Dɛn Ne Assortment Optimization?

Assortment optimization ne nteɛsoɔ a ɛdi saa nsonsonoeɛ yi ho dwuma. Ɛde nea Asafo ti no si gyinae no bata nea adetɔfo hu ankasa wɔ shelf - so denam data, adesua a ɛkɔ so, ne store-level execution so. Saa akwankyerɛ yi ka nea ɛyɛ, nea enti a akwan dodow no ara di nkogu, sɛnea wɔde bedi dwuma, ne sɛnea wɔsusuw nea efi mu ba no ho asɛm.

Assortment Optimization vs. Assortment Nhyehyɛe: Nsonsonoe bɛn na ɛwɔ mu?

Wɔtaa de nsɛmfua abien yi di dwuma de sesa wɔn ho wɔn ho. Wɔkyerɛkyerɛ akwan horow a ɛsono emu biara titiriw mu.

Nhyehyɛe kyerɛ sɛnea ɛsɛ sɛ wo assortment no yɛ. Optimization hwɛ hu sɛ ɛyɛ - ankasa na ɛkɔ so nya nkɔso bere a tebea horow sesa no.

Nneɛma Abiɛsa a Wɔsi Assortment Gyinaesi Na Wɔyera

Adetɔnfo dodow no ara de sika kɛse hyɛ ɔfa a edi kan no mu. Nsonsonoe akɛse a ɛwɔ adwumayɛ mu no te abien a aka no mu.

Strategic Layer: Nea Wɔtɔn

Eyi ne baabi a category-level gyinaesi ahorow si: nneɛma bɛn na wonya shelf space, sɛnea private label kari pɛ wɔ ɔman no brands ho, ne dwuma a category biara di wɔ store nhyehyɛe no nyinaa mu. Wɔsi gyinae wɔ ha wɔ Asafo ti no mu, a gua so nsɛm ne akansi mu nsusuwii na ɛkanyan, na ɛsakra wɔ kyinhyia tenten mu.

Asiane no: data a wɔaboaboa ano no kata mpɔtam hɔ nsakrae so. Ebia ade bi a ɔman no tɔn a wogye tom no nyɛ adwuma yiye wɔ sotɔɔ ahorow 40% mu na ɛnyɛ adwuma boro so wɔ 30% foforo mu. Sɛ wɔkyekyem pɛpɛɛpɛ a, wɔde sɛnkyerɛnne no sie.

Tactical Layer: Baabi ne Sɛnea Wɔtɔn No

Tactical layer no kyerɛ strategy ase kɔ beae-nhyehyɛe pɔtee mu: store clustering, planogram design, ne merchandising mmara. Eyi ne baabi a assortment bɛyɛ local ankasa - high-density urban store wɔ soronko space anohyeto, nantew akwantu nhyehyɛe, ne adetɔfo asɛmpatrɛw sen suburban format.

Asiane no: gyinaesi ahorow a ɛwɔ saa gyinabea yi da so ara gyina nsusuwii ahorow so kɛse sen sɛ wɔde wɔn ho bɛto store-gyinabea nsɛnkyerɛnne ankasa so. Assortments betumi ayɛ sɛ ɛyɛ papa-localized wɔ krataa so bere a ɛda so ara yɛ nea ɛnteɛ wɔ adeyɛ mu.

Operational Layer: Nea Ɛduru Adetɔfoɔ no nkyɛn Ankasa

Eyi ne baabi a assortment optimization di nkonim anaasɛ edi nkogu komm. Adwumayɛ fã no da honam fam nokwasɛm a adetɔfo hyia adi: nneɛma bɛn na ɛwɔ shelf so, sɛ ebia wɔyɛ planogram ahorow no yiye, sɛ ebia wotumi hu nsɛm a wɔde hyɛ nkurɔfo nkuran, ne sɛ ebia wɔkyere stockouts na wosiesie ntɛmntɛm.

Sɛ enni real-bere visibility mu store execution, upstream gyinaesi biara yɛ fã bi asusuw. Mfiridwuma ho nimdeɛ te sɛɛlɛtrɔnik shelf nkyerɛwde ahorowna wɔde IoT sensor ahorow di dwuma kɛse de to saa visibility gap yi mu - a ɛkyere shelf tebea ahorow no ara sen sɛ wɔde wɔn ho bɛto nsaano akontaabu a ɛntaa nsi dodo sɛ wobetumi ayɛ ho biribi so.

