Paid media and measurementEvidence check: 4.5/5
Blake, T., Nosko, C., & Tadelis, S. (2015). Econometrica, 83(1), 155–174.
Plain-English take: The study found little measurable short-run benefit from brand-keyword ads in this study. For your marketing: This shows why platform-attributed conversions can overstate incrementality and why brand and non-brand traffic, user familiarity and holdout design should be analyzed separately.
Question: What is the causal short-run effect of paid search advertising on eBay, and how does it vary across consumers?
Design: Large-scale randomized geographic field experiment. eBay users and search activity in the markets included in the experiment. The findings concern one large, established online marketplace and the specific campaigns and time horizon studied.
The authors report little measurable short-run benefit from brand-keyword ads in this setting. Non-brand ads had positive responses among some newer and infrequent users, while many ad-attributed actions came from customers whose behavior was not changed by advertising; estimated average returns were negative under the studied conditions.
Econometrica · 201510.3982/ECTA12423Publisher record and author-hosted paper
Paid media and measurementEvidence check: 2.5/5
Bayer, E., Srinivasan, S., Riedl, E. J., & Skiera, B. (2020). International Journal of Research in Marketing, 37(4), 789–804.
Plain-English take: Paid search had a more positive effect on sales than offline advertising, while display advertising had a relatively more positive effect on Tobin's q than offline advertising, consistent with different short- and long-run economic roles. For your marketing: Separate near-term sales objectives from longer-horizon brand or firm-value goals when comparing channels.
Question: How do paid search and online display advertising relate to firm performance and firm value compared with offline advertising?
Design: Firm-level econometric analysis of advertising mix and performance. Firms with observed online and offline advertising and financial outcomes; the abstract reviewed does not provide the sample size or specific industries.
Paid search had a more positive effect on sales than offline advertising, while display advertising had a relatively more positive effect on Tobin's q than offline advertising, consistent with different short- and long-run economic roles.
International Journal of Research in Marketing · 202010.1016/j.ijresmar.2020.02.002Publisher abstract
Paid media and measurementEvidence check: 4.5/5
Gordon, B. R., Zettelmeyer, F., Bhargava, N., & Chapsky, D. (2019). Marketing Science, 38(2), 193–225.
Plain-English take: Commonly used observational approaches based on industry-available data often failed to accurately measure the true advertising effect in these comparisons. For your marketing: Treat attribution models as hypotheses and validate them against holdouts where possible.
Question: Can commonly used observational methods recover the causal effect of online ads as measured by randomized experiments?
Design: Comparison of observational estimators against randomized field experiments. Fifteen Facebook campaigns, each with at least one million users and business-relevant conversion tracking, selected to compare measurement methods.
Commonly used observational approaches based on industry-available data often failed to accurately measure the true advertising effect in these comparisons.
Marketing Science · 201910.1287/mksc.2018.1135Publisher abstract and author-hosted paper
Paid media and measurementEvidence check: 4.0/5
Morozov, I., & Tuchman, A. (2024). Marketing Science, 43(5), 986–1001.
Plain-English take: Ad content can polarize consumer responses. For your marketing: Test how creative changes demand across audience segments, not only average clicks.
Question: How does the content of a display ad affect what consumers search for and buy?
Design: Preregistered incentive-compatible randomized field experiment. Visitors to an experimental online bookstore; the publisher abstract does not state the sample size.
Ad content can polarize consumer responses. Depending on how many consumers favor the advertised genre, the same content can increase or decrease the total number of searches and purchases for the advertised book.
Marketing Science · 202410.1287/mksc.2023.0138Publisher abstract; full text access may vary
Paid media and measurementEvidence check: 4.0/5
Johnson, G. A., Lewis, R. A., & Reiley, D. H. (2017). Marketing Science, 36(1), 43–53.
Plain-English take: The paper's main point is that more granular data do not necessarily make advertising effects easier to estimate and examines tradeoffs between measurement precision and experimentation resources. For your marketing: Plan experiments around outcome variance and expected effect size.
Question: How do the amount and structure of available data affect the power and cost of online advertising experiments?
Design: Large-scale randomized advertising field experiment and statistical analysis. Yahoo! advertising experiment at large scale; the publisher search result reports evidence that retailer ads increased sales by 3.6% in the studied campaign.
The paper highlights that more granular data do not necessarily make advertising effects easier to estimate and examines tradeoffs between measurement precision and experimentation resources.
