首页 > 资料专栏 > IT > IT技术 > 人工智能 > salesforce_人工智能如何影响收入和购物(英文)2018_25页

salesforce_人工智能如何影响收入和购物(英文)2018_25页

zhuwenz***
V 实名认证
内容提供者
热门搜索
资料大小:5847KB(压缩后)
文档格式:WinRAR
资料语言:中文版/英文版/日文版
解压密码:m448
更新时间:2019/7/7(发布于江苏)

类型:积分资料
积分:25分 (VIP无积分限制)
推荐:升级会员

   点此下载 ==>> 点击下载文档


文本描述
Personalization in Shopping Data from 150 Million Shoppers on How AI Impacts Revenue and the Shopper Journey 15+ 2 3 4500+ Table of Contents Key Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 03 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 04 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 05 The Impact of Personalization on Conversion Rates and Shopper Spending . . . . . . . . . . . . . . . . . . . . . . 06 Innovation Q&A with a Data Science Visionary . . . . . . . . . . . . . . . . . . . . . . . . 10 Trailblazer Story: Stonewall Kitchen . . . . . . . 12 The Unique Traits of Shoppers That Engage with Recommendations . . . . . . . . . . . . . . . . . . 14 Trailblazer Story: PacSun . . . . . . . . . . . . . . . . . 17 5 Best Practices to Implement Personalization . . . . . . . . . . . . . . 19 Trailblazer Story: Room & Board . . . . . . . . . . 21 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Resources to Learn More . . . . . . . . . . . . . . . . . 24Key Findings The following is a summary of key personalization trends, findings, and benchmarks.1Product recommendations drive revenue. Visits where the shopper clicked a recommendation comprise just 7% of all visits, but 24% of orders and 26% of revenue. Page 6 Shopper spend soars with personalization. Purchases where a recommendation was clicked saw a 10% higher average order value, and the per-visit spend of a shopper who clicks a recommendation is five times higher. Page 7 Different devices mean different behaviors. For example, mobile shoppers that click recommendations complete orders at a higher rate than desktop shoppers. Page 8 A winning combination: personalization and site search. Shoppers that use search and click a recommendation convert 3.7 times more often than those that only search -- and 4.2 times more on mobile. Page 95Recommendations are directly linked to longer shopping visits. Shoppers that clicked a product recommendation spent an average of 12.9 minutes on-site vs. 2.9 minutes for those that didn't click recommendations. Page 14 Your recommendation-clicking shoppers have a high propensity to buy. Only 6% of shoppers clicked a recommendation, yet 37% of shoppers that placed an order clicked a recommendation. Page 16 Recommendations lead to sales. Nearly one in four products bought by recommendation-clickers came from recommended items. Page 1626374PERSONALIZATION IN SHOPPINGcommercecloud|3Methodology The Personalization in Shopping Report is an analysis of more than 150 U.S.-based direct-to-consumer digital commerce sites that used personalization on-site from March to June of 2017. The data includes shopping activity across more than 250 million visits, which created more than half a billion dollars in gross merchandise value. THIS REPORT ANALYZES SHOPPING ACTIVITY ACROSS250 150Million visits to ecommerce websitesTo qualify for inclusion in the analysis set, a digital commerce site must have transacted throughout the analysis period of March to June and met a monthly minimum visit threshold. Additional data hygiene factors were applied to ensure consistent metric calculation. Data footnotes are noted inline as appropriate in the report to provide additional clarity on analysis. This report is not indicative of the operational performance of Salesforce Commerce Cloud or its reported financial metrics, including GMV growth and comparable customer GMV growth.Million shoppers$550 320Million worth of ordersMillion shopper clicksPERSONALIZATION IN SHOPPINGcommercecloud|4Introduction Ask someone how they shop today, and you'll encounter nuances and technologies that were barely on a retailer's radar ten years ago. From artificial intelligence (AI)-powered product recommendations to receiving deliveries in less than an hour, shopping is being disrupted to an unprecedented extent.Technology makes it possible to connect with customers in new and exciting ways. But with increasingly mobile and task-switching shoppers navigating expanding product catalogs, retailers need to be ready with personalized content and intelligent recommendations. They must show shoppers exactly what they want to see, efficiently and in real time. They also must consider how to leverage the power of