Nea Enti a Amanneɛ kwan so Assortment Optimization Di Nkonim

Wɔayɛ assortment akwan dodow no ara yiye-wɔ krataa so. Ɛha na wɔsɛe wɔ nneyɛe mu.

Failure Mode 1: Abakɔsɛm mu Data Yɛ Nea Ɛyɛ Paara ma Bere a Atwam no

Adetɔn abakɔsɛm kyerɛ wo nea adetɔfo tɔɔ wɔ tebea horow a na ɛwɔ hɔ saa bere no - ne assortment a na ɛwɔ hɔ, wɔ bo a wɔde sii hɔ no mu. Entumi nkyerɛ wo nea adetɔfo pɛ nanso wɔantumi anhu. Wɔ akuw a ɛkɔ ntɛmntɛm-mu no, bere a su bi bɛda adi pefee wɔ abakɔsɛm data mu no, mfɛnsere a wɔde bɛyɛ ade no taa atwam dedaw.

Failure Mode 2: Gyinaesi a Wɔde Ahyɛ Mfinimfini, Mpɔtam Hɔ Nokwasɛm

Sɛ wɔyɛ assortment gyinaesi nyinaa wɔ adwumayɛbea ti a, store-level nuance nya averaged away. Wobetumi ayi ade a ɔman no tɔn a ɛnyɛ papa nanso ɛyɛ adwuma denneennen wɔ sotɔɔ ahorow pɔtee bi mu no afi hɔ. A standardized planogram gets deployed wɔ sotɔɔ ahorow a ɛsono kɛse shelf nsusuwii ne adetɔfo dodow.

Failure Mode 3: Data Silos Ma Gyinaesi a Enni Mu Ba

Adetɔn ahyehyɛde ahorow yɛ data wɔ nhyehyɛe ahorow pii - point-a ɛfa-adetɔn, nneɛma a wɔakora so, nokwaredi, e-aguadi, ne in-store sensor ahorow so. Category managers yɛ adwuma fi data set biako mu. Supply chain adwuma fi foforo hɔ. Store adwumayɛ fi nkyem abiɛsa mu biako. Saa adwene ahorow yi mu biara nni hɔ a edi mũ, na gyinaesi ahorow a wosi fi silo biako biara mu no bɛma ɔhaw ahorow a wotumi hu wɔ foforo mu nkutoo aba.

Failure Mode 4: Planogram Compliance no Ba fam Sen sɛnea Asafo ti no Susuw

Planogram de mfaso ma sɛ wɔyɛ no yiye na wɔyɛ no daa nkutoo a. Wɔ aguadidan dodow no ara mu no, ɛsono sɛnea wodi mmara so kɛse wɔ sotɔɔ ahorow - mu na mpɛn pii no Asafo ti no nnim kosi sɛ wɔbɛsusuw. Sɛ woresusuw ade bi shelf adwumayɛ ho a egyina adetɔn ho nsɛm so, nanso saa ade no akɔ bay gyinabea a ɛnteɛ wɔ wo sotɔɔ ahorow 20% mu asram abiɛsa a, wo adwumayɛ ho nsɛm no nyɛ nea wotumi de ho to so. Nteaseɛsɛnea wɔtaa yɛ shelf data foforono bata sɛnea saa susuw ahorow yi yɛ pɛpɛɛpɛ no ho tẽẽ.

Failure Mode 5: Omnichannel Nsɛnkyerɛnne no Kɔ a Wɔnkenkan

Intanɛt so adetɔfo nneyɛe yɛ nneɛma ahorow ho nyansa a ɛyɛ fɛ a honam fam adetɔnfo dodow no ara bu wɔn ani gu so. Zero-results searches wɔ wo e-commerce platform so kyerɛ wo nea adetɔfoɔ rehwehwɛ a wo nkura. High-browse, low-adetɔ nhyehyɛe da ahwehwɛde a ebia ɛbɛhwehwɛ wɔ-store nhwehwɛmu ansa na wɔasakra adi. Adetɔfoɔ a ɔhwehwɛ adeɛ bi wɔ intanɛt so, ɔhunu sɛ ɛnni hɔ, na ɔfiri hɔ no nnya data biara wɔ in-store system - nanso saa data a enni hɔ no ankasa yɛ sɛnkyerɛnne, sɛ wokyekyere dwumadie no sɛ wobɛkyere a. Mfiaseɛ ne sɛ wode wo intanɛt so hwehwɛ ne browse data bɛka wo category planning adwumayɛ ho, mpo wɔ ɔkwan a ɛnyɛ ɔkwan pa so.