Marketing Science · 201710.1287/mksc.2016.0998Publisher record and abstract; full text access may vary
Paid media and measurementEvidence check: 4.0/5
Hoban, P. R., & Bucklin, R. E. (2015). Journal of Marketing Research, 52(3), 375–386.
Plain-English take: The paper uses randomized exposure data to estimate how display advertising affects multiple purchase-funnel outcomes rather than relying only on last-click attribution. For your marketing: Measure how campaigns affect more than the final conversion event, while preserving the experimental control that supports causal estimates.
Question: How does Internet display advertising affect consumer movement through the purchase funnel?
Design: Randomized field experiment with purchase-funnel model. Online visitors to one financial-tools provider who were included in the display-ad experiment; the publisher preview does not report sample size.
The paper uses randomized exposure data to estimate how display advertising affects multiple purchase-funnel outcomes rather than relying only on last-click attribution.
Journal of Marketing Research · 201510.1509/jmr.13.0277Publisher abstract; full text access may vary
Paid media and measurementEvidence check: 4.5/5
Huang, S., Aral, S., Hu, Y. J., & Brynjolfsson, E. (2020). Marketing Science, 39(6), 1142–1165.
Plain-English take: The experiment finds meaningful differences across customers in social advertising lift across categories and user characteristics. For your marketing: When testing paid social, segment hypotheses by product and audience rather than assuming platform-wide average effectiveness.
Question: How does social advertising effectiveness vary across product categories and consumer characteristics?
Design: Large-scale randomized field experiment. A large online social network and advertising across 71 product categories; the paper reports variation by product and user status.
The experiment finds meaningful heterogeneity in social advertising lift across categories and user characteristics. The authors report that social cues increase effectiveness more for some product types and users than for others.
Marketing Science · 202010.1287/mksc.2020.1240Publisher record and author manuscript
Paid media and measurementEvidence check: 2.5/5
Huang, S., & Lin, S. (2025). Journal of Marketing Research. https://doi.org/10.1177/00222429241307608.
Plain-English take: Displaying more likes increased click-through rates for lesser-known brands but not for well-known brands. For your marketing: Social proof may matter more when brand familiarity is low, but the effect can depend on audience behavior.
Question: Does displaying a larger number of likes on an ad cause more users to like or click it?
Design: Large-scale field experiment, three online lab studies, and theoretical model. WeChat social advertising plus online lab studies; the accessible abstract does not report a sample size.
Displaying more likes increased click-through rates for lesser-known brands but not for well-known brands. It had a stronger effect on like rates among more socially engaged users.
Journal of Marketing Research · 202510.1177/00222429241307608Publisher abstract; full text access may vary
Paid media and measurementEvidence check: 2.5/5
Braun, M., & Moe, W. W. (2013). Marketing Science, 32(5), 753–767.
Plain-English take: The model distinguishes exposure histories and creative content; simulations suggest advertisers can increase website visits and conversions by varying the creative shown to a person. For your marketing: Treat creative sequence and repetition as testable variables.
Question: How do display-ad creative variation and each person's exposure history relate to visits and conversions?
Design: Individual-level advertising response model with policy simulations. Online display-ad exposures and subsequent website behavior in the advertiser data analyzed by the paper; the abstracted record does not provide sample size.
The model distinguishes exposure histories and creative content; simulations suggest advertisers can increase website visits and conversions by varying the creative shown to a person.
Marketing Science · 201310.1287/mksc.2013.0802Publisher record and bibliographic abstract
Paid media and measurementEvidence check: 2.5/5
Li, C., Luo, X., Zhang, C., & Wang, X. (2017). Marketing Science, 36(5), 762–779.
Plain-English take: The study found that response was higher and faster in sunny than cloudy conditions, and lower and slower in rainy than cloudy conditions. For your marketing: This is a reminder that context can moderate campaign response.
Question: Do local weather conditions and advertising tone change consumers’ response to mobile promotions?
Design: Large-scale mobile promotion field data analysis. Mobile promotion campaigns observed across weather conditions; the publisher summary reports comparative response timing and purchase response but not a full sample description.
The authors report that response was higher and faster in sunny than cloudy conditions, and lower and slower in rainy than cloudy conditions. Message tone altered these weather-related response patterns.
Marketing Science · 201710.1287/mksc.2017.1044Publisher record and INFORMS research summary
Paid media and measurementEvidence check: 4.0/5
Goli, A., Reiley, D. H., & Zhang, H. (2025). Marketing Science, 44(2), 327–352.
Plain-English take: Consumption responses took more than a year to stabilize, while subscription responses stabilized in under six months. For your marketing: This shows why ad load can have delayed effects and why short tests may miss them.
