LATAM Project Study · Shopper augmentation and incremental value · Version 2

Do Shopper Surveys Add Predictive Value?

A matched-shop study of shopper segments, detailed information domains, temporal linkage, and incremental predictive value.

Latin American micro-retailExpanded segment comparison
2,595shopper interviews
1,048linked shops
4shopper segments compared
4.01%protected-test RMSE gain
Abstract

Testing whether shopper voices improve a shop-level model

This study evaluates whether shopper survey information adds predictive value beyond a completed shopkeeper-only model of relative commercial activity. A conservative temporal linkage rule matched shoppers to shops using the same probable interviewer code, same calendar date, an absolute time gap of no more than 90 minutes, and exactly one eligible shopkeeper anchor. The rule linked 1,883 of 2,595 shopper interviews to 1,048 shops. Shopper responses were transformed and aggregated to the shop level. A supplementary comparison first compressed shoppers into four behavioral segments: Deliberate Planners, Impulse Explorers, Loyal Optimizers, and Overspenders. On 841 linked development shops, adding three nonredundant segment shares improved repeated-cross-validation RMSE from 0.7700 to 0.7676 and R² from 0.2634 to 0.2680. The richer 30-feature store-choice and desired-capabilities domain performed better, reaching RMSE 0.7610 and R² 0.2806; adding segments to that domain changed performance only marginally to RMSE 0.7604 and R² 0.2818. In the original governed evaluation, the parsimonious 109-feature Elastic Net—79 shopkeeper predictors plus those 30 store-choice predictors—was locked before the protected test. On 207 matched test shops, it reduced MAE by 4.69%, reduced RMSE by 4.01%, and increased R² from 0.2154 to 0.2771. Shopper segments therefore provide a compact predictive summary, while detailed store-choice information remains the stronger and better-validated shopper augmentation.

1

Research question and incremental design

Does aggregated shopper information improve prediction beyond shopkeeper information when both models are trained and evaluated on exactly the same shops?

The incremental design matters. Comparing a 1,291-shop shopkeeper model with a smaller linked-shop model would mix feature value with sample composition. This study therefore rebuilt a matched shopkeeper benchmark on the 841 linked development shops and compared every shopper-augmented candidate on shared outer and inner folds.

The shopper domains are also plausible in the regional retail setting. A Colombian neighborhood-store bulletin connects store scale with assortment, shopping frequency, and ticket patterns,[3] while a Brazilian bakery-sector study emphasizes customer experience, portfolio diversification, digital channels, and loyalty as competitive concerns.[4] These external sources provide context only; the incremental-value claim is estimated exclusively from the matched project sample.

Interpretation boundary

The estimand is predictive improvement among shops with accepted shopper links. It is not the causal effect of shopper preferences, a representative estimate for every retailer, or proof that adding a digital capability will raise commercial activity.

2

Linking shoppers to shops

The apparent store code was identified as a probable interviewer or fieldwork code. Because interviewers could visit several shops in one day, code-only linkage was not credible. Timing and uniqueness were therefore required.

Figure 1
Conservative shopper-to-shop linkage funnel2,595shopper interviews1,883uniquely linkedshoppers1,048shops with linkedshoppers841development shops207protected test shops
The linkage rule was locked before shopper feature engineering and did not use the commercial-activity outcome or predictive performance.
Linkage outcomeShopper interviewsShare
Uniquely linked1,88372.56%
Ambiguous within 90 minutes250.96%
Same-date candidate outside window34413.26%
No same-interviewer/date candidate34313.22%
Locked linkage: same probable interviewer code, same date, absolute gap ≤90 minutes, and one eligible shopkeeper anchor. No nearest-neighbor or sequential override was allowed.

Coverage was broad but not complete: 81.18% of shops had at least one linked shopper. Linked shops had a median of two shoppers and a maximum of five, which limits the precision of shop-level shopper averages.

3

Shopper feature engineering

The shopper questionnaire contained 214 source columns organized into 57 question families. Administrative fields, linkage variables, conditional free text, ambiguous storage fields, and current transaction spending were excluded from the primary model. The spending field was reserved because it could behave as a contemporaneous proxy for the outcome.

Figure 2
From questionnaire storage to locked shopper predictors214source columns136prepared shopperresponses219shop-level aggregates174eligible shopperfeatures30selected domainfeatures
Category rules and feature eligibility were learned from linked development shops only. Protected-test categories did not define feature construction.
Shopper domainCandidate featuresEligible featuresRole in final model
Current trip and purchase behavior7668Compared
General attitudes and psychographics5427Not selected
Shopper and household composition2724Not selected
Store choice and desired capabilities3330Selected
Store experience and interaction2925Not selected

Numeric responses and ratings were averaged within shop. Multiple-response indicators became shop-level selection proportions. Categorical responses became proportions after rare development categories were pooled and one reference category was omitted. Documented refusal codes became analytic missingness and were median-imputed inside training folds.

