零售营销成功的5种技术解决方案

© Oliver WymanINTRODUCTIONImagine a world where technology has taken over most human activity in retailmanagement, speeding up the innovation cycle and providing customers a fullypersonalized, multi-touchpoint shopping experience. It is easy to dream aboutthe future. However, it is harder is to estimate the pace of change, distinguish betweenfads and trends, and make the right technology bets. That is why it is importantto understand what technology could mean for the heart of retail:merchandisingand sourcing.The effective use of technology is crucial now, as macro-economic and geopolitical forcesdisrupt supply chains and drive inflation, with significant impact on customers’budgets.As a result, the industry is faced with a mix shift towards lower priced, lower marginalternatives, challenging the overall economics of retailers. Meanwhile, service retailersface added pressure from the continued growth of discounters and rising costs in areassuch as labor and energy. Technology could mitigate — at least in part — these risks.Our worldwide experience in consulting and developing software for retailers suggestsfive critical areas of focus for technology in merchandising andsourcing.Exhibit 1: Five critical areas of focus for technology in merchandising and sourcingPriorities1Further automate themerchandising levers2Bring an E2E cost view tooptimise the business3Look at your business througha customer lens4Build a high-techinnovation function5Leverage sourcing scaleand get more for lessSource: Oliver Wyman AnalysisDreaming about the future is easy. It is harder estimating thepace of change and making the right technology bets Why do this?•Make better decisions•Free up time and resource by automating•Reduce business complexity and operating cost•Win-win with customers by passing on savings•Drive growth with profitablecustomers•Increase loyalty•Differentiate! Offer products customers cannot getanywhere else•Best possible prices in branded•Amazing product and price in OL © Oliver WymanDemystifying the role of AILet us clarify the role of AI. Some may have the impression that AI could soon fully automateretailer's headquarters. We acknowledge that Artificial Intelligence (AI) has the potentialto revolutionize the retail industry's approach to merchandising and sourcing. However,the reality of a completely AI-driven merchandising department is still far off in the future.Instead, retailers must strive to become AI-compatible, and start applying it in select areas.We believe that retailers should prioritize incorporating structured analytics anddeterministic models, as ample room for improvement still exists. In addition, structuredanalytics provide deeper insights into the underlying causes and drivers of decisions, whichis essential for successful adoption as these decisions have significant impact on a retailer'seconomics. Commercial directors are unlikely to rely solely on “black-box AI answers” whenmaking critical business decisions.However, there are applications where AI offers value already now. AI offers immediatevalue in areas where vast amounts of unstructured data are available, the “why” is lessimportant, and a human filter is available to make final decisions with intuition.Examplesof activities that AI excels at include identifying high-risk customers for churn, conductinga first run of product matching, or discovering innovative product ideasglobally.Exhibit 2: Examples of structured analytics and AI use caseTopicExample structured analytics use casePricing•Deterministic model of price elasticityand volume switching•Competitor price indicesPromo•Deterministic model of ROI: Uplift,cannibalization, etc•Supplier funding scenariosRanging•Deterministic model of rande switchingand optimization•Supplier cost-price and range scenarionsSourcing•Analysis to get to 4-net costs•Should cost models, and analyzingvolumeeffectsLifecyle•Understanding WHY customersare churning and producingmitigation strategiesSource: Oliver Wyman Analysis Example AI use case•Forecasting competitor reactionto price changes•More sophisticated elasticity models•Product matching•Forecasting consumer reaction to promomechanic, timing, communication•Finding novel promo ideas from aroundthe world — trawling muchunstructured data•Creative range recommendations,which learn•Finding novel product ideas from aroundthe world, trawling much unstructured data•Broad data analysis to predict upcomingsupplier cost increases•Predicting customers at risk of churn © Oliver WymanPriority 1FURTHER AUTOMATE THE MERCHANDISING LEVERSWith varying success, retailers are on a journey to “escape from excel-catraz”. They areadopting task and decision automation, incorporating advanced analytics, and graduallyshifting towards a more autonomous system. Traditional retailers — including those thathave made progress in automation — still have “humans in the loop”. This means that humanactivity is still essential to run core processes, creating a barrier to scaling a