Question: Can ad load be personalized to improve subscription and advertising revenue, and how quickly do users respond?
Design: Large-scale randomized field experiment and heterogeneous-treatment analysis. More than seven million Pandora users in an 18-month experiment.
Consumption responses took more than a year to stabilize, while subscription responses stabilized in under six months. A personalized allocation was estimated to raise subscription profits by 7% without reducing ad profits; the modeled strategy lowered average consumer welfare by 2%.
Marketing Science · 202510.1287/mksc.2022.0357Publisher abstract; full text may require access
Paid media and measurementEvidence check: 3.5/5
Mochon, D., Johnson, K., Schwartz, J., & Ariely, D. (2017). Journal of Marketing Research, 54(2), 306–317.
Plain-English take: Acquired likes had a positive causal effect on offline behavior when the firm used the page for paid promoted posts. For your marketing: This is a reminder to measure paid social against business outcomes and distinguish paid exposure from organic engagement.
Question: Do acquiring Facebook page likes affect customer behavior outside Facebook, and under what use of the page?
Design: Field experiment. Customers of a firm with a Facebook page; the publisher abstract does not state the sample size or business category.
Acquired likes had a positive causal effect on offline behavior when the firm used the page for paid promoted posts. The study found no effect when customers interacted organically with the page.
Journal of Marketing Research · 201710.1509/jmr.15.0409Publisher abstract; full text may require access
Paid media and measurementEvidence check: 4.0/5
Sahni, N. S. (2016). Journal of Marketing Research, 53(4), 459–478.
Plain-English take: Ads increased sales likelihood for the advertiser and also for nonadvertised restaurants that served the same cuisine and had high ratings. For your marketing: When evaluating category advertising, account for the possibility that your spend increases category interest while also helping alternatives.
Question: How do online ads affect both the advertiser and competing options considered by customers?
Design: Randomized online field experiments. Restaurants listed on a restaurant-search site; detailed sample counts are not given in the accessible abstract.
Ads increased sales likelihood for the advertiser and also for nonadvertised restaurants that served the same cuisine and had high ratings. Spillovers were largest at lower ad intensity and disappeared as intensity increased.
Journal of Marketing Research · 201610.1509/jmr.14.0274Publisher abstract; full text may require access
Paid media and measurementEvidence check: 4.0/5
Wernerfelt, N., Tuchman, A., Shapiro, B. T., & Moakler, R. (2025). Marketing Science, 44(2), 268–286.
Plain-English take: The median cost per additional customer was estimated at $38.16 at baseline and $49.93 under the median loss in effectiveness when offsite data could not be used, a 31% increase. For your marketing: This may be useful for thinking about privacy-related measurement loss and incrementality, especially for small advertisers.
Question: How much does offsite tracking data improve ad delivery effectiveness for advertisers?
Design: Large-scale randomized advertising experiment. More than 70,000 Facebook and Instagram advertiser experiments across over 160 countries.
The median cost per incremental customer was estimated at $38.16 at baseline and $49.93 under the median loss in effectiveness when offsite data could not be used, a 31% increase. The estimated loss was larger for small advertisers; purchase-optimized ads generated more long-term customers per dollar than click-optimized ads.
Marketing Science · 202510.1287/mksc.2023.0274Publisher abstract; full text may require access
Paid media and measurementEvidence check: 2.5/5
Johnson, G. A., Lewis, R. A., & Nubbemeyer, E. I. (2017). Journal of Marketing Research, 54(6), 867–884.
Plain-English take: In the demonstration campaign, retargeting lifted website visits by 17.2% and purchases by 10.5%. For your marketing: This is a foundational reference for asking an ad platform or analytics team how holdouts can measure incremental impact at lower cost.
Question: Can online ad effectiveness be measured more economically while retaining the precision of conventional experiments?
Design: Advertising measurement method demonstrated in a retailer campaign. Retailer display-retargeting campaign; the publisher page reports recording over 100 million predicted ghost ads per day.
In the demonstration campaign, retargeting lifted website visits by 17.2% and purchases by 10.5%. The method is presented as achieving comparable precision to intent-to-treat and public-service-announcement tests at substantially lower experimental cost.
Journal of Marketing Research · 201710.1509/jmr.15.0297Publisher abstract; full text may require access
Paid media and measurementEvidence check: 2.5/5
Li, H. (Alice), & Kannan, P. K. (2014). Journal of Marketing Research, 51(1), 40–56.