Selected information

Alternative places considered, reasons for choosing the store, and the importance of payment or digital capabilities.

Excluded from the final augmentation

Broad psychographics, household composition, general store ratings, and many current-purchase indicators did not add stable enough signal for their complexity.

4

Shopper segments as a compact alternative

A supplementary analysis asked whether four interpretable shopper profiles could summarize useful shopper information with fewer inputs than the detailed questionnaire domains.

Each linked shopper was assigned to one of four mutually exclusive profiles: Deliberate Planners, Impulse Explorers, Loyal Optimizers, or Overspenders. Individual memberships were averaged within shop, producing segment shares such as 0, 0.5, or 1. Because the four shares sum to one, Loyal Optimizers was retained as the reference and the other three shares entered the model. The comparison used the same 841 linked development shops, fixed Elastic Net settings of alpha = 0.1L1 ratio = 0.1, identical 5-fold cross-validation splits repeated five times, and the same preprocessing for every specification.

Repeated-CV specificationShopper inputsRMSEMAER²Result
Shopkeeper only00.77000.61250.2634Reference
Shopkeeper + segment shares30.76760.60870.2680Better than shopkeeper alone
Shopkeeper + store choice300.76100.60510.2806Most of the shopper gain
Shopkeeper + store choice + segments330.76040.60380.2818Small additional gain
Figure 3
Repeated-CV RMSE: segments improve the baseline, store choice adds more0.7600.7650.770Shopkeeper only0.7700+ segment shares0.7676+ store choice0.7610+ store choice + segments0.7604
Lower RMSE is better. Segment shares improve the shopkeeper-only benchmark, but the detailed store-choice domain produces the larger gain. Adding segments after store choice lowers RMSE by only 0.0006.

The segment-only model reduced RMSE by 0.31% and MAE by 0.62% relative to shopkeeper information alone, confirming that the profiles retain some predictive content. The store-choice domain reduced RMSE by 1.17% and MAE by 1.21% in the same comparison. The combined specification was numerically best, but its improvement beyond store choice alone was very small: 0.0006 RMSE, 0.0013 MAE, and 0.0012 R².

Interpretation: segmentation compresses shopper behavior into a useful three-variable summary. The 30 store-choice features preserve more detail and remain the principal shopper-information block. The supplementary segment comparison was performed after the original model-selection exercise, so it is reported as exploratory development evidence and does not replace the locked protected-test result.
5

Matched evaluation design

Five shared outer folds estimated performance, while four inner folds selected Elastic Net settings. Every model saw the same rows. Shopkeeper preprocessing was preserved; all shopper features were numeric shop-level means or proportions and received training-fold median imputation followed by standardization.

841matched development shops
5 × 4outer × inner folds
1,920inner Elastic Net fits

The full comparison included a mean baseline, matched shopkeeper model, shopper-only model, five shopkeeper-plus-domain models, and the complete 253-feature augmented model. The protected test outcome remained unavailable until one fixed specification was recorded.

6

Development evidence and model lock

Fixed specificationFeaturesOOF MAEOOF RMSEOOF R²RMSE change vs shopkeeper
Shopkeeper only790.60070.76050.2902reference
+ Current trip and purchase1470.59780.75780.2952+0.35%
+ Store choice and capabilities1090.59130.75170.3066+1.16%
+ All shopper domains2530.59520.75320.3037+0.95%
Figure 4
Fixed-setting development OOF RMSEStore choice + capabilities0.752All shopper domains0.753Current trip + purchase0.758Shopkeeper only0.760
Lower is better. The 30-feature store-choice block gave the strongest fixed-setting result with far fewer shopper inputs than the complete block.

The fixed store-choice model improved pooled development MAE by 1.57% and RMSE by 1.16%, with lower RMSE in four of five folds. Development bootstrap intervals crossed zero, so the evidence was treated as promising rather than conclusive.

Why the smaller model was selected: the full shopper block used 144 additional shopper predictors beyond the selected domain but performed slightly worse. Parsimony reduced questionnaire burden and the risk of fitting noise.

The locked candidate contained 109 predictors and fixed Elastic Net settings of alpha = 0.1L1 ratio = 0.1.

7

Protected-test incremental value

4.69%MAE improvement
4.01%RMSE improvement
+0.0617R² change
Protected-test metricMatched shopkeeperAugmentedChange
MAE (log target)0.68940.65714.69% lower
RMSE (log target)0.85390.81974.01% lower
R²0.21540.2771+0.0617
Pearson correlation0.46450.5271+0.0626
Spearman correlation0.40980.4877+0.0779
Median multiplicative error factor1.77341.73002.45% lower
Figure 5
Protected test: matched shopkeeper vs augmented model0.6890.657MAE0.8540.820RMSE0.2150.277R²ComparisonSelected model
For MAE and RMSE, lower is better; for R², higher is better. The augmented model improved all three metrics on the locked 207-shop test sample.