AI to anticipate their customers' next moves. That's why personalized shopping experiences are the future of commerce -- and the only way to compete amidst consumers' expanding expectations. Looking for proof This report analyzes the shopping activity of more than 150 million shoppers and their 250 million visits to ecommerce websites -- that includes $550 million worth of orders and 320 million shopper clicks. We'll look closely at this data as it relates to personalized product recommendations and how these recommendations impact conversion rate, average order value, per-visit spend, and more. You'll also learn about how to implement successful personalization tactics -- without an army of data scientists -- and hear stories from retail trailblazers. From this research, you'll discover that personalization is critical for a better shopper experience and increasing your sales dramatically and intelligently.YOU MIGHT ALSO LIKEPERSONALIZATION IN SHOPPINGcommercecloud|5The Impact of Personalization on Conversion Rates and Shopper Spending Providing your customers with a personalized shopping experience is now the cost of entry to retail. Fifty percent of consumers say they're likely to switch brands if a company doesn't anticipate their needs, and 58% of consumers say technology has significantly changed their expectations of how companies should interact with them.1 But personalization is more than a boon for customer experience; it is driving digital commerce sales. In this section, we take a deep dive into the business implications of AI-driven product recommendations. You'll learn the tangible financial results that retailers tracked when shoppers engaged with their recommendations. For the purposes of this research, a recommendation refers to a suggested product that a customer sees and can click or tap to reach a product detail page or category page, or within a cart (75% of total recommendations, 16%, and 5%, respectively). And if you're thinking AI seems out of reach, don't worry -- the basic recommendations you may already be generating on your site or encountering when you shop or browse Netflix are a form of AI. Eventually, more and more aspects of your site will be individualized, down to the content and offers. But implementing product recommendations, whether you manually generate or automate them, is a smart and profitable way to begin leveraging the power of personalization. 1. State of the Connected CustomerPersonalization and Revenue Today, most visits to an ecommerce site don't result in recommendation clicks -- but the visits that do are extremely productive. Visits where the shopper clicked a recommendation comprise just 7% of visits, but create 24% of orders and 26% of revenue.VISITS WITH RECOMMENDATION CLICKS7% VISITS24% ORDERS26% REVENUEPERSONALIZATION IN SHOPPINGcommercecloud|6SHOPPERS THAT CLICK RECOMMENDATIONS ARE: $96.00$96.00YOUR CARTXSSMLXLYOU MIGHT ALSO LIKEMORE LIKELY TO CREATE A SHOPPING CART4.5 MORE LIKELY TO COMPLETE THEIR PURCHASE4.5 $24.00 $37.00SUB TOTAL: $157.00 SHIPPING: $00.00 TOTAL: $157.00CHECKOUTIt's telling that such a small percentage of total visits accounts for such a large portion of total orders and revenue. Retailers have long surmised that recommendations were an important component of an ecommerce site, but this data shows how closely linked recommendations and high revenue are. The product detail page (PDP) is the new battleground, with 30% of site visits starting on the PDP.2 By shifting shoppers' attention to recommended products on PDPs and other key recommendation zones like category pages, retailers can guide shoppers toward exactly what they want in a digital sea of maybes -- with the potential to significantly increase orders and revenue in the process.shoppers are among your most engaged: Shoppers that clicked recommendations were 4.5 times more likely to create a shopping cart than those who didn't click recommendations.Shoppers that click recommendations spend 5 times more per visit.After clicking a few recommendations and filling up a cart, many of these shoppers proceed to convert at a higher rate than non-recommendation clickers. Site visits including recommendation clicks saw a 4.5 times higher conversion rate than visits where no recommendation was clicked. When one of these recommendation-clicking shoppers decides to buy, how does his or her order stack up Purchases where a recommendation was clicked saw a 10% higher average order value overall and 8% on mobile. Additionally, the per-visit spend of a shopper who clicks a recommendation is 5 times higher overall and 4.7 times higher on mobile. See the next section for more details about recommendation clicks across devices.From Carts to Conversion With site visits including recommendation clicks comprising just 7% of all visits but 24% of orders and 26% of revenue, you may be curious about tracking these numbers back to the baskets and carts where they originate. Your recommendation-clicking 2. 