Plain-English take: Attribution based on estimated channel effects differed significantly from common metrics. For your marketing: This is a useful framework for considering channel sequence, carryover, and retargeting harm when last-click reports over-credit later touches.
Question: How can channel contribution be estimated while accounting for customer consideration, visits, carryover, and cross-channel effects?
Design: Individual-level attribution model validated in a field study. Individual customer touchpoint histories and subsequent website purchases; the accessible abstract does not give sample size or business category.
Attribution based on estimated channel effects differed significantly from common metrics. The model also identified cases where retargeting could decrease conversion probability.
Journal of Marketing Research · 201410.1509/jmr.13.0050Publisher abstract; full text may require access
Paid media and measurementEvidence check: 2.5/5
Zhou, M., Abhishek, V., Kennedy, E. H., Srinivasan, K., & Sinha, R. (2025). Information Systems Research, 36(1), 225–238.
Plain-English take: Email advertising was associated with increased sales across channels through higher purchase probability and a wider variety of products bought; effects varied across product types and customer segments. For your marketing: This encourages measuring email's full downstream value rather than clicks alone.
Question: Do email ads affect sales beyond online clicks, including purchases in physical stores?
Design: Observational causal-estimation study using customer-level behavior and transactions. High-dimensional data from a large office-supplies retailer linking online behavior to item-level physical-store purchases.
Email advertising was associated with increased sales across channels through higher purchase probability and a wider variety of products bought; effects varied across product types and customer segments.
Information Systems Research · 202510.1287/isre.2020.0557Publisher abstract; full text may require access
Paid media and measurementEvidence check: 4.5/5
Frick, T. W., Belo, R., & Telang, R. (2023). Management Science, 69(3), 1665–1686.
Plain-English take: Ads increased purchase and return probabilities, but consumers targeted more aggressively were not more responsive to ads. For your marketing: Highlights the difference between predicting who will buy and identifying who changes behavior because of an ad.
Question: Do ad platforms target users who are persuadable, or users who were already likely to purchase?
Design: Large-scale randomized field experiment. 208,538 consumers eligible for retargeting at a European consumer-electronics e-retailer over 49 days.
Ads increased purchase and return probabilities, but consumers targeted more aggressively were not more responsive to ads. The authors estimate the platform's bidding optimization reduced advertiser profit and overall welfare by about 6% relative to a firm-optimal regime.
Management Science · 202310.1287/mnsc.2022.4438Publisher abstract; full text may require access
Paid media and measurementEvidence check: 3.5/5
Valenti, A., Miller, C. J., & Tucker, C. E. (2025). Management Science, 71(10), 8586–8603.
Plain-English take: Combining contextual targeting with retargeting improved website visits, engagement, and soft conversions. For your marketing: This is a useful test idea for combining audience intent with page context and matching the creative to a meaningful next action.
Question: Do contextual targeting and retargeting complement one another, and does ad content affect their combined performance?
Design: Field experiment with a supporting laboratory experiment. Online advertising campaign for an automobile manufacturer; the accessible abstract does not state sample size.
Combining contextual targeting with retargeting improved website visits, engagement, and soft conversions. Ads prompting users to customize an offering were especially effective in the combined condition.
Management Science · 202510.1287/mnsc.2023.02310Publisher abstract; full text may require access
Paid media and measurementEvidence check: 3.5/5
Sahni, N. S. (2015). Quantitative Marketing and Economics, 13, 203–247.
Plain-English take: Ads increased purchase likelihood and effects carried over to later purchase occasions. For your marketing: This may be useful when testing retargeting frequency and lifecycle messaging cadence.
Question: Does spacing repeated ad exposures change purchase likelihood compared with bunching exposures together?
Design: Online field experiments with a memory-based response model. Online consumers exposed to digital advertising; detailed sample counts are not reported in the retrieved abstract.
Ads increased purchase likelihood and effects carried over to later purchase occasions. Spacing exposures apart increased purchase likelihood relative to bunching them, even when some exposures moved away from the purchase occasion.
Quantitative Marketing and Economics · 201510.1007/s11129-015-9159-9Publisher abstract; full text access may vary
Paid media and measurementEvidence check: 4.0/5
Grahl, J., Hinz, O., Rothlauf, F., Abdel-Karim, B. M., & Mihale-Wilson, C. (2023). Journal of Business Research, 166, 114133.
Plain-English take: Likes increased revenue mainly during non-work hours, when they raised likelihood of buying; the effect was strongest for already popular products. For your marketing: Treat visible popularity as context-dependent social proof.
