[{"data":1,"prerenderedAt":981},["ShallowReactive",2],{"content-query-6VbgYtPaxS":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"heading":10,"prompt":11,"tags":15,"files":17,"nav":6,"presets":18,"gallery":36,"body":38,"_type":974,"_id":975,"_source":976,"_file":977,"_stem":978,"_extension":979,"sitemap":980},"/tools/customer-segmentation","tools",false,"","Customer Segmentation Tool","Segment customers online from Excel or CSV data. Run RFM analysis, clustering, and profile high-value or at-risk groups with AI.","Customer Segmentation",{"prefix":12,"label":13,"placeholder":14},"Segment customers","Describe your customer data and segmentation goal","e.g. 800 customers with purchase frequency, recency, and total spend; RFM segmentation into Champions, Loyal, At Risk, Lost; scatter plot colored by segment; segment profiles with mean metrics; size of each segment",[16],"statistics",true,[19,25,30],{"label":20,"prompt":21,"dataset_url":22,"dataset_title":23,"dataset_citation":24},"RFM Customer Segmentation","RFM segmentation: compute Recency (days since last purchase), Frequency (total orders), Monetary (total spend) for each customer; score 1–5 on each dimension; assign segments (Champions, Loyal, At Risk, Hibernating, Lost); scatter plot of Frequency vs Monetary colored by segment; segment size and mean metrics table","https://data.cdc.gov/api/views/iuq5-y9ct/rows.csv?accessType=DOWNLOAD","NHANES Mental Health Assessment","CDC",{"label":26,"prompt":27,"dataset_url":28,"dataset_title":29,"dataset_citation":24},"K-Means Behavioral Clustering","K-means clustering (k=4) on customer behavioral features (sessions, pages_viewed, purchase_count, avg_order_value); use elbow method to validate k; scatter plot of first 2 PCA components colored by cluster; cluster profile table with mean feature values; name each cluster based on behavior","https://data.cdc.gov/api/views/dppn-5tm3/rows.csv?accessType=DOWNLOAD","NCHS Health and Nutrition Examination Survey",{"label":31,"prompt":32,"dataset_url":33,"dataset_title":34,"dataset_citation":35},"Health Program Participant Segments","Segment program participants by engagement and health outcome: cluster on visit_frequency, adherence_rate, outcome_score; k-means with k=3; PCA scatter plot colored by cluster; identify high-engagement/high-outcome vs low-engagement/low-outcome clusters; compare segment sizes and mean outcomes","https://ourworldindata.org/grapher/happiness-cantril-ladder.csv","Self-Reported Life Satisfaction","Our World in Data",[37],"/img/tools/customer-segmentation.png",{"type":39,"children":40,"toc":963},"root",[41,50,90,121,126,132,198,204,212,392,400,488,493,499,649,655,812,818,872,878,915,921,931,953],{"type":42,"tag":43,"props":44,"children":46},"element","h2",{"id":45},"what-is-customer-segmentation",[47],{"type":48,"value":49},"text","What Is Customer Segmentation?",{"type":42,"tag":51,"props":52,"children":53},"p",{},[54,60,62,67,69,74,76,81,83,88],{"type":42,"tag":55,"props":56,"children":57},"strong",{},[58],{"type":48,"value":59},"Customer segmentation",{"type":48,"value":61}," divides a customer base into distinct groups — segments — whose members share similar characteristics, behaviors, or needs. Rather than treating all customers identically, segmentation enables tailored strategies: different retention campaigns for at-risk customers, different upsell offers for champions, different reactivation messages for dormant accounts. The most widely used data-driven segmentation framework is ",{"type":42,"tag":55,"props":63,"children":64},{},[65],{"type":48,"value":66},"RFM analysis",{"type":48,"value":68},": classifying customers by ",{"type":42,"tag":55,"props":70,"children":71},{},[72],{"type":48,"value":73},"Recency",{"type":48,"value":75}," (how recently they purchased), ",{"type":42,"tag":55,"props":77,"children":78},{},[79],{"type":48,"value":80},"Frequency",{"type":48,"value":82}," (how often they purchase), and ",{"type":42,"tag":55,"props":84,"children":85},{},[86],{"type":48,"value":87},"Monetary value",{"type":48,"value":89}," (how much they spend). These three dimensions capture