2017 Fashion Shopping Focus ReportPERSONALIZATION IN SHOPPINGcommercecloud|7Personalization Across Devices Shoppers value relevant recommendations -- and in micro-moments, those short periods of time between activities, personalization connects on-the-go shoppers with the best product for them in the shortest amount of time. When we talk about smaller screens, however, it's not only about mobile phones, although they get all the press. Tablet sales may have slowed in recent years, but tablet shoppers may still be among your most valuable. Tablet shoppers that clicked a recommendation had a conversion rate of 7.3%, compared to 4.7% for mobile and 9.8% for desktop. These tablet shoppers that clicked a recommendation also accounted for 30% of total tablet orders and 33% of revenue, despite comprising only 9% of total visits. Still, visits by desktop shoppers that click recommendations are strongest when it comes to visit duration (15 minutes on average), conversion rate (9.8%), and their propensity to create a basket (24% do so). Meanwhile, mobile shoppers that click recommendations complete orders at a higher rate than desktop shoppers (25% vs. 23%, respectively). So when looking at personalization across devices, it's clear that every device matters and benefits from a personalization strategy. If you want to strengthen your overall conversion rate, personalization is the answer -- across every device, visits including recommendation clicks have markedly higher conversion rates.RECOMMENDATION CLICK STATISTICSCONVERSION RATE YESVISIT SHARE YESBASKET RATE YESORDER SHARE YESREVENUE SHARE YESAVERAGE ORDER VALUE NO YES4.6HIGHER CLICKED ON A RECOMMENDATION 4.3HIGHER 4.3HIGHER 4.1HIGHER OVERALL COMPUTER7% 7% 7% 9% MOBILE24% 22% 26% 31% TABLET24% 23% 25% 30%26% 25% 26% 33%$ 117 $ 127 $ 101 $ 118$ 129 $ 140 $ 109 $ 136PERSONALIZATION IN SHOPPINGcommercecloud|8A Winning Combination: Personalization and Site Search One reason that personalized recommendations work so well is because they guide busy customers toward precisely what they want on increasingly overloaded websites. Shoppers don't have the time or patience to wade through pages and pages of products -- so personalization shortens the time between the initial impulse to browse and finding the right item. Another tool that helps shorten this time to product-finding nirvana is site search. For surgical shoppers, those customers who already know what they're looking for, search makes it easy to zero in on the perfect product on a site. And for shoppers in the exploration and discovery phase, search narrows down a world of options to the best few. Site-search metrics are similar to the visit and revenue contributions of recommendations; 10% of visits are from shoppers engaging with site search, and those visits yield 25% of revenue.3 The lesson for retailers is clear: Help customers find what they want, be it through search or personalized suggestions, and shoppers will thank you with higher conversion rates. These numbers relate to searching for a product and clicking on a product recommendation on a PDP, category page, or within a cart, not personalized search results.Through our analysis of 250 million site visits for this report, the combination of search and personalization emerged as a powerful force on conversion:Shoppers that use search and click a recommendation convert 3.7 times more often than those that only search, and 4.2 times more on mobile.Shoppers that use search and click a recommendation convert 2.1 times more often than those that only click on a recommendation.3. Unified Commerce and Shopper-First RetailingPERSONALIZATION IN SHOPPINGcommercecloud|9Innovation Q&A with a Data Science Visionary Feeling inspired by the possibilities of personalization but unsure where to begin Read our Q&A with Rama Ramakrishnan, SVP of Data Science at Salesforce Commerce Cloud, as he demystifies the key concepts of personalization for the non-data scientists among us.they actually care about, personalization is