Question: Does displaying product likes affect purchases and revenue, and does shopping time moderate the effect?
Design: Randomized controlled field experiment. 112,087 product-detail-page views across 9,559 hedonic products in an online store.
Likes increased revenue mainly during non-work hours, when they raised purchase propensity; the effect was strongest for already popular products. The study found no significant increase in spending conditional on a purchase.
Journal of Business Research · 202310.1016/j.jbusres.2023.114133Publisher abstract; full text may require access
Paid media and measurementEvidence check: 3.5/5
Waisman, C., Sahni, N. S., Nair, H. S., & Lin, X. (2025). Marketing Science, 44(2), 437–456.
Plain-English take: Ignoring concurrent competitive actions can misestimate advertising lift by a factor of two or more, potentially changing budget decisions. For your marketing: This is a useful caution for campaign tests run in competitive auctions: your measured lift may depend on what other advertisers are doing simultaneously.
Question: How can advertisers measure ad effects when multiple firms are experimenting at the same time on a platform?
Design: Advertising experiment and estimation-method study. Advertising experiments on JD.com, a large e-commerce platform and digital-ad publisher; exact sample counts are not in the abstract.
Ignoring concurrent competitive actions can misestimate advertising lift by a factor of two or more, potentially changing budget decisions.
Marketing Science · 202510.1287/mksc.2022.0194Publisher abstract; full text may require access
Paid media and measurementEvidence check: 3.5/5
Ellickson, P. B., Kar, W., & Reeder, J. C. III. (2023). Marketing Science, 42(4), 704–728.
Plain-English take: Content and framing had heterogeneous effects along the funnel. For your marketing: This is a detailed example of measuring message components against both engagement and purchase outcomes.
Question: Which individual email-promotion components affect opens and purchases, and how do responses vary across customers?
Design: Observational causal estimation with double machine learning. 34 email promotions sent to more than 1.3 million customers of an online apparel retailer over two months in 2015.
Content and framing had heterogeneous effects along the funnel. Clearance framing outperformed product-specific discount framing in the reported setting; top-of-funnel engagement did not always predict purchases. Off-policy evaluation estimated about an 11% revenue gain from shifting better clearance offers toward highly engaged customers.
Marketing Science · 202310.1287/mksc.2022.1401Publisher abstract; full text may require access
Paid media and measurementEvidence check: 4.5/5
Minah H. Jung, Adam L. Alter (2020). Name Similarity Encourages Generosity: A Field Experiment in Email Personalization. Marketing Science. https://doi.org/10.1287/mksc.2019.1220.
Plain-English take: Full surname matching increased opens, clicks, donations and total money; shared initial also increased generosity. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Does individualized name matching improve email response and giving?
Design: Randomized field experiment manipulating name matching in email communications.. DonorsChoose.org; N=30,297 potential donors; surname, initial, gender, ethnicity and country-of-origin matching.
Full surname matching increased opens, clicks, donations and total money; shared initial also increased generosity.
Marketing Science · 202010.1287/mksc.2019.1220Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 4/5
Xingyu Li, Jens Grahl, Oliver Hinz (2022). How Do Recommender Systems Lead to Consumer Purchases? A Causal Mediation Analysis of a Field Experiment. Information Systems Research. https://doi.org/10.1287/isre.2021.1074.
Plain-English take: Personalized recommendations increased purchase propensity 12.4% and basket value 1.7%; effects were largely mediated by consideration-set changes, especially breadth. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Through what causal paths do personalized recommendations induce purchases?
Design: Randomized controlled field experiment with causal mediation analysis.. Website of an online book retailer; consumer purchase behavior and consideration-set breadth/depth.
Personalized recommendations increased purchase propensity 12.4% and basket value 1.7%; effects were largely mediated by consideration-set changes, especially breadth.
Information Systems Research · 202210.1287/isre.2021.1074Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 4.5/5
Albert Valenti, Shuba Srinivasan, Gokhan Yildirim, Koen Pauwels (2024). Direct mail to prospects and email to current customers? Modeling and field-testing multichannel marketing. Journal of the Academy of Marketing Science. https://doi.org/10.1007/s11747-023-00962-2.
Plain-English take: Field experiment provided causal support for reallocating direct mail toward prospects/high-value segments and email toward current customers (as described by the paper). For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do email and direct mail differ by customer value and prospect/current-customer status?
Design: Hierarchical empirical model plus randomized four-cell field experiment (control, email, direct mail, both).. Beauty retailer and apparel retailer; two years of purchase history; beauty experiment stratified across six value segments and capped at 33,000 mail pieces.