the most predictive behavioral signals for customer lifetime value and churn risk without requiring complex modeling.",{"type":42,"tag":51,"props":91,"children":92},{},[93,98,100,105,107,112,114,119],{"type":42,"tag":55,"props":94,"children":95},{},[96],{"type":48,"value":97},"K-means clustering",{"type":48,"value":99}," provides a complementary algorithmic approach when the segmentation dimensions extend beyond RFM or when you want the data to reveal its own natural groupings. K-means partitions customers into k clusters by minimizing within-cluster variance on the input features. The ",{"type":42,"tag":55,"props":101,"children":102},{},[103],{"type":48,"value":104},"elbow method",{"type":48,"value":106}," (plotting within-cluster sum of squares vs k) or ",{"type":42,"tag":55,"props":108,"children":109},{},[110],{"type":48,"value":111},"silhouette score",{"type":48,"value":113}," helps choose the optimal k. For datasets with many features, ",{"type":42,"tag":55,"props":115,"children":116},{},[117],{"type":48,"value":118},"PCA",{"type":48,"value":120}," (principal component analysis) reduces dimensionality before clustering, improving cluster separation and enabling 2D visualization where each point is a customer and its position reflects the two strongest axes of variation in the data.",{"type":42,"tag":51,"props":122,"children":123},{},[124],{"type":48,"value":125},"A concrete example: an e-commerce company applies RFM scoring to 800 customers. Champions (high R, F, M: 180 customers, 22% of base) generate 48% of revenue and have a 12-month retention rate of 91%. \"At Risk\" customers (low R, high F, M: 210 customers) have stopped purchasing recently despite historically high engagement — they are the highest-priority reactivation target. \"Lost/Inactive\" customers (low on all three: 170 customers) have not purchased in over 12 months and have low lifetime value; a lightweight win-back email costs little and reactivates ~5% of this group. The frequency vs monetary scatter plot clearly separates the four segments into distinct quadrants.",{"type":42,"tag":43,"props":127,"children":129},{"id":128},"how-it-works",[130],{"type":48,"value":131},"How It Works",{"type":42,"tag":133,"props":134,"children":135},"ol",{},[136,147,163],{"type":42,"tag":137,"props":138,"children":139},"li",{},[140,145],{"type":42,"tag":55,"props":141,"children":142},{},[143],{"type":48,"value":144},"Upload your data",{"type":48,"value":146}," — provide a CSV or Excel file with one row per customer and columns for purchase history metrics (recency, frequency, spend) or behavioral features (sessions, pages viewed, product categories browsed).",{"type":42,"tag":137,"props":148,"children":149},{},[150,155,157],{"type":42,"tag":55,"props":151,"children":152},{},[153],{"type":48,"value":154},"Describe the segmentation",{"type":48,"value":156}," — e.g. ",{"type":42,"tag":158,"props":159,"children":160},"em",{},[161],{"type":48,"value":162},"\"RFM segmentation into 4 groups; scatter plot frequency vs monetary colored by segment; mean metrics per segment; identify Champions and At Risk customers\"",{"type":42,"tag":137,"props":164,"children":165},{},[166,171,173,180,182,188,190,196],{"type":42,"tag":55,"props":167,"children":168},{},[169],{"type":48,"value":170},"Get full results",{"type":48,"value":172}," — the AI writes Python code using ",{"type":42,"tag":174,"props":175,"children":177},"a",{"href":176},"https://pandas.pydata.org/",[178],{"type":48,"value":179},"pandas",{"type":48,"value":181},", ",{"type":42,"tag":174,"props":183,"children":185},{"href":184},"https://scikit-learn.org/",[186],{"type":48,"value":187},"scikit-learn",{"type":48,"value":189},", and ",{"type":42,"tag":174,"props":191,"children":193},{"href":192},"https://plotly.com/python/",[194],{"type":48,"value":195},"Plotly",{"type":48,"value":197}," to score and segment customers, produce the scatter visualization with shaded quadrant regions, and compute per-segment summary statistics",{"type":42,"tag":43,"props":199,"children":201},{"id":200},"required-data-format",[202],{"type":48,"value":203},"Required Data