MKR 7000 Module 4 Analysis and Report on a Subscription Test Example

From 27 Percent Who Say Yes to 2 Percent Who Sign Up: Reporting a Subscription Test

Student Name

Master of Business Administration, Thomas Edison State University

MKR-7000: Marketing Research

Mentor Name

Month Day, Year

From 27 Percent Who Say Yes to 2 Percent Who Sign Up: Reporting a Subscription Test

Executive Summary

Hearthstone Market's research supports a limited launch of the prepared-meal subscription at $69 a week, beginning in the stores with the most family households. Stated interest was substantial and fell predictably with price. Actual sign-ups in test stores were far lower than stated intent, as research on intentions predicts, but renewal was healthy and cannibalization of the prepared foods counters was modest. A phased launch in 10 stores, with clear criteria for expanding, balances the evidence of real demand against the gap between what customers said and what they did.

Survey Results

The survey drew 612 completes from 5,000 invitations, a 12.2 percent response rate. Respondents resembled the full invited sample on loyalty spending and visit frequency, though they slightly over-represented the largest stores, so results were weighted by store. Intent is reported as the top-two-box share, those answering probably or definitely would subscribe.

A chi-square test of intent by price group gives a value of 13.3 with two degrees of freedom, significant at the 0.01 level, so the differences are unlikely to be chance. Intent was highest among busy families, 41 percent at $59, and lowest among older couples. The weeknight convenience scale correlated 0.38 with intent, confirming that the subscription appeals to time-pressed households. The most important features were reliable pickup times, adjustable portion sizes and weekly menu variety; delivery mattered less than expected, which lowers the cost of a launch based on store pickup.

Field Test Results

The four test stores offered the plan at $69 for four weeks. Sign-ups reached 21 per 1,000 loyalty households, about 2.1 percent. Of those subscribers, 64 percent renewed after the first four weeks. To measure cannibalization, prepared foods counter sales were compared using difference-in-differences: counter sales in test stores fell 3 percent from the pre-period while those in matched control stores rose 1 percent, a net effect of about 4 percentage points, smaller than the subscription revenue added. Because the comparison subtracts the change in matched stores, it removes chain-wide trends and seasonal effects, the main threats that a simple before-and-after comparison would leave in place (Shadish et al., 2002).

Interpreting the Gap Between Intent and Behavior

Twenty-seven percent stated intent at $69 against 2.1 percent actual sign-up in four weeks is a large gap, but not a surprising one. Morwitz et al. (2007) found that stated intentions predict behavior less well for new products and that only a fraction of stated intent converts. Babin (2019) advises treating survey intent as a relative measure, useful for comparing prices or segments, rather than as a forecast. Used that way, the survey ranks prices and segments correctly, while the field test supplies the realistic level of demand. Four weeks is also short; sign-ups may grow as awareness spreads, which the phased launch will reveal.

Projected Economics

Applying the test's 2.1 percent sign-up rate and 64 percent renewal to the 10 stores with the most family households, about 46,000 loyalty households, suggests roughly 970 initial subscribers and about 620 continuing after the first month. At $69 a week and an estimated 30 percent contribution margin, 620 subscribers contribute about $667,000 a year before fixed costs, after allowing for the counter-sales effect. That supports a phased launch but not yet the full $1.2 million chain-wide investment.

Recommendation

Launch at $69 in the 10 stores with the highest share of family households, using store pickup, with adjustable portions and a rotating menu. Set expansion criteria in advance: sign-ups above 2.5 percent of loyalty households and renewal above 60 percent after three months justify extending to the remaining stores. Track cannibalization monthly with the same control-store comparison. Revisit the $59 price for a targeted family promotion, since intent there was highest, if margins allow.

Limitations

Results come from one season and four test stores, so they may not hold in summer or in every trade area. Survey respondents were loyalty members with email, who may differ from other customers. The renewal rate covers only one cycle. These limits support the phased approach rather than a full launch.

Presenting the Results

The report will be presented to the leadership team in a 20-minute session with five slides: the recommendation and criteria, intent by price, intent by household type, the field test results and the projected economics. Each slide states its finding in the title so that the deck reads as an argument. Charts will use sorted bars for intent by segment and a simple before-and-after comparison of test and control stores for counter sales, so that the difference-in-differences logic is visible without statistical vocabulary. Detailed tables, the questionnaire and the weighting method will sit in an appendix for anyone who wants to check the work. Questions likely to arise, such as why sign-ups were so far below stated intent, have prepared answers drawn from the interpretation section.

Conclusion

The research shows real but modest demand, strongest among busy families at lower prices, with good early retention and limited cannibalization. A phased launch with criteria set in advance turns those findings into a decision while protecting the company from overestimating demand.

References

Babin, B. J. (2019). Essentials of marketing research (7th ed.). Cengage.

Morwitz, V. G., Steckel, J. H., & Gupta, A. (2007). When do purchase intentions predict sales? International Journal of Forecasting, 23(3), 347-364. https://doi.org/10.1016/j.ijforecast.2007.05.015

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.