Online Advertising Open lecture at Warsaw University January 7/8, 2011

Publish in

Documents

6 views

Please download to get full document.

View again

of 95
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Please interrupt me at any point!. Online Advertising Open lecture at Warsaw University January 7/8, 2011. Ingmar Weber Yahoo! Research Barcelona ingmar@yahoo-inc.com. Disclaimers & Acknowledgments.
Transcript
Please interrupt me at any point!Online AdvertisingOpen lecture at Warsaw UniversityJanuary 7/8, 2011Ingmar WeberYahoo! Research Barcelonaingmar@yahoo-inc.comDisclaimers & Acknowledgments
  • This talk presents the opinions of the author. It does not necessarily reflect the views of Yahoo! Inc. or any other entity.
  • Algorithms, techniques, features, etc. mentioned here might or might not be in use by Yahoo! or any other company.
  • Some of the slides in this lecture are based on slides for “Introduction to Computational Advertising”, given by A. Broder and V. Josifovski at Stanford University. http://www.stanford.edu/class/msande239/
  • Goals of this Presentation
  • Give an overview of the two main types of online advertising; (i) search advertising and (ii) display advertising
  • Explain the key technical aspects behind with a focus on computational aspects
  • This time: more breadth
  • Next time: more depth (you tell me where!)
  • Types of Online Advertising
  • Search Advertising
  • Display Advertising
  • E-mail Advertising
  • Classifieds
  • Sponsorships
  • Part 1Part 2Part 0Setting the SceneDifferent Advertising ObjectivesBrand AdvertisingYou’re not expected to buy a rolex watch tomorrow.What’s different?Direct MarketingTries to cause an (almost) immediate reaction.US Online Spending share by objective What’s bigger?Branding or direct response?Lots of $$$ (or zloty)Poland’s state deficit in 2010: ~$11 billionPoland’s agriculture GDP: ~$32 billionPart 1Search AdvertisingThe Life of an Ad - Terminology“impression”/“pageview”<script type="text/javascript" src="http://www.yahoo.com/conversion.js"> </script> “click”“click-through rate”:(# clicks)/(# impressions)“landing page”“target page”“conversion” or “action”“conversion rate”:(# conversions)/(# page visits)“tracking code”Search Advertising
  • Advertisements are sold in auctions
  • Advertisers bid on search terms [show live]
  • Different payment models
  • CPC (cost per click)
  • Advertiser pays $X when an ad gets clicked
  • CPA (cost per action)
  • Advertiser pays $Y when a click on an ad leads to a (trans-)action/purchase
  • CPM (cost per mille [page impressions])
  • Advertiser pays $Z for 1,000 ad displaymentsde-facto standardgrowing popularityused for display adsBidding for search termsAdvertisers compete for search terms “warsaw hotels”, “online advertising”, …A click has a different value for different advertisers depends on profit margin and on conversion rateThere’s a ranked list of sponsored search results Assumption: higher ranking => more clicks (CTR)Advertisers bid for a (good) slot in the results $ 0.01 per click - $ 100.00 per clickSearch engine decides the order/inclusion slots are assigned to (successful) biddersWhen a user clicks on a sponsored search result … … payment is made by the advertiserSearch engines need to decide: * How should the slots be assigned? * How much should be paid per click?Advertisers need to decide: * How much to bid?MyComputer.com99% of web site visitors don’t purchase anything 1% buy a computer - conversion rate (from click to transaction)Profit per computer sold $100Expected profit per visitor $1 – value of a single visit/clickHow would you do it?Guess the most expensive search term?How much do people typically pay?How much do people typically pay?How much does X cost?
  • Try to guess some expensive key words
  • Clear (commercial) intent
  • Very high value for new customer
  • Keyword tool
  • Small competition …
  • The winner is …
  • Mesothelioma
  • Exercise
  • Build six teams
  • Think of terms to bid on (exact match) and corresponding ads. You can choose the target page!
  • You’ll get 5 EUR per team to target the US&Canada search market