Sɛnea AI Ma Assortment Gyinaesi Tu Tu mpɔn

Manual assortment management wɔ sotɔɔ ɔhaha pii ne SKU mpempem du du mu adu anohyeto a ɛwɔ nea spreadsheets ne bere ne bere mu nhwehwɛmu betumi aboa. AI boa wɔ akwan pɔtee bi a wotumi susuw so.

Store-level ahwehwɛdeɛ nkɔmhyɛ.Amanneɛ kwan so nkɔmhyɛ yɛ adwuma wɔ banner anaa cluster level. Mfiri adesua nhwɛsoɔ tumi ma nkɔmhyɛ wɔ ankorankoro sotɔɔ ne SKU gyinabea, bu akontaa fa mpɔtam hɔ nneɛma - mpɔtam hɔ nnipa dodoɔ, akansi a ɛbɛn, mmerɛ mu micro-nkɔsoɔ - a nhwɛsoɔ a ɛtrɛ no yɛ average away. Saa granularity yi ne nea ɛma localized assortment gyinaesi ahorow yɛ defensible mmom sen sɛ wobesusuw.

SKU ntease a wɔde ma.Ɛnyɛ ade biara na ɛma wonya ne kwan. AI nhwɛsoɔ tumi kyerɛ SKU ahodoɔ a ɛredi shelf real estate ne inventory capital a enni mfasoɔ a ɛfata - a ɛbu akontaa fa margin ntoboa, substitution effects, ne basket impact ho. Nsonsonoe a ɛho hia no ne brɛoo-movers a wɔsom niche nokwaredi ne slow-movers a wɔnyɛ adwuma yiye kɛkɛ ntam. AI tumi kyerɛ nsonsonoe a ɛda abien no ntam wɔ nsenia a nsaano nhwehwɛmu ntumi nyɛ so.

Dynamic bo a wɔbɔ ne nkɔso a ɛne ne ho hyia.Assortment gyinaesi ahorow nni hɔ a atew ne ho afi bo a wɔbɔ ho. AI-a wɔde di dwumabo a wɔbɔ a ɛyɛ nnambetumi de nkɔso dwumadi ne assortment adwumayɛ ahyia wɔ bere ankasa mu - a ɛtew nea wɔayɛ ho nhyehyɛe ne nea adetɔfo bua ankasa wɔ shelf level ntam a ɛnhyia.

Nnipakum a wɔhwɛ wɔn so.Kɔmputa so anisoadehu ne sensor data betumi ahu planogram deviations a enhia sɛ wɔyɛ nsaano akontaabu a edi mũ. Nkɔso a aba wɔshelf label mfiridwuma ho nimdeɛama automated shelf-state monitoring ayɛ nea ɛyɛ mmerɛw kɛse ma mfinimfini-size adetɔnfo, ɛnyɛ nkɔnsɔnkɔnsɔn akɛse nkutoo.

Anamɔn Anum-Nhyehyɛe a Wɔde Di Dwuma

Adetɔnfo dodow no ara nim sɛ assortment optimization ho hia. Kakraa bi na wɔwɔ mfiase a emu da hɔ. Wɔayɛ saa nhyehyɛe yi sɛnea ɛbɛyɛ a wobetumi de adi dwuma wɔ nsenia biara mu.

Anamɔn 1: Audit Wo Mprempren Assortment

Ansa na wobɛma biribiara ayɛ papa no, fa nnyinasosɛm a ɛyɛ nokware si hɔ. Dɛn ne wo mprempren stockout rate sɛnea category ne store te? SKU ahorow bɛn na ɛrema decile a ɛwɔ ase a wɔtɔn wɔ anammɔn ahinanan biara mu no aba? Ɛhe na nsonsonoe kɛse a ɛda nneɛma ahorow a wɔayɛ ho nhyehyɛe ne shelf a ɛwɔ hɔ ankasa ntam? Sɛ wo ntumi mfa data a wotumi de ho to so mmua saa nsɛmmisa yi a, ɛno ankasa ne ade a ɛho hia sen biara a woahu - ne sɛnkyerɛnne a ɛkyerɛ sɛ wode sika bɛto nea wotumi hu mu ansa na wode sika ahyɛ nnwinnade a ɛma nneɛma yɛ yiye mu. A nhyehyɛemfitiaseɛ ROI akontabuobetumi aboa ma wɔakyerɛ baabi a nkɛntɛnso mu nsonsonoe a ɛkorɔn-wɔ ansa na wɔde wɔn ho ahyɛ ɔkwan biara so.