Format",{"type":42,"tag":51,"props":205,"children":206},{},[207],{"type":42,"tag":55,"props":208,"children":209},{},[210],{"type":48,"value":211},"For RFM segmentation:",{"type":42,"tag":213,"props":214,"children":215},"table",{},[216,240],{"type":42,"tag":217,"props":218,"children":219},"thead",{},[220],{"type":42,"tag":221,"props":222,"children":223},"tr",{},[224,230,235],{"type":42,"tag":225,"props":226,"children":227},"th",{},[228],{"type":48,"value":229},"Column",{"type":42,"tag":225,"props":231,"children":232},{},[233],{"type":48,"value":234},"Description",{"type":42,"tag":225,"props":236,"children":237},{},[238],{"type":48,"value":239},"Example",{"type":42,"tag":241,"props":242,"children":243},"tbody",{},[244,272,312,352],{"type":42,"tag":221,"props":245,"children":246},{},[247,258,263],{"type":42,"tag":248,"props":249,"children":250},"td",{},[251],{"type":42,"tag":252,"props":253,"children":255},"code",{"className":254},[],[256],{"type":48,"value":257},"customer_id",{"type":42,"tag":248,"props":259,"children":260},{},[261],{"type":48,"value":262},"Unique identifier",{"type":42,"tag":248,"props":264,"children":265},{},[266],{"type":42,"tag":252,"props":267,"children":269},{"className":268},[],[270],{"type":48,"value":271},"C1234",{"type":42,"tag":221,"props":273,"children":274},{},[275,284,289],{"type":42,"tag":248,"props":276,"children":277},{},[278],{"type":42,"tag":252,"props":279,"children":281},{"className":280},[],[282],{"type":48,"value":283},"recency_days",{"type":42,"tag":248,"props":285,"children":286},{},[287],{"type":48,"value":288},"Days since last purchase",{"type":42,"tag":248,"props":290,"children":291},{},[292,298,299,305,306],{"type":42,"tag":252,"props":293,"children":295},{"className":294},[],[296],{"type":48,"value":297},"12",{"type":48,"value":181},{"type":42,"tag":252,"props":300,"children":302},{"className":301},[],[303],{"type":48,"value":304},"180",{"type":48,"value":181},{"type":42,"tag":252,"props":307,"children":309},{"className":308},[],[310],{"type":48,"value":311},"420",{"type":42,"tag":221,"props":313,"children":314},{},[315,324,329],{"type":42,"tag":248,"props":316,"children":317},{},[318],{"type":42,"tag":252,"props":319,"children":321},{"className":320},[],[322],{"type":48,"value":323},"frequency",{"type":42,"tag":248,"props":325,"children":326},{},[327],{"type":48,"value":328},"Total number of orders",{"type":42,"tag":248,"props":330,"children":331},{},[332,338,339,345,346],{"type":42,"tag":252,"props":333,"children":335},{"className":334},[],[336],{"type":48,"value":337},"8",{"type":48,"value":181},{"type":42,"tag":252,"props":340,"children":342},{"className":341},[],[343],{"type":48,"value":344},"2",{"type":48,"value":181},{"type":42,"tag":252,"props":347,"children":349},{"className":348},[],[350],{"type":48,"value":351},"45",{"type":42,"tag":221,"props":353,"children":354},{},[355,364,369],{"type":42,"tag":248,"props":356,"children":357},{},[358],{"type":42,"tag":252,"props":359,"children":361},{"className":360},[],[362],{"type":48,"value":363},"monetary",{"type":42,"tag":248,"props":365,"children":366},{},[367],{"type":48,"value":368},"Total spend",{"type":42,"tag":248,"props":370,"children":371},{},[372,378,379,385,386],{"type":42,"tag":252,"props":373,"children":375},{"className":374},[],[376],{"type":48,"value":377},"1240",{"type":48,"value":181},{"type":42,"tag":252,"props":380,"children":382},{"className":381},[],[383],{"type":48,"value":384},"89",{"type":48,"value":181},{"type":42,"tag":252,"props":387,"children":389},{"className":388},[],[390],{"type":48,"value":391},"8500",{"type":42,"tag":51,"props":393,"children":394},{},[395],{"type":42,"tag":55,"props":396,"children":397},{},[398],{"type":48,"value":399},"For k-means / behavioral