driving real growth. It's nothing less than a retail superpower.A: Rama: If your catalog has more than a few dozen products, customers may not easily find what they're looking for. And if today's impatient shopper can't find what they want in a click or two, they will leave you. That's why personalization is super important. Despite its clear importance, a fair number of ecommerce sites still don't use personalization. I am shocked by this and believe these sites are doing themselves a grave disservice. Q: Implementing personalization can seem daunting. What's the simplest way for a retailer to get started A: Rama: Take a look at your competitors. I guarantee that your bestin-class peers are already using personalization, so show these sites to your boss. Then dip your toes in the water. On your product detail page, start by incorporating a simple recommendation zone showing products that customers may also like. By recommending productssimilar to the product they're looking at, you've quickly given them more choices and paths forward to purchase. It doesn't have to be all or nothing -- test gradually. Try recommendations on a few product detail pages, not your entire site. Start small and, if something doesn't work or your strategy needs further development, you can revert your site back quickly. After you've experienced success with early personalization efforts, you can test predictive search, which gives two people searching for the same thing different search results. The possibilities of personalization are endless.PERSONALIZATION IN SHOPPINGQ: Why is personalization necessary in today's ecommerce environmentBy helping shoppers quickly find the productscommercecloud|10Q: What's your advice for merchants who are nervous to give up control to AI A: Rama: Most product recommendation systems -- including Commerce Cloud -- let merchants override or constrain recommendations when needed. For example, you could tell the system not to recommend certain products or brands. You'll find that the rules are flexible enough to allay any concerns, so don't let these qualms sway you from trying personalization. You'd be leaving money on the table if you don't give it a try. Q: What's the future of AI in retail A: Rama: When personalization becomes so woven into the fabric of the shopping experience that the shopper doesn't realize he or she is being personalized to, that's when we have arrived. The future of AI will be that seamless, pervasive impact on every shopper journey. Over the next few years, I anticipate a world where two different visitors to an ecommerce site receive completely different and tailored experiences, from personalized navigation and personalized pricing to personalized promotions. As long as shoppers have any sort of prior behavior we can learn from, their experience will be unique.Q: How can retailers change their mindsets to prepare for this kind of technology A: Rama: It might sound crazy coming from a data scientist, but abandon the thought that you must collect a lot of data first. You likely already have enough data to enact some powerful personalization. So you just have to dive in, and your efforts will become smarter over time.TrailblazerSTORYStonewall Kitchen Fuels Growth with Einstein, No Data Scientist Required Stonewall Kitchen was born in 1991 at a farmers market in Portsmouth, New Hampshire. Founders Jim Stott and Jonathan King created a jam reminiscent of Jim's grandmother's blueberry pie. Since then, Stonewall has grown from producing a few dozen jars for the weekly farmers market to 75,000 jars daily. Specialty and grocery stores nationwide sell Stonewall products, from salsas to pancake mixes, and Stonewall runs 10 company stores. With online personalization, Stonewall is now delighting the palates of more customers than ever before.When a customer visits a Stonewall store, they are greeted by an associate who recommends products just for them. Stonewall's team knew AI could bring this experience online. Ian Marquis, ecommerce manager, says, AI is a feature we knew would be increasingly important in ecommerce. We knew personalization was an important part of our future. With a five-person ecommerce team, Commerce Cloud Einstein helps Stonewall power 1-to-1 personalization without hours of manual merchandising or segmentation. Ian explains, No one knows our brand better than we do, but Einstein enables digital depth, variety, and novelty of experience that we couldn't deliver consistently otherwise.。。。。。。