Automated promotion forecasting and methods therefor

1. A computer-implemented method for performing promotion optimization that is performed by a promotion optimization system, containing one or more processors, a promotion module, an administration module, and a monitoring module, the method, comprising:

receiving a set of variable values for a product, a desired channel, and a criteria;

accessing an experimental knowledge database of previous promotions, wherein the previous promotions have prior variable values and results on the criteria;

clustering variables by promotional concept, wherein the promotional concepts include promotion type, presence of a person in a promotion image, gender of the person, promotion format, promotion channel, and numerosity behavioral economic principle using machine learned clustering algorithms;

aggregating the promotional concepts into pools of a given size by shared attributes;

calculating an impact each value of the set of variable values has on the criteria by comparison of the prior variable value to a baseline experiment information for the set of variable values, and the prior variable value's associated result, that is in the same pool as the each value of the set of variable values;

generating a plurality of test promotions automatically for the channel;

aggregating the impact of the values of each test promotion to generate a predicted result for each test promotion;

ranking the plurality of test promotions by predicted result against the criteria;

proposing a subset of the plurality of test promotions with the highest ranking to improve administration of the plurality of test promotions; and

administering the proposed subset of test promotions in real-time wherein the real-time administration of test promotions includes concurrent testing of the plurality of test promotions at the same time to a plurality consumers, to improve selection of the variable values such that impactful values are tested.

2. The method of claim 1 , wherein the desired channel includes social media sites, in-video placement, emails, mobile browser, and loyalty accounts.

3. The method of claim 1 , wherein the criteria is one of increased volume, increased profitability, and increased customer awareness.

4. The method of claim 3 , wherein the criteria may further include promotion type.

5. The method of claim 4 , wherein the promotion types includes percent off, amount off, and buy one quantity get a second quantity free.

6. The method of claim 1 , wherein insights are isolated effects of each variable value in the previous promotions to determine if a combination of variable values will result in a desired response.

7. The method of claim 6 , wherein the previous promotions are stored in the experimental knowledge database of past multivariate promotions and their performance.

8. The method of claim 7 , wherein the performance for each multivariate promotion is compared against a baseline experiment.

9. The method of claim 8 , wherein the baseline experiment is tested in the same time period as experimental promotions.

10. The method of claim 1 , wherein the classifying is by type of promotion, gender of the person in an image, existence of a human in the image, format of promotion, and channel of promotion.

11. The method of claim 10 , wherein the classification uses behavioral economic principle.

12. The method of claim 1 , further comprising determining a predicted percentage lift for a particular variable value of a past promotion.

13. The method of claim 12 , further comprising aggregating and weighting the predicted percentage lift for each variable value to arrive at a predicted result for the test promotion.

14. The method of claim 1 , wherein the given size of the pools is configurable by a user.