Field experiment provided causal support for reallocating direct mail toward prospects/high-value segments and email toward current customers (as described by the paper).
Journal of the Academy of Marketing Science · 202410.1007/s11747-023-00962-2Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 4.5/5
Eva Ascarza, Peter Ebbes, Oded Netzer, Matthew Danielson (2017). Beyond the Target Customer: Social Effects of Customer Relationship Management Campaigns. Journal of Marketing Research. https://doi.org/10.1509/jmr.15.0442.
Plain-English take: Targeting customers affected the behavior of their social connections, showing CRM campaign effects beyond directly targeted customers. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Do CRM campaigns affect people connected socially to the targeted customer?
Design: Randomized CRM field experiment with network/social-connection analysis.. Nearly 6,000 customers of a mobile telecommunications provider and their social connections.
Targeting customers affected the behavior of their social connections, showing CRM campaign effects beyond directly targeted customers.
Journal of Marketing Research · 201710.1509/jmr.15.0442Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3.5/5
Alexander Bleier, Maik Eisenbeiss (2015). Personalized Online Advertising Effectiveness: The Interplay of What, When, and Where. Marketing Science. https://doi.org/10.1287/mksc.2015.0930.
Plain-English take: Effectiveness depends jointly on what is personalized, when the ad appears, and where it appears; well-targeted personalization can increase responsiveness. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do message, timing, and placement interact in personalized online advertising?
Design: Two field experiments testing personalization, timing, placement/context, and advertising effectiveness.. Consumers receiving personalized online display advertising in commercial field settings.
Effectiveness depends jointly on what is personalized, when the ad appears, and where it appears; well-targeted personalization can increase responsiveness.
Marketing Science · 201510.1287/mksc.2015.0930Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 4/5
Jana Liaukonyte, Theodore Teixeira, Kenneth C. Wilbur (2015). Television Advertising and Online Shopping. Marketing Science. https://doi.org/10.1287/mksc.2014.0899.
Plain-English take: Television advertising changes online shopping activity, with effects varying by ad characteristics and shopping stage. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How does television advertising affect online shopping and which funnel outcomes respond?
Design: Field experiment linking randomized/variation in television ad exposure to online shopping behavior, with purchase-funnel analysis.. Online shopping sessions and television advertising for a retail advertiser.
Television advertising changes online shopping activity, with effects varying by ad characteristics and shopping stage.
Marketing Science · 201510.1287/mksc.2014.0899Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3.5/5
Weijia Dai, Hyunjin Kim, Michael Luca (2023). Which Firms Gain from Digital Advertising? Evidence from a Field Experiment. Marketing Science. https://doi.org/10.1287/mksc.2023.1436.
Plain-English take: Digital advertising gains differ across restaurant types; the experiment identifies which businesses benefit more from advertising. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Which types of firms gain most from digital advertising?
Design: Large-scale field experiment randomizing digital advertising for restaurants and comparing heterogeneous treatment effects.. Restaurants and their online customers on a digital platform.
Digital advertising gains differ across restaurant types; the experiment identifies which businesses benefit more from advertising.
Marketing Science · 202310.1287/mksc.2023.1436Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 2.5/5
Jacquelyn S. Thomas, Chaoqun Chen, Dawn Iacobucci (2022). Email Marketing as a Tool for Strategic Persuasion. Journal of Interactive Marketing. https://doi.org/10.1177/10949968221095552.
Plain-English take: Email can operate as a strategic persuasion tool; response depends on message strategy rather than sending alone. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How can email marketing be used strategically to persuade customers?
Design: Empirical analysis of email marketing campaigns examining strategic persuasion and campaign response.. Commercial email recipients/campaigns in the study’s field setting.
Email can operate as a strategic persuasion tool; response depends on message strategy rather than sending alone.
Journal of Interactive Marketing · 202210.1177/10949968221095552Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 2.5/5
Tom DeWitt, Doan T. Nguyen, Roger Marshall (2008). Exploring Customer Loyalty Following Service Recovery: The Mediating Effects of Trust and Emotions. Journal of Service Research. https://doi.org/10.1177/1094670507310767.
Plain-English take: Trust and both positive and negative emotions mediated the relationship between perceived justice and loyalty. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do perceived justice, trust and emotions explain customer loyalty after service recovery?
Design: Structural-equation-model test of a cognitive-appraisal model.. Customers evaluating service recovery; the accessible publisher abstract does not state industry or sample size.
Trust and both positive and negative emotions mediated the relationship between perceived justice and loyalty.
