on superficial variables, Etsy is now presenting results that offer both categories (to expose more content) and single items, as the image below reflects.

Etsy’s search evolution is fascinating because it represents a larger problem or story — the challenge of organizing huge amounts of data and presenting it in useful ways for end users. What’s also interesting is that Etsy went old-school as it made the leap to a new system.

The company first tried a machine learning approach but discovered that it didn’t have a strong enough training set of data to go down that path. It wound up hiring a library scientist to create a structured taxonomy. Etsy also enlisted its sellers to help better organize, tag and categorize their own items. Now that it has created this new model and data set, it can later use machine learning techniques to further improve search and content ranking.

The new search capabilities provide more context and at the same time allow people to go much deeper into categories both with queries and subsequent refinements and filters.

Since the beta site went live for part of Etsy’s audience a month ago, the company has seen an improvement of more than 10 percent in user engagement. Mobile web engagement is higher. Mobile now represents more than 50 percent of Etsy’s traffic (apps + mobile Web combined).

The new search capabilities should be live for everyone on the site starting today.

Shopzilla's Rebrand As Connexity Opens Search Marketing Technology To Retailers

Shopzilla, best known for its comparison shopping engine, changed its name in September to Connexity Inc. to reflect a change in focus. The company still operates its consumer-facing sites, but now makes its B2C assets and shopping data available to marketers as managed services. As part of the rebrand, the company rolled out new and expanded search marketing services.

The company is essentially turning its own internal systems into a search marketing technology stack for retailers with PLAs, text ads and SEO offerings.

For search text ads, Connexity’s platform automatically generates keywords and scores them based on conversion performance. Ad groups are dynamically organized, and bids adjustments are automated daily.

Connexity’s product feed management platform automatically fills in missing product attributes and organizes product groupings and manages bids.

The platform also offers dynamic creative optimization for text ads and PLAs — testing product titles and dynamically pausing keywords that don’t perform.

The SEO platform is designed to support large retailers managing millions of products and pages and facing issues around duplicate content, faceted navigation, internal search, deep crawling and pagination. The platform offers real-time page scoring, advanced crawler management and dynamic XML sitemap generation tools among other features.

Last year, the company says it managed more than $60 million in ad spend on paid text ads and product listing ads, bidding on over 1.8 billion keywords for 100 million products. Connexity claims that $60 million in ad spend generated $1 billion in gross merchandise sales.

“Now marketers can leverage all that we have learned and built in our paid search operations to optimize every aspect of their campaigns,” says Bob Michaelian, EVP Marketing Services at Connexity. “With our proprietary bid management, automated keyword generation/landing page selection, and advanced ways to group and structure campaigns, we are now a one-stop-shop for all things search in retail.”

The company also plans to grow its programmatic display arm. Connexity harnesses data from its own sites, including Shopzilla and Bizrate, along with its product marketplace which syndicates across 3,000 publishers to create and target shopper audience segments.

Ahead Of Black Friday, Google Adds Shopping Details For Mobile Users

Google says that shopping searches from smartphones have increased 3.5 times compared to last year and that Google Shopping is already sending more mobile traffic to retailer sites than it did during the first week of December last year. To capitalize on this growth, Google is now providing more detailed product information available from Google Shopping on mobile devices.

When users search for a specific product on their smartphones or tablets there will likely be a “See more” arrow below the Google Shopping ad on search results (note that text ads are being driven to the bottom of the page). Clicking on the arrow shows more information about the product, where to buy it online, which stores have it available nearby and customer reviews.

Some products, like this Lego spaceship now have 360 degree views available from Google Shopping on mobile devices.

Lastly, Google is expanding Local Inventory ads to tablets. These ads are now available across all devices.

Google is clearly aiming to drive more mobile product searches straight to Google Shopping this holiday season, and beyond. With the huge growth in mobile search volume, retailers should make note of these changes to ensure their listings are prominent in Google Shopping and understand that mobile text ads for specific products are likely to see less and less action if this format succeeds.

Target App Now Features In-Store Product Search And Inventory Maps

Just in time for the frenzied holiday shopping season, Target is introducing product inventory search into its mobile app. The functionality is powered by Point Inside but doesn’t rely on beacons or other indoor location technologies.

Following a 40-store trial earlier this year, Point Inside announced this morning that Target will be rolling out its “StoreMode” product inventory search and mapping capability in all the company’s nearly 2,000 North American retail locations.

Smartphone owners will be able to use the updated app to search for local product inventory when not in the store. However, once users enter Target stores — store presence will be passively detected through geofencing — they will have access to a range of search and find capabilities. The app also provides a type-ahead/auto-complete feature.

Items can be quickly be added to shopping lists (also accessible outside the store). But in-store product locations will automatically be appended to items on the list. Accordingly, every product is mapped to its in-store aisle location, which can be accessed on a map. Point Inside also told me that all of Target’s Black Friday “doorbuster” items will be plotted on the interactive map, as well.

Below are some screens provided by Point Inside reflecting the basic elements of the upgraded user experience.

Point Inside has already deployed similar capabilities with Lowe’s and grocery chain Meijer.

The company said that more than 70 percent of retailer mobile app usage is focused on the in-store experience. Historically retail apps have been small-screen versions of their e-commerce sites. They’re just starting to realize there’s much more value in evolving these apps to in-store shopping assistants.

Point Inside’s Pete Coleman also indicated, based on data he’s seen, that between 10 percent and 30 percent of consumers leave stores without buying a desired product, because they can’t find it, even though that product is actually on the shelf.

Offering an in-store search or mapping capability dramatically increases user app engagement and retention, says Point Inside’s Coleman. Despite the fact that nearly 90 percent of user time is spent with apps, most consumers don’t have many retailer apps on their smartphones. These types of in-store features can help boost the numbers.

Indeed, nearly all of the top 100 retailers are deploying or experimenting with some sort of mobile-enhanced in-store user experience. Many retailers are deploying Bluetooth beacons and specifically Apple’s iBeacon spec (though Point Inside’s solution doesn’t rely on beacons). These top 100 retailers collectively represent more than $2.5 trillion in annual sales in the US.

An enhanced in-store experience is one of brick-and-mortar’s top weapons against pure-play e-commerce. In addition, expedited payment systems (e.g., Apple Pay) will also be part of this improved in-store experience among those that “get it right.” Despite the recent MCX payments controversy Target remains one of Apple’s Pay’s marquee partners.

Walmart also recently updated its app with a new in-store product search capability. Both Walmart’s and Target’s in-app product search features can be monetized through advertising. That’s not currently happening but it’s likely to happen over time as a new potential retail revenue stream — and digital version of in-store merchandising.

It also goes without saying that the search data that Target obtains from its customers, in the aggregate, will provide extremely valuable insights about product demand and user behavior on a local, regional and national basis.