clustering:",{"type":42,"tag":213,"props":401,"children":402},{},[403,421],{"type":42,"tag":217,"props":404,"children":405},{},[406],{"type":42,"tag":221,"props":407,"children":408},{},[409,413,417],{"type":42,"tag":225,"props":410,"children":411},{},[412],{"type":48,"value":229},{"type":42,"tag":225,"props":414,"children":415},{},[416],{"type":48,"value":234},{"type":42,"tag":225,"props":418,"children":419},{},[420],{"type":48,"value":239},{"type":42,"tag":241,"props":422,"children":423},{},[424,447],{"type":42,"tag":221,"props":425,"children":426},{},[427,435,439],{"type":42,"tag":248,"props":428,"children":429},{},[430],{"type":42,"tag":252,"props":431,"children":433},{"className":432},[],[434],{"type":48,"value":257},{"type":42,"tag":248,"props":436,"children":437},{},[438],{"type":48,"value":262},{"type":42,"tag":248,"props":440,"children":441},{},[442],{"type":42,"tag":252,"props":443,"children":445},{"className":444},[],[446],{"type":48,"value":271},{"type":42,"tag":221,"props":448,"children":449},{},[450,467,472],{"type":42,"tag":248,"props":451,"children":452},{},[453,459,461],{"type":42,"tag":252,"props":454,"children":456},{"className":455},[],[457],{"type":48,"value":458},"feature_1",{"type":48,"value":460}," … ",{"type":42,"tag":252,"props":462,"children":464},{"className":463},[],[465],{"type":48,"value":466},"feature_n",{"type":42,"tag":248,"props":468,"children":469},{},[470],{"type":48,"value":471},"Any numeric behavioral features",{"type":42,"tag":248,"props":473,"children":474},{},[475,481,482],{"type":42,"tag":252,"props":476,"children":478},{"className":477},[],[479],{"type":48,"value":480},"sessions",{"type":48,"value":181},{"type":42,"tag":252,"props":483,"children":485},{"className":484},[],[486],{"type":48,"value":487},"avg_order_value",{"type":42,"tag":51,"props":489,"children":490},{},[491],{"type":48,"value":492},"Alternatively, provide raw transaction-level data and ask the AI to compute RFM metrics per customer before segmenting.",{"type":42,"tag":43,"props":494,"children":496},{"id":495},"interpreting-the-results",[497],{"type":48,"value":498},"Interpreting the Results",{"type":42,"tag":213,"props":500,"children":501},{},[502,518],{"type":42,"tag":217,"props":503,"children":504},{},[505],{"type":42,"tag":221,"props":506,"children":507},{},[508,513],{"type":42,"tag":225,"props":509,"children":510},{},[511],{"type":48,"value":512},"Output",{"type":42,"tag":225,"props":514,"children":515},{},[516],{"type":48,"value":517},"What it means",{"type":42,"tag":241,"props":519,"children":520},{},[521,537,553,569,585,601,617,633],{"type":42,"tag":221,"props":522,"children":523},{},[524,532],{"type":42,"tag":248,"props":525,"children":526},{},[527],{"type":42,"tag":55,"props":528,"children":529},{},[530],{"type":48,"value":531},"Segment assignment",{"type":42,"tag":248,"props":533,"children":534},{},[535],{"type":48,"value":536},"Which segment each customer belongs to",{"type":42,"tag":221,"props":538,"children":539},{},[540,548],{"type":42,"tag":248,"props":541,"children":542},{},[543],{"type":42,"tag":55,"props":544,"children":545},{},[546],{"type":48,"value":547},"Scatter plot",{"type":42,"tag":248,"props":549,"children":550},{},[551],{"type":48,"value":552},"Customers plotted on two key dimensions; color = segment; shaded regions show quadrant boundaries",{"type":42,"tag":221,"props":554,"children":555},{},[556,564],{"type":42,"tag":248,"props":557,"children":558},{},[559],{"type":42,"tag":55,"props":560,"children":561},{},[562],{"type":48,"value":563},"Segment size",{"type":42,"tag":248,"props":565,"children":566},{},[567],{"type":48,"value":568},"Number and % of customers in each segment — informs campaign reach",{"type":42,"tag":221,"props":570,"children":571},{},[572,580],{"type":42,"tag":248,"props":573,"children":574},{},[575],{"type":42,"tag":55,"props":576,"children":577},{},[578],{"type":48,"value":579},"Segment profiles",{"type":42,"tag":248,"props":581,"children":582},{},[583],{"type":48,"value":584},"Mean recency, frequency, monetary (or feature means) per segment — characterizes each group",{"type":42,"tag":221,"props":586,"children":587},{},[588,596],{"type":42,"tag":248,"props":589,"children":590},{},[591],{"type":42,"tag":55,"props":592,"children":593},{},[594],{"type":48,"value":595},"Revenue concentration",{"type":42,"tag":248,"props":597,"children":598},{},[599],{"type":48,"value":600},"% of total revenue attributable to each segment — typically Champions generate 