Anamɔn 2: Kyerɛkyerɛ Wo Store Clusters mu

Ɛnyɛ sotɔɔ ahorow nyinaa na ɛsɛ sɛ wɔde nneɛma ahorow koro kɔ, nanso nneɛma ahorow a ɛyɛ soronko koraa ma sotɔɔ biara no yɛ nea wontumi nni so wɔ adwumayɛ mu. Store clustering bridges saa extremes yi denam mmeae a wɔde akuwakuw a ɛwɔ ahwehwɛde profile ahorow a ɛte sɛ nea ntease wom no so. Wɔde clustering a etu mpɔn asi adetɔ suban ankasa - basket composition, category velocity, shopper mission patterns - a ɛnyɛ nnipa dodow a wosusuw sɛ ɛyɛ so. Adetɔnfo dodow no ara de akuw anan kosi awotwe na ɛyɛ adwuma, a egyina network kɛse ne format ahorow so. Nɔma a ɛfata ne nea ɛsono sɛnea akuw biara yɛ n’ade wɔ ɔkwan soronko so ankasa a ɛbɛma wɔatumi ayɛ nneɛma ho nhyehyɛe soronko bi.

Anamɔn 3: Fa Wo Data Fibea ahorow no Bom

Assortment optimization yɛ papa te sɛ data a ɛma no aduan no nkutoo. Anyɛ yiye koraa no, wuhia SKU-level adetɔn data denam sotɔɔ biara so a anyɛ yiye koraa no ɛwɔ asram 12 abakɔsɛm, mprempren nneɛma a wɔakora so, ne nsusuwii bi a ɛkyerɛ sɛnea shelf wɔ hɔ. Asɛmmisa a ɛfa sɛdeɛ wɔkyere shelf data - sɛ ɛnam nsaano amanneɛbɔ, ESL nhyehyɛeɛ, anaa IoT sensor - so nya data foforɔ ne ahotosoɔ so nkɛntɛnsoɔ tẽẽ. Ntease a ɛwɔnkitahodi akwan a wɔfa so kyere shelf datayɛ gyinaesi a mfaso wɔ so a wosisi ntɛm. Data nkabom a ɛyɛ pɛ nyɛ ade a ɛsɛ sɛ wodi kan yɛ ansa na woafi ase - nanso ɛsɛ sɛ wote wo data no mu nsonsonoe ne latency ase ansa na woagye ne output adi.

Anamɔn 4: Set Optimization Mmara ne Guardrails

AI models ne optimization algorithms hia anohyeto ahorow. Ɛnyɛ sɛ ɛsɛ sɛ wɔde wɔn ankasa si gyinae biara. Kyerɛkyerɛ pefee gyinaesi ahorow a ebetumi akɔ so ankasa - te sɛ replenishment triggers ma high-velocity SKUs - ne nea ɛhwehwɛ sɛ nnipa hwɛ mu, te sɛ delisting a product from a cluster. Guardrails nso bɔ wɔn ho ban fi mfomso a automated systems di bere a data nwie pɛyɛ no ho. Nhwɛso a wɔtaa de di dwuma: algorithm kamfo kyerɛ sɛ wonyi ade bi fi hɔ efisɛ ne tɔn sua, bere a nea ɛde ba ankasa ne stockouts a ɛkɔ so daa a adetɔn ho nsɛm no nkyerɛ nsonsonoe a ɛda nea wɔhwehwɛ a ɛba fam no ntam.Bo ne nea ɛwɔ hɔ a wɔda no adi mfomsoyɛ adwumayɛ mu huammɔdi kwan a ɛfa ho a ɛfata sɛ yɛte ase ansa na wɔde automation aba.