Journal of Service Research · 200810.1177/1094670507310767Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3/5
Andy Chin Woon Fook, Omkar Dastane (2021). Effectiveness of Loyalty Programs in Customer Retention: A Multiple Mediation Analysis. Asia-Pacific Journal of Management Research and Innovation. https://doi.org/10.1177/22786821211000182.
Plain-English take: Loyalty programs positively affected retention; brand association mediated more strongly than satisfaction, with parallel, distinct and sequential mediation patterns differing. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Do loyalty programs improve retention, and are brand association and satisfaction mediators?
Design: Quantitative questionnaire study; CFA, SEM and bootstrapped multiple mediation analysis.. 313 Malaysian national-car users selected by convenience sampling.
Loyalty programs positively affected retention; brand association mediated more strongly than satisfaction, with parallel, distinct and sequential mediation patterns differing.
Asia-Pacific Journal of Management Research and Innovation · 202110.1177/22786821211000182Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3/5
Keng-Boon Ooi, others (2023). Social media influencers: An effective marketing approach?. Journal of Business Research. https://doi.org/10.1016/j.jbusres.2023.113773.
Plain-English take: Mobile convenience, interactivity, and influencer credibility shape attitudes; product/service attitude fully mediates influencer-attitude effects on actual purchase behavior. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do mobile convenience, interactivity, and influencer credibility lead to actual purchase behavior?
Design: Cross-sectional questionnaire study using structural equation modeling; includes actual purchase behavior outcome.. Mobile users in a developing-country context; 100 usable respondents are reported in indexed study descriptions.
Mobile convenience, interactivity, and influencer credibility shape attitudes; product/service attitude fully mediates influencer-attitude effects on actual purchase behavior.
Journal of Business Research · 202310.1016/j.jbusres.2023.113773Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 2.5/5
K.Y. Koay, C.W. Teoh, P.C. Soh (2021). Instagram influencer marketing: perceived social media marketing activities and online impulse buying. First Monday. https://doi.org/10.5210/fm.v26i9.11598.
Plain-English take: The study estimates relationships between perceived social-media marketing activities and online impulse buying. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Do perceived Instagram marketing activities by influencers predict online impulse buying?
Design: Empirical questionnaire/PLS-SEM study of perceived influencer marketing activities and impulse buying.. Instagram users in a consumer social-commerce context.
The study estimates relationships between perceived social-media marketing activities and online impulse buying.
First Monday · 202110.5210/fm.v26i9.11598Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3.5/5
Alexander Bleier, Maik Eisenbeiss (2015). The Importance of Trust for Personalized Online Advertising. Journal of Retailing. https://doi.org/10.1016/j.jretai.2015.04.002.
Plain-English take: Greater personalization depth increased click-through only for a more trusted retailer and reduced it for a less trusted retailer; trust also shaped usefulness/reactance in the lab. For your marketing: Use to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Does retailer trust moderate the effectiveness and privacy/reactance consequences of personalized retargeting?
Design: Quasi-experimental retargeting field study followed by scenario-based online experiment.. Consumers exposed to retargeting banners from retailers; real-world field data plus lab participants.
Greater personalization depth increased click-through only for a more trusted retailer and reduced it for a less trusted retailer; trust also shaped usefulness/reactance in the lab.
Journal of Retailing · 201510.1016/j.jretai.2015.04.002Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3/5
Haoyan Sun, Ming Fan, Yong Tan (2020). An Empirical Analysis of Seller Advertising Strategies in an Online Marketplace. Information Systems Research. https://doi.org/10.1287/isre.2019.0874.
Plain-English take: Both tools increased traffic; only sponsored search significantly increased sales. Sponsored search worked better for low-reputation traffic and high-reputation sales; endorsement sales effects appeared only for high-reputation sellers. For your marketing: Use it to shape a local channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do sponsored search and social-media endorsement affect seller traffic and sales, and how does reputation moderate effects?
Design: Observational marketplace study with controls for sellers’ self-selection into sponsored search/social endorsement.. Sellers at a retail e-commerce platform; traffic, sales, advertising tools, and seller reputation.
Both tools increased traffic; only sponsored search significantly increased sales. Sponsored search worked better for low-reputation traffic and high-reputation sales; endorsement sales effects appeared only for high-reputation sellers.
Information Systems Research · 202010.1287/isre.2019.0874Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 2.5/5
Anthony Burton-Jones, Gordon S. Hubona (2006). The Mediation of External Variables in the Technology Acceptance Model. Information Systems Research. https://doi.org/10.1287/isre.1060.0080.