40–60%",{"type":42,"tag":221,"props":602,"children":603},{},[604,612],{"type":42,"tag":248,"props":605,"children":606},{},[607],{"type":42,"tag":55,"props":608,"children":609},{},[610],{"type":48,"value":611},"RFM scores",{"type":42,"tag":248,"props":613,"children":614},{},[615],{"type":48,"value":616},"1–5 scores on each dimension; combined score drives segment assignment",{"type":42,"tag":221,"props":618,"children":619},{},[620,628],{"type":42,"tag":248,"props":621,"children":622},{},[623],{"type":42,"tag":55,"props":624,"children":625},{},[626],{"type":48,"value":627},"Elbow/silhouette plot",{"type":42,"tag":248,"props":629,"children":630},{},[631],{"type":48,"value":632},"For k-means: guides the choice of optimal number of clusters",{"type":42,"tag":221,"props":634,"children":635},{},[636,644],{"type":42,"tag":248,"props":637,"children":638},{},[639],{"type":42,"tag":55,"props":640,"children":641},{},[642],{"type":48,"value":643},"PCA scatter",{"type":42,"tag":248,"props":645,"children":646},{},[647],{"type":48,"value":648},"2D projection of all features; each point = customer; cluster boundaries visible",{"type":42,"tag":43,"props":650,"children":652},{"id":651},"example-prompts",[653],{"type":48,"value":654},"Example Prompts",{"type":42,"tag":213,"props":656,"children":657},{},[658,674],{"type":42,"tag":217,"props":659,"children":660},{},[661],{"type":42,"tag":221,"props":662,"children":663},{},[664,669],{"type":42,"tag":225,"props":665,"children":666},{},[667],{"type":48,"value":668},"Scenario",{"type":42,"tag":225,"props":670,"children":671},{},[672],{"type":48,"value":673},"What to type",{"type":42,"tag":241,"props":675,"children":676},{},[677,694,710,727,744,761,778,795],{"type":42,"tag":221,"props":678,"children":679},{},[680,685],{"type":42,"tag":248,"props":681,"children":682},{},[683],{"type":48,"value":684},"RFM segmentation",{"type":42,"tag":248,"props":686,"children":687},{},[688],{"type":42,"tag":252,"props":689,"children":691},{"className":690},[],[692],{"type":48,"value":693},"RFM segmentation; score R, F, M 1–5; assign segments; scatter F vs M colored by segment; segment table",{"type":42,"tag":221,"props":695,"children":696},{},[697,701],{"type":42,"tag":248,"props":698,"children":699},{},[700],{"type":48,"value":97},{"type":42,"tag":248,"props":702,"children":703},{},[704],{"type":42,"tag":252,"props":705,"children":707},{"className":706},[],[708],{"type":48,"value":709},"k-means on 5 behavioral features; elbow method for optimal k; PCA scatter colored by cluster; cluster profiles",{"type":42,"tag":221,"props":711,"children":712},{},[713,718],{"type":42,"tag":248,"props":714,"children":715},{},[716],{"type":48,"value":717},"Segment revenue share",{"type":42,"tag":248,"props":719,"children":720},{},[721],{"type":42,"tag":252,"props":722,"children":724},{"className":723},[],[725],{"type":48,"value":726},"what % of revenue comes from each segment? bar chart of revenue by segment",{"type":42,"tag":221,"props":728,"children":729},{},[730,735],{"type":42,"tag":248,"props":731,"children":732},{},[733],{"type":48,"value":734},"At-risk identification",{"type":42,"tag":248,"props":736,"children":737},{},[738],{"type":42,"tag":252,"props":739,"children":741},{"className":740},[],[742],{"type":48,"value":743},"which customers moved from Champions to At Risk in the last 6 months? list them with their RFM scores",{"type":42,"tag":221,"props":745,"children":746},{},[747,752],{"type":42,"tag":248,"props":748,"children":749},{},[750],{"type":48,"value":751},"Segment over time",{"type":42,"tag":248,"props":753,"children":754},{},[755],{"type":42,"tag":252,"props":756,"children":758},{"className":757},[],[759],{"type":48,"value":760},"compute RFM segments for Q1 and Q2; how many customers migrated between segments? Sankey diagram",{"type":42,"tag":221,"props":762,"children":763},{},[764,769],{"type":42,"tag":248,"props":765,"children":766},{},[767],{"type":48,"value":768},"High-value threshold",{"type":42,"tag":248,"props":770,"children":771},{},[772],{"type":42,"tag":252,"props":773,"children":775},{"className":774},[],[776],{"type":48,"value":777},"what