Anamɔn 5: Sua, Sua, na San Yɛ

Assortment optimization yɛ adeyɛ a ɛkɔ so, ɛnyɛ bere biako-dwuma. Fa nhwehwɛmu a ɛyɛ daa - si hɔ bosome mmiɛnsa biara anyɛ yie koraa no ma gyinaesie a ɛfa akwankyerɛ ho, bosome biara ma nsakraeɛ a ɛfa ɔkwan a wɔfa so yɛ adwuma ho. Yɛ nhyehyeɛ a ɛfa nsɛm a wɔde ma ho nhama wɔ mfimfini ɔfa akuo ntam na sie-level adwumayɛ data. Fa nhyehyeɛ kyinhyia biara di dwuma sɛ sɔhwɛ: yɛ nsusuwii hunu, fa nsakrae bi di dwuma, susuw nea ebefi mu aba, fa saa adesua no di dwuma wɔ kyinhyia a edi hɔ no mu. Ahyehyɛde ahorow a wonya mfaso kɛse fi saa nhyehyɛe yi mu no nyɛ wɔn a wɔwɔ nnwinnade a ɛyɛ nwonwa sen biara. Wɔyɛ wɔn a wɔanya su sɛ wobesua biribi afi data mu bere nyinaa.

KPI ahorow nsia a wɔde susuw Assortment Optimization ho

Di metrics asia no nyinaa akyi wɔ store level, ɛnyɛ sɛ wɔaka abom kɛkɛ. Network-level averages taa de sotɔɔ ahorow a ɔhaw ahorow no mu yɛ den kɛse - ne baabi a optimization hokwan akɛse wɔ hɔ no sie.

Assortment Optimization Wɔ Intanɛt ne Honam fam akwan horow so

Wɔ adetɔnfo a wɔyɛ adwuma wɔ honam fam ne dijitaal akwan horow so fam no, wontumi nni gyinaesi ahorow a ɛfa nneɛma ahorow ho ho dwuma wɔ ɔkwan a ɛyɛ soronko so.Adetɔnbea ahorow noasesa: adetɔfo tu fa akwan horow ntam ntɛmntɛm, na nsɛm a efi akwan biara so no betumi ama wɔahu gyinaesi ahorow wɔ ɔkwan foforo no so.

Online sɛ assortment sɛnkyerɛnne.Zero-results searches wɔ wo e-commerce platform so yɛ tẽẽ kyerɛ sɛ assortment gaps - adetɔfoɔ ka kyerɛ wo deɛ wɔpɛ pɛpɛɛpɛ a wo nkura. High-browse, low-adetɔ nhyehyɛe betumi akyerɛ nneɛma a adetɔfo pɛ sɛ wɔn ankasa hwehwɛ mu ansa na wɔatɔ, a ɛwɔ nkyerɛkyerɛmu ma in-store ranging. Sɛnea kyerɛ noMcKinsey nhwehwɛmu, mprempren adetɔfo bɛboro 70% hwɛ kwan sɛ wobenya osuahu ahorow a wɔayɛ ama wɔn ankasa - akwanhwɛ a ɛfa nneɛma a ɛwɔ hɔ ho te sɛ nkitahodi ho.

Nkabom vs. nsonsonoe ahorow assortment.Sɛ ebia ɛsɛ sɛ wo intanɛt ne in{0}}store assortments no hyia a, egyina wo store format ne adetɔfo suban so. Assortment a wɔaka abom ma adwumayɛ yɛ mmerɛw na ɛma ahwehwɛde ho data a ɛho tew ba, nanso ɛhyɛ honam fam sotɔɔ ahorow ma wɔde intanɛt so nhoma a wɔahyehyɛ a ɛyɛ den a format dodow no ara ntumi nnye no kɔ. Ɔkwan a ɛsono - a honam fam sotɔɔ ahorow de curated, high-velocity core bere a intanɛt kwan no di dua tenten - ho dwuma no yɛ adwuma yiye bere a akwan abien no som adetɔ asɛmpatrɛw ahorow a ɛsono ankasa. Gyinaesi nhyehyɛe no yɛ mmerɛw: sɛ adetɔfo taa hwehwɛ wɔ intanɛt so na wɔdan kɔ-store mu a, alignment ho hia. Sɛ intanɛt ne sotɔɔ mu adetɔfoɔ yɛ atiefoɔ a wɔda nsow kɛseɛ a, nsonsonoeɛ bɛtumi ayɛ yie.