  • Ads will go live around 18h00 today (Friday) and we’ll look at the results tomorrow (Saturday) around 16h00
  • Exercise
  • All ads will run under my account
  • All keywords have to be “distinct” (system doesn’t allow self-competition)
  • Assigned in reversing round robin fashion (1,2,3,3,2,1,1,2,3,…)
  • Max 5 key words and 1 ad per team
  • The team with the largest number of clicks by 16h00 on Saturday wins
  • Please, no cheating
  • Pricing of Ads
  • How was it done?
  • What was wrong with that?
  • How is it done now?
  • Does that solve all problems?
  • Historic Overture mechanismSlot assignment by bid orderAssign the slots in the order of the bid valueshigher bid => higher slotWhen a user clicks, you pay your bid valueYou bid $1.00 per click? - You pay $1.00 per click!Simple. - Intuitive. - Used for many years.What’s wrong with this?End of story? – No, because …Difficult for advertisers to “play” this “game”:There’s no equilibrium!Scenario:
  • Two available ad slots with CTR 5% and 4% respectively
  • Three bidders with valuations $20, $18, $10 per click
  • What happens? Bidder 2 bids $10.01 to beat Bidder 1 and to get a slot Bidder 1 will not pay more than $10.02 Then bidder 2 bids $10.03 Then bidder 1 bids $10.04 … and the fun continues until $14 … when it all collapses back to $10.01Difficult to “play” this game optimally.Potential feeling of “being cheated”.End of story? – And no, because …Ads can have different motivations
  • Motivating an action/purchase/click
  • Simply placing/marketing a brand
  • ebay could afford to bid for every term …... because no one will click the ad!“Buy * on ebay!”* = world peace, grandmother, happiness, …ebay cares more about page impressionsWant to get rid of high-bidding free riders.Addressing the first problem: Second price auctionIf only a single slot exists, do the following:Assign the slot to the highest bidder.Ex: Slot goes to Bidder 1 who bid $17.Let him pay the second highest bid. Ex: Bidder 1 pays $15, Bidder 2’s bid.Theorem (Vickrey ‘61): Bidding truthfully is a dominant strategy in this setting.(c.f. stamp auctions 1878+)Second Price Auction ExplainedThis ad slot is worth €1 to me.He’s “lying”.I bid €0.80!Your title hereYour cool adtext goes here.Loses item. But could have bid €1.00.Pays €0.70. But could have bid €1.00.Loses item. Should have bid €1.00.www.domain.comI bid €0.90!I bid €1.50!I bid €0.70!Bidding “truthfully” is always best.Regardless of what others do.Only works for a single slot …Addressing the first problem:Generalized second price auctionIf many slots exist, do the following:Assign the slots in (decreasing) order of the bids.Let each one pay the next (lower) bid.Called: Generalized second price (GSP) auctionIs bidding “truthfully” a dominant strategy?Are there any dominant strategies?Addressing the first problem:Generalized second price auctionSame scenario again:
  • Two available ad slots with CTR 5% and 4% respectively
  • Three bidders with valuations $20, $18, $10 per click
  • What happens if everyone bids truthfully ($20, $18, $10 respectively)? Bidder 1: ($20-$18)*0.05 = $0.10 profit per page impression Bidder 2: ($18-$10)*0.04 = $0.32 profit per page impression Bidder 3: $0.00 profit per page impression If bidder 1 bids $11 instead … … his profit is ($20-$10)*0.04 = $0.40 per page impressionBidding “truthfully” is not a dominant strategy in GSP.In fact, no dominant strategy exists for GSP.So, still saw-tooth under GSP?As long as you bid less than the higher bid, your payment doesn’t change …… but the guy above gets charged more. So:Bidder 2 increases bid to stay just slightly below bidder 1 No difference for his position/payment But payment of other bidder 1 goes upBidder 1 can “retaliate” by underbidding bidder 2 Bidder 1 now pays less (for a worse slot) Bidder 2 now pays more (for a better slot)Bidder 1 and bidder 2 have swapped position and (kind of) bids.“locally envy-free” if these games don’t happen.Locally envy-free equilibria“Internet Advertising and the GSP Auction: Selling Billions of Dollars Worth of Keywords”, Edelman et al., 2006A (pure Nash) equilibrium is locally envy-free if for any rank i: ®i sg(i) – p(