recency/frequency/spend thresholds define the top 20% of customers by LTV?",{"type":42,"tag":221,"props":779,"children":780},{},[781,786],{"type":42,"tag":248,"props":782,"children":783},{},[784],{"type":48,"value":785},"Churn prediction",{"type":42,"tag":248,"props":787,"children":788},{},[789],{"type":42,"tag":252,"props":790,"children":792},{"className":791},[],[793],{"type":48,"value":794},"which customers have not purchased in 90+ days but previously bought monthly? flag for reactivation",{"type":42,"tag":221,"props":796,"children":797},{},[798,803],{"type":42,"tag":248,"props":799,"children":800},{},[801],{"type":48,"value":802},"Geographic segments",{"type":42,"tag":248,"props":804,"children":805},{},[806],{"type":42,"tag":252,"props":807,"children":809},{"className":808},[],[810],{"type":48,"value":811},"compute RFM segments by region; compare segment distribution across regions; heatmap",{"type":42,"tag":43,"props":813,"children":815},{"id":814},"assumptions-to-check",[816],{"type":48,"value":817},"Assumptions to Check",{"type":42,"tag":819,"props":820,"children":821},"ul",{},[822,832,842,852,862],{"type":42,"tag":137,"props":823,"children":824},{},[825,830],{"type":42,"tag":55,"props":826,"children":827},{},[828],{"type":48,"value":829},"Data completeness",{"type":48,"value":831}," — RFM analysis requires complete transaction history; if historical data is truncated (only last 12 months available but some customers were acquired 3 years ago), frequency and monetary scores will underestimate long-tenure customers; use a consistent observation window for all customers",{"type":42,"tag":137,"props":833,"children":834},{},[835,840],{"type":42,"tag":55,"props":836,"children":837},{},[838],{"type":48,"value":839},"Scaling before clustering",{"type":48,"value":841}," — k-means is sensitive to feature scale; always standardize (z-score) features before clustering so that a feature measured in dollars does not dominate features measured in sessions; describe whether to standardize",{"type":42,"tag":137,"props":843,"children":844},{},[845,850],{"type":42,"tag":55,"props":846,"children":847},{},[848],{"type":48,"value":849},"Outlier customers",{"type":48,"value":851}," — very high-spend or very high-frequency customers will distort k-means cluster centers; consider capping extreme outliers (e.g., at the 99th percentile) or treating them as a separate \"VIP\" segment before running clustering",{"type":42,"tag":137,"props":853,"children":854},{},[855,860],{"type":42,"tag":55,"props":856,"children":857},{},[858],{"type":48,"value":859},"Segment stability",{"type":48,"value":861}," — k-means clusters are sensitive to initialization and can vary between runs; run multiple initializations (k-means++) to ensure stability; verify that the segment profiles are interpretable and consistent before acting on them",{"type":42,"tag":137,"props":863,"children":864},{},[865,870],{"type":42,"tag":55,"props":866,"children":867},{},[868],{"type":48,"value":869},"RFM score cut-points",{"type":48,"value":871}," — quintile-based RFM scoring (top 20% = 5) is robust but can create arbitrary boundaries; customers just above or below a quintile boundary are nearly identical but receive different scores; consider using percentile ranges or fuzzy membership",{"type":42,"tag":43,"props":873,"children":875},{"id":874},"related-tools",[876],{"type":48,"value":877},"Related Tools",{"type":42,"tag":51,"props":879,"children":880},{},[881,883,889,891,897,899,905,907,913],{"type":48,"value":882},"Use the ",{"type":42,"tag":174,"props":884,"children":886},{"href":885},"/tools/cohort-retention",[887],{"type":48,"value":888},"Cohort Retention Analysis",{"type":48,"value":890}," tool to track how segment membership evolves over time — do Champions retain at 90% per month while At Risk customers churn at 30%? Use the ",{"type":42,"tag":174,"props":892,"children":894},{"href":893},"/tools/lead-scoring-model",[895],{"type":48,"value":896},"Lead Scoring Model",{"type":48,"value":898}," to score new prospects (before their first purchase) based on acquisition channel and initial