Baabi a ɛsɛ sɛ wufi ase.Ade a ɛyɛ adwuma sen biara a wobɛfa so akɔ mu ne sɛ wode wo e-commerce zero-results search data bɛka wo category nhyehyɛe nhwehwɛmu no ho. Mfiridwuma foforo biara nhia - ɔsram biara hwehwɛ nsɛmmisa a entumi nyɛ yiye a category managers ahwɛ mu a wɔde kɔ amannɔne no betumi ada assortment gaps a ɛwɔ-store adetɔn data mu no adi da. Pairing eyi neshelf-level data a wɔkyere no yiyewɔ honam fam sotɔɔ ahorow mu no yɛ ɔkwan a wɔatoto mu wɔ intanɛt so nsɛnkyerɛnne ne in-store execution ntam.

Nea Eyi Te Wɔ Nneyɛe Mu

Nsɛm a edidi so yi kyerɛ sɛnea assortment optimization nnyinasosɛm ahorow no di dwuma wɔ aguadidan ahorow mu. Eyinom yɛ nhwɛso ahorow a ɛyɛ mfatoho, ɛnyɛ adwumakuw pɔtee bi a wɔayɛ ho nhwehwɛmu.

Grocery: mpɔtam hɔ ahwehwɛde masking wɔ aggregate data mu.Ɔmantam bi a wɔtɔn nneɛma wɔ ɔkwan a ɛyɛ nwonwa so de aggregate category data di dwuma de yɛ assortments ho nhyehyɛe. Mmusuakuw aduan akuw - wɔn a wɔyɛ adwuma denneennen wɔ mpɔtam pɔtee bi - no nni hɔ bere nyinaa efisɛ wɔn adetɔn no yɛ mmerɛw bere a wɔabobɔw akɔ frankaa no so no. Ɔkwan a egyina cluster- so a wɔkyekyee wɔ basket composition ankasa so da no adi sɛ nea na ɛte sɛ category ahwehwɛde a ɛba fam wɔ store akuw bi mu no yɛ structural data aggregation problem mmom. Sɛ wɔyɛ nsakrae wɔ saa sotɔɔ ahorow no nsusuwso ahorow mu ma ɛda mpɔtam hɔ adetɔ nneyɛe adi a, ɛto nsonsonoe no mu. Ade a ɛma wotumi yɛ adwuma no nyɛ mfiridwuma foforo - ɛyɛ disaggregating demand data denam store so sen sɛ wɔde banner bɛyɛ. Nneɛma a wotumi hu yiye denam nnwinnade te sɛɛlɛtrɔnik shelf nkyerɛwde a ɛwɔ sotɔɔ ahorow a wɔtɔn nnuan mufoa susuw a ɛkɔ so sɛ ebia wɔreyɛ saa assortments a wɔayɛ nsakrae no ankasa no so.

Fashion: tenten-dua SKU sohwɛ.Ntade titiriw tɔnfo bi de SKU mpempem pii a ɛyɛ nnam kɔ ​​bere biara mu. Nhwehwɛmu a wɔyɛe wɔ adwumayɛ mu da no adi sɛ, nneɛma a wɔde yɛ adwuma no fã kɛse bi ma wonya sika a wonya no mu kyɛfa ketewaa bi a ɛnsɛ bere a wɔde nhyehyɛe, nneɛma a wɔakora so, ne nneɛma a wɔde bɛhyɛ mu ma no di dwuma no. Nhwehwɛmu no tetew akuw abien a wɔnyɛ adwuma yiye mu: SKU ahorow a wonni adetɔfo anokwafo a wobetumi ahu wɔn ne baabi a enye-to-gyinabea ntoboa, ne SKU ahorow a wɔn dodow nyinaa sua nanso wɔsan tɔ nneɛma pii wɔ adetɔfo fã pɔtee bi mu. Wɔde kuw a edi kan no fi hɔ nkakrankakra. Wɔde nea ɛto so abien no sie a wɔayɛ nsakrae wɔ baabi a wɔkyekyɛ mu. Nea afi mu aba ne tighter range a ɛyɛ mmerɛw sɛ wobedi na ɛnyɛ den sɛ ɛbɛma gyinaesi ɔbrɛ aba wɔ shelf level.

