Optimal pricing and advertisement slot allocation

• US 8,473,347 B1
• Filed: 11/19/2009
• Issued: 06/25/2013
• Est. Priority Date: 10/12/2009
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Koningstein, Ross
• Primary Examiner(s): Retta, Yehdega
• Application Number:
  US12/621,878
• Time in Patent Office: 1,314 Days
• Field of Search: 705/14.4, 705/14.43, 705/7.31
• US Class Current: 705/14.43
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

An allocation subsystem allocates advertisement slots based on cumulative demand of the advertisement slots. The cumulative demand can be based on a softened demand curve for advertisers that have submitted bids for allocation of the advertisement slot. The demand curves can be adjusted to regulate a variance of the advertisers spend rate throughout a budget period.

Claims

1. A computer-implemented method, comprising:

identifying, by a processing device, a derating variance over a specified period, the derating variance being a measure of variance for an effective bid over the specified period;

determining, by a processing device, whether the derating variance exceeds a threshold variance;

increasing, by a processing device, a slope of a demand curve to generate an adjusted demand curve, in response to the derating variance exceeding the threshold variance;

decreasing, by a processing device, the slope of the demand curve to generate an adjusted demand curve, in response to the derating variance not exceeding the threshold variance;

and

allocating, by a processing device, advertisement slots based on the adjusted demand curve.

2. The method of claim 1, further comprising:

determining, by a processing device, that the derating did not stabilize within a threshold time; and

decreasing, by a processing device, the slope of the demand curve in response to the determination that the derating did not stabilize within the threshold time to generate the adjusted demand curve.

3. The method of claim 1, further comprising computing the demand curve as a softened demand curve.

4. The method of claim 3, wherein computing the demand curve comprises:

accessing a maximum bid for the advertiser, the maximum bid being a maximum price that the advertiser agrees to pay for allocation of advertisement slots;

identifying a cumulative supply of advertisement slots;

and

generating the softened demand curve based on the maximum bid, the softened demand curve specifying a plurality of pairs of price and demand for the advertiser.

5. The method of claim 4, wherein allocating advertisement slots comprises:

identifying a common price at which the cumulative supply satisfies cumulative demand for advertisement slots, the common price being a price at which a page revenue satisfies a page revenue threshold; and

allocating the advertisement slot to the advertiser based on a portion of the cumulative supply demanded by the advertiser at the common price.

6. The method of claim 4, further comprising:

accessing a budget for the advertiser and a budget period;

computing a target spend rate based on the budget and the budget period, the target spend rate being a spend rate at which the budget will be exhausted at substantially the end of the budget period;

computing an actual spend rate based on an expense amount incurred for advertisement slots though a portion of the budget period;

determining that the actual spend rate is not within a threshold margin of the target spend rate; and

in response to the determination that the actual spend rate is not within the threshold margin of the target spend rate, computing an effective bid based on a derating factor.

7. The method of claim 6, further comprising computing the derating factor based on a difference between the actual spend rate and the actual spend rate.

8. The method of claim 6, wherein the effective bid is increased in response to an actual spend rate that is less than the target spend rate.

9. The method of claim 6, wherein the effective bid is decreased in response to an actual spend rate that is greater than the target spend rate.

10. A system comprising:

a datastore storing bids for advertisers, the bids being maximum amounts that the advertisers agree to pay for an advertisement slot; and

an allocation subsystem coupled to the datastore, the allocation subsystem comprising at least one processor configured to regulate variance of an effective bid for an advertiser based on an adjusted demand curve for the advertiser, the adjusted demand curve having a slope that is adjusted in response to a derating variance measured over a specified period.

11. The system of claim 10, wherein the allocation subsystem is further configured to increase the slope of the demand curve in response to a determination that the derating variance exceeds a threshold variance.

12. The system of claim 11, wherein the allocation subsystem is further configured to decrease the slope of the demand curve in response to a determination that the derating variance is less than the threshold variance and did not stabilize within a threshold time.

13. The system of claim 10, wherein the allocation subsystem is further configured to allocate advertisement slots based on the adjusted demand curve.

14. The system of claim 10, wherein the allocation subsystem is further configured to compute the demand curve based on a maximum bid specified for the advertiser and a cumulative supply of advertisement slots.

15. The system of claim 14, wherein the allocation subsystem is further configured to identify a common price at which the cumulative supply satisfies a cumulative demand for advertisement slots and allocate advertisement slots at the common price.

16. The system of claim 14, wherein the datastore further stores budgets for advertisers over a budget period and the allocation subsystem is further configured to compute an effective bid for an advertiser based on a difference between an actual spend rate and a target spend rate, the target spend rate being a spend rate at which the budget will be exhausted at substantially the end of the budget period.

17. A computer-readable medium encoded with a computer program comprising instructions that when executed operate to cause a computer to perform operations including:

identifying a derating variance over a specified period, the derating variance being a measure of variance for an effective bid over the specified period;

determining whether the derating variance exceeds a threshold variance;

increasing a slope of a demand curve to generate an adjusted demand curve, in response to the derating variance exceeding the threshold variance;

decreasing the slope of the demand curve to generate an adjusted demand curve, in response to the derating variance not exceeding the threshold variance;

and

allocating advertisement slots based on the adjusted demand curve.

18. The computer-readable medium of claim 17, further comprising instructions that cause a computer to perform operations including:

determining that the derating did not stabilize within a threshold time; and

decreasing the slope of the demand curve in response to the determination that the derating did not stabilize within the threshold time to generate the adjusted demand curve.

19. The computer-readable medium of claim 17, further comprising instructions that cause a computer to perform operations including:

accessing a maximum bid for the advertiser, the maximum bid being a maximum price that the advertiser agrees to pay for allocation of advertisement slots;

identifying a cumulative supply of advertisement slots;

and

generating the softened demand curve based on the maximum bid, the softened demand curve specifying a plurality of pairs of price and demand for the advertiser.

20. The computer-readable medium of claim 17, further comprising instructions that cause a computer to perform operations including:

accessing a budget for the advertiser and a budget period;

computing a target spend rate based on the budget and the budget period, the target spend rate being a spend rate at which the budget will be exhausted at substantially the end of the budget period;

computing an actual spend rate based on an expense amount incurred for advertisement slots though a portion of the budget period;

determining that the actual spend rate is not within a threshold margin of the target spend rate; and

in response to the determination that the actual spend rate is not within the threshold margin of the target spend rate, computing an effective bid based on a derating factor.