behavior, analogous to the way RFM scores existing customers. Use the ",{"type":42,"tag":174,"props":900,"children":902},{"href":901},"/tools/pca",[903],{"type":48,"value":904},"PCA — Principal Component Analysis",{"type":48,"value":906}," tool to reduce high-dimensional customer feature sets before clustering — PCA identifies the most informative axes of variation for visualization. Use the ",{"type":42,"tag":174,"props":908,"children":910},{"href":909},"/tools/ab-test-calculator",[911],{"type":48,"value":912},"A/B Test Calculator",{"type":48,"value":914}," to test whether a targeted campaign sent to an \"At Risk\" segment significantly improved re-purchase rates compared to a control group.",{"type":42,"tag":43,"props":916,"children":918},{"id":917},"frequently-asked-questions",[919],{"type":48,"value":920},"Frequently Asked Questions",{"type":42,"tag":51,"props":922,"children":923},{},[924,929],{"type":42,"tag":55,"props":925,"children":926},{},[927],{"type":48,"value":928},"How many segments should I create?",{"type":48,"value":930},"\nFor RFM-based segmentation, 4–6 actionable segments is the practical sweet spot — enough to differentiate strategies (Champions vs Loyal vs At Risk vs Lost) without creating so many segments that marketing cannot act on each one differently. For k-means, let the elbow method guide k, but validate that each cluster has a distinct, interpretable profile and sufficient size to be actionable (at least 5% of the customer base). A common mistake is over-segmenting: 12 clusters where only 3–4 have meaningfully different profiles add complexity without insight.",{"type":42,"tag":51,"props":932,"children":933},{},[934,939,944,946,951],{"type":42,"tag":55,"props":935,"children":936},{},[937],{"type":48,"value":938},"What is the difference between RFM segmentation and k-means clustering?",{"type":42,"tag":55,"props":940,"children":941},{},[942],{"type":48,"value":943},"RFM",{"type":48,"value":945}," uses expert-defined rules (score recency, frequency, and monetary on 1–5 scales, then assign named segments by score combinations) — it is interpretable, deterministic, and produces segments with clear business meaning. ",{"type":42,"tag":55,"props":947,"children":948},{},[949],{"type":48,"value":950},"K-means",{"type":48,"value":952}," is data-driven — it finds the natural clusters in whatever features you provide, without requiring prior knowledge of what the segments should look like. RFM is better when you have a clear framework and want predictable, stable segments. K-means is better when you have many features beyond R/F/M and want the data to reveal unexpected groupings. In practice, RFM is most common for retention marketing; k-means is used for product personalization or when behavioral features are richer.",{"type":42,"tag":51,"props":954,"children":955},{},[956,961],{"type":42,"tag":55,"props":957,"children":958},{},[959],{"type":48,"value":960},"My Champions segment is tiny (\u003C 5% of customers) — is that normal?",{"type":48,"value":962},"\nYes — RFM Champions (high on all three dimensions) typically represent 10–25% of customers but generate a disproportionate share of revenue (40–60%). If your Champions segment is very small, check the quintile cut-points: are they too strict? Consider using top-30% thresholds instead of top-20%. If truly only 2–3% of customers qualify as Champions, it may reflect a business model with low repeat purchase rates (one-time buyers) where frequency and recency scores are uniformly low — in which case, segment primarily on monetary value and recency rather than frequency.",{"title":7,"searchDepth":964,"depth":964,"links":965},2,[966,967,968,969,970,971,972,973],{"id":45,"depth":964,"text":49},{"id":128,"depth":964,"text":131},{"id":200,"depth":964,"text":203},{"id":495,"depth":964,"text":498},{"id":651,"depth":964,"text":654},{"id":814,"depth":964,"text":817},{"id":874,"depth":964,"text":877},{"id":917,"depth":964,"text":920},"markdown","content:tools:088.customer-segmentation.md","content","tools/088.customer-segmentation.md","tools/088.customer-segmentation","md",{"loc":4},1775502475444]