Plain-English take: External variables influenced usage largely through perceived usefulness and ease of use, supporting mediation rather than treating external variables as direct usage drivers. For your marketing: Use it to shape a local channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How do external variables enter TAM—directly or through usefulness/ease-of-use beliefs?
Design: Empirical survey study using structural equation modeling to compare direct and mediated effects.. Information-system users in an organizational usage context.
External variables influenced usage largely through perceived usefulness and ease of use, supporting mediation rather than treating external variables as direct usage drivers.
Information Systems Research · 200610.1287/isre.1060.0080Publisher record and abstract retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 3/5
Jai-Yeol Son, Sung S. Kim, Frederick J. Riggins (2006). Consumer Adoption of Net-Enabled Infomediaries: Theoretical Explanations and an Empirical Test. Journal of the Association for Information Systems. https://doi.org/10.17705/1jais.00094.
Plain-English take: Intention to use the infomediary was a function of perceived efficiency, perceived effectiveness, and perceived ease of use. Shoppers low in asset specificity and high in uncertainty appreciated the infomediary benefits more than other shoppers. For your marketing: Use it to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: Why and how do online shoppers adopt a new infomediary website, and how do the infomediary's utilitarian benefits and the shopper's transaction environment shape adoption intention?
Design: Empirical test of a conceptual model using data from an online questionnaire; the publisher page says the model was tested with online shoppers.. 367 online shoppers evaluating adoption/use of a net-enabled infomediary website that searches and compares online retailers.
Intention to use the infomediary was a function of perceived efficiency, perceived effectiveness, and perceived ease of use. Shoppers low in asset specificity and high in uncertainty appreciated the infomediary benefits more than other shoppers.
Journal of the Association for Information Systems · 200610.17705/1jais.00094Publisher abstract/metadata retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 2.5/5
Qiang Yan, Shuang Wu, Lingli Wang, Pengfei Wu, Hejie Chen, Guohong Wei (2016). E-WOM from e-commerce websites and social media: Which will consumers adopt?. Electronic Commerce Research and Applications. https://doi.org/10.1016/j.elerap.2016.03.004.
Plain-English take: EC-eWOM usefulness and credibility positively influenced EC-eWOM adoption but negatively influenced SM-eWOM adoption. EC-eWOM adoption negatively affected SM-eWOM adoption and mediated usefulness/credibility effects; cognitive level and involvement moderated effects. For your marketing: Use it to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: When consumers encounter eWOM from e-commerce websites and social media simultaneously, how do they choose/adopt the two sources, and how do usefulness, credibility, cognition, and involvement shape that process?
Design: Empirical questionnaire-survey analysis of a conceptual model based on cognitive-cost theory; the publisher page says the questionnaire had demographic items and 30 construct items.. Consumers exposed to eWOM from e-commerce websites (EC-eWOM) and social media (SM-eWOM) in a social-commerce/online-shopping context; the fetched publisher page does not state the respondent count.
EC-eWOM usefulness and credibility positively influenced EC-eWOM adoption but negatively influenced SM-eWOM adoption. EC-eWOM adoption negatively affected SM-eWOM adoption and mediated usefulness/credibility effects; cognitive level and involvement moderated effects.
Electronic Commerce Research and Applications · 201610.1016/j.elerap.2016.03.004Publisher abstract/metadata retrieved; consult the linked paper for full methods and tables.
Paid media and measurementEvidence check: 4/5
Marcel Goic, Andrea Rojas, Ignacio Saavedra (2021). The Effectiveness of Triggered Email Marketing in Addressing Browse Abandonments. Journal of Interactive Marketing. https://doi.org/10.1016/j.intmar.2021.02.002.
Plain-English take: Browse-abandonment emails increased online-channel sales and sales in the triggered category; longer navigation histories and wider-assortment recommendations were associated with larger conversions/revenues. For your marketing: Use it to shape a locally controlled channel or message test; account for audience, timing and attribution limits before reallocating budget.
Question: How effective are triggered emails at converting customers who browse but leave without purchasing?
Design: Randomized experiment: approximately half of browse-abandoning customers received automated emails and the other half received no message.. Customers of a multichannel retailer who had browsed the website and abandoned before purchase.
Browse-abandonment emails increased online-channel sales and sales in the triggered category; longer navigation histories and wider-assortment recommendations were associated with larger conversions/revenues.
Journal of Interactive Marketing · 202110.1016/j.intmar.2021.02.002Publisher abstract/metadata retrieved; consult the linked paper for full methods and tables.