The augmented model produced a lower absolute error for 113 of 207 shops and a higher error for 94. Rank correlation increased from 0.410 to 0.488. The result is meaningful but modest: the shopper block improves the existing prediction rather than replacing shopkeeper operational information.

8

Paired uncertainty

A country-stratified paired bootstrap resampled the 207 protected-test shops 5,000 times. Each bootstrap sample compared both models on the same shops.

Figure 6
Protected-test error reduction with 95% paired intervalsMAE reduction0.0324 [0.0083, 0.0562]RMSE reduction0.0343 [0.0111, 0.0575]no difference
Positive values favor the augmented model. Both intervals remain above zero; 99.5% of samples improved MAE and 99.84% improved RMSE.

The test evidence is stronger than the development estimate, but the interval remains conditional on this linked-shop sample and the fitted models. It does not represent uncertainty from collecting an entirely new multi-country survey and repeating every modeling decision.

9

Interpretation and product implications

The evidence supports a layered use of shopper information: a shopkeeper foundation, a detailed shopper enhancement when linked responses are available, and a compact segment lens for communication.

Core shopkeeper diagnostic

Available wherever a shopkeeper completes the operational questionnaire. It remains the foundation for predicting relative shop performance across the broader shop sample.

Shopper-enhanced diagnostic

Available when recent shopper feedback can be linked and aggregated. Store alternatives, reasons for choosing the shop, and desired payment or digital capabilities provide the strongest tested shopper contribution.

Segment lens for interpretation. The four shopper profiles offer a concise way to describe a shop’s linked customer mix. They improved prediction beyond shopkeeper information alone and can support portfolio summaries, presentations, and team discussion. The detailed store-choice block remains the preferred predictive input because it retained more information and received the governed protected-test evaluation.

The augmented module should report shopper coverage beside every prediction: how many linked shopper surveys informed the shop-level averages, when they were collected, and whether coverage meets a minimum operational threshold. That keeps a one-shopper estimate from looking as stable as an estimate based on several shopper voices.

The patterns can guide questions for shopkeepers—for example, what alternatives shoppers considered or which payment capabilities they wanted. They should be presented as predictive signals and opportunities for investigation, rather than automatic claims that changing one response will cause higher performance.

10

Limitations

11

Conclusion

Shopper information adds predictive value at two useful levels: broad behavioral segments provide a compact improvement, while detailed store-choice responses deliver the stronger and better-validated gain.

On the shared 841-shop development sample, adding segment shares to the shopkeeper model improved RMSE from 0.7700 to 0.7676, MAE from 0.6125 to 0.6087, and R² from 0.2634 to 0.2680. Store-choice information produced the larger improvement, reaching RMSE 0.7610, MAE 0.6051, and R² 0.2806. Combining store choice and segments was numerically best at RMSE 0.7604, MAE 0.6038, and R² 0.2818, but the extra gain over store choice alone was very small. The segment framework is therefore valuable as an understandable summary of shopper mix, while the detailed domain preserves more predictive detail.

The original governed evaluation provides the strongest evidence for practical use. On 207 protected-test shops, the locked store-choice model reduced MAE by 4.69%, reduced RMSE by 4.01%, and raised R² from 0.2154 to 0.2771 relative to the matched shopkeeper model. This supports a practical architecture for Latin American micro-retail: keep the shopkeeper model as the widely available foundation, add detailed shopper feedback when reliable links exist, and use shopper segments as a clear communication layer for understanding the customers around each shop.

R

References

  1. Eslava, M., Meléndez, M., Ulyssea, G., Urdaneta, N., & Flores, I. (2024). Firms and Inequality in Latin America. IDB Working Paper 1587, Inter-American Development Bank.
  2. Shmueli, G., Bruce, P. C., Gedeck, P., Yahav, I., & Patel, N. R. (2023). Machine Learning for Business Analytics: Concepts, Techniques, and Applications in R (2nd ed.). John Wiley & Sons.
  3. Tienda Registrada. (2024). Así se comportan las tiendas de barrio según su tamaño. Boletín 11, five-city Colombia summary through October 31, 2024.
  4. Euromonitor International. (2025). Mercado de Panificação no Brasil: Oportunidades e Desafios. Research commissioned by FIPAN.
  5. Project shopkeeper and shopper survey instruments, governed analytical artifacts, and group-defined shopper segment comparison (2024–2026). Private course-project materials; only aggregate findings are reported.