Convenience retail: kum ahoɔhare sɛ nsonsonoe.Nkɔnsɔnkɔnsɔn ketewa-format convenience chain yɛ adwuma wɔ mmeae a square foot biara yɛ kɛse-stakes na stockout ho ka yɛ kɛse denam inventory buffers a ɛba fam so. Adeɛ a ɛto ano hyeɛ no nyɛ assortment nhyehyɛeɛ - ɛyɛ berɛ a ɛda stockout a ɛrekɔ so ne store associate bi a ɔbua no ntam. Saa nsonsonoe no a wɔbɛtew so denam automated shelf monitoring so, sen sɛ wɔde wɔn ho bɛto nsaano nhwehwɛmu a wɔayɛ ho nhyehyɛe so no, ɛwɔ nkɛntɛnso tẽẽ na wotumi susuw wɔ in-store a ɛwɔ hɔ ma high-margin impulse categories.

Nsɛm a Wɔtaa Bisa

Dɛn ne assortment optimization wɔ retail mu?

Assortment optimization yɛ adeyɛ a ɛkɔ so paw na wɔtew nneɛma a wɔde afrafra a wɔde ma wɔ sotɔɔ biara mu no mu na ama wɔatɔn, mfaso a wonya, ne adetɔfo akomatɔyam kɛse. Nea ɛnte sɛ bere biako-bere mu nsɛso nhyehyɛe no, ɛka bere ankasa -bere data ne adwumayɛ mu nhwehwɛmu a ɛkɔ so bom ma ɛma nneɛma a wɔpaw no ne ahwehwɛde ankasa hyia.

Nsonsonoe bɛn na ɛda assortment nhyehyɛe ne assortment optimization ntam?

Assortment nhyehyeɛ yɛ berɛ ne berɛ mu, a ɛwɔ mfimfini - a ɛtaa yɛ mmerɛ anaa afe biara - a ɛkyerɛ nneɛma a ɛsɛ sɛ wɔde kɔ a egyina abakɔsɛm mu nsɛm so. Assortment optimization yɛ nea ɛkɔ so. Ɛde real-bere nsɛnkyerɛnne ka ho na ɛkora-level adwumayɛ data de sesa assortment no bere a tebea horow sesa. Nhyehyɛe na ɛde akwankyerɛ a edi kan no si hɔ; optimization ma ɛkɔ so yɛ calibrate.

Ɔkwan bɛn so na AI ma assortment optimization tu mpɔn?

AI ma store-level demand forecasting a ɛkɔ akyiri sen cluster averages no tumi yɛ adwuma, ɛkyerɛ SKU ahorow a ɛnyɛ adwuma yiye bere a ɛrebu akontaa wɔ substitution effects ho, ɛma planogram nyansahyɛ ahorow a egyina mprempren adetɔn ahoɔhare so, na ɛyɛ real-bere sɛnkyerɛnne - wim tebea, mpɔtam hɔ nsɛm a esisi, akansifo dwumadi - a nsaano nhyehyɛe kyinhyia ahorow ntumi mfa nka ho wɔ bere mu mfa nyɛ ho adwuma.

Dɛn ne nneɛma a ɛtaa ma assortment optimization di nkogu?

Nneɛma anum a ɛtaa di huammɔ: wɔde wɔn ho to abakɔsɛm mu data a entumi nnye mprempren ahwehwɛde so dodo -; gyinaesi a ɛwɔ mfinimfini-si a ɛyera mpɔtam hɔ nsakrae; siloed data nhyehyɛe ahorow a ɛma mfonini a enni mũ ba; planogram a wodi so no sua sen sɛnea Asafo ti no susuw; ne sɛ wɔamfa intanɛt so ahwehwɛde nsɛnkyerɛnne a ɛda nsonsonoe a wontumi nhu wɔ in-store adetɔn data nkutoo mu anhyɛ mu.

KPI bɛn na ɛsɛ sɛ midi akyi ma assortment optimization?

Metrics a mfasoɔ wɔ so paa ne stockout rate, sell-through rate, SKU productivity (sika a wɔnya anaa margin wɔ unit of shelf space biara mu), planogram compliance rate, category margin contribution, ne cluster demand alignment (nsonsonoeɛ a ɛda nhyehyɛeɛ a wɔayɛ ne sell ankasa-through wɔ cluster level). Di eyinom nyinaa akyi wɔ sotɔɔ mu, ɛnyɛ sɛ wɔaka abom kɛkɛ.

Bere tenten ahe na egye sɛ wɔde di dwuma?

Mpɛn pii no, wobetumi ayɛ mfitiaseɛ akontabuo ne cluster{0}}based optimization framework wɔ asram kakraa bi mu denam data a ɛwɔ hɔ dada so. AI-driven continuous optimization a ɛyɛ nwonwa kɛse hwehwɛ sɛ wonya data fapem a ɛyɛ den na ebetumi agye asram 12 kosi 18 ansa na ayɛ adwuma koraa. Ɛkame ayɛ sɛ bere nyinaa sɛ wofi ase fi akontaabu no so a, ɛda nkonimdi ahorow a wobetumi anya ntɛmntɛm adi ansa na mfiridwuma foforo biara ho ahia.

So aguadifo nketewa betumi anya mfaso afi assortment optimization mu?

Aane. Nnyinasosɛm no di dwuma ɛmfa ho sɛ scale - nteaseɛ a ɛfa nneɛma bɛn na ɛma wɔn atenaeɛ, di stockout mpɛn dodoɔ a wɔdi akyi, ne feedback loops a wɔkyekyere wɔ adetɔn data ne adeɛ ho gyinaesie ntam no yɛ nea nteaseɛ wɔ mu ma adwumayɛ kɛseɛ biara. Ebia adetɔnfo nketewa renhia adwumayɛbea AI nhyehyɛe ahorow; free anaa low-cost analytics nnwinnade betumi aboa optimization a mfaso wɔ so a egyina data a wɔwɔ dedaw so. Paw a wɔpaw nonifa so shelf label ano aduruyɛ mfiase biako a mfaso wɔ so a wɔde bɛma data a wɔkyere atu mpɔn a wɔmfa sika kɛse nsie wɔ nnwuma mu.

Data bɛn na ɛsɛ sɛ mifi ase?

Anyɛ yie koraa no: SKU-level adetɔn data by store a anyɛ yie koraa no asram 12 abakɔsɛm, mprempren inventory levels, ne shelf a ɛwɔ hɔ ho susudua bi - mpo nsaano stockout amanneɛbɔ. Efi saa fapem yi so no, wobɛtumi ayɛ akontabuo a nteaseɛ wom, ahunu wo hokwan a ɛkorɔn-nsunsuansoɔ, na woayɛ data nkɔsoɔ kwankyerɛ. Data a edi mũ nyɛ ade a ɛsɛ sɛ wodi kan yɛ. Optimization a mfaso wɔ so betumi aba wɔ data a ɛnyɛ pɛ mu, bere tenten a wote ase na wubu ne nsonsonoe ho akontaa no.

Baabi a Wobefi Ase

Assortment optimization de boɔ a ɛkyɛn so ma berɛ a ɛyɛ adwuma sɛ loop a ɛkɔ so - hwehwɛ adwumayɛ mu, siesie afiri afrafra no, yɛ adwuma wɔ-store mu, susuw nea ɛfiri mu ba, na ɛsan yɛ bio. Ɛnyɛ nea adetɔnfo a wɔkyekye saa tumi yi yiye sen biara no ne wɔn a wodi kan de wɔn sika hyɛ nnwinnade a ɛkɔ akyiri sen biara mu. Wɔyɛ wɔn a wɔde nokware data a ɛfa baabi a wɔn mprempren assortment no di nkogu ho na efi ase, na wɔkyekye ahyehyɛde no su ahorow a wɔde bɛyɛ adwuma wɔ saa data no so bere nyinaa.

Sɛ worehyɛ aseɛ firi mfitiaseɛ a, nneyɛeɛ anan yɛ adwuma ntɛm ara: yɛ stockout ne SKU adwumayɛ ho nhwehwɛmu denam data a wowɔ dedaw so; hwɛ wo sotɔɔ akuakuo nkyerɛaseɛ mu tia adetɔ suban ankasa sene sɛ wobɛfa nnipa dodoɔ a wɔsusu sɛ ɛwɔ hɔ; fa wo e-commerce zero-aba hwehwɛ data no bata wo category nhyehyeɛ adwumayɛ kwan no ho; na kyerɛkyerɛ assortment gyinaesi ahorow a ɛsɛ sɛ wɔyɛ no automated sen sɛnea onipa hwɛ mu ansa na wɔakum no.

Wobetumi ayɛ eyinom mu biara ansa na wɔatɔ mfiridwuma foforo biara - na emu biara bɛma wɔahu baabi a mfiridwuma mu sika a wɔde bɛto mu no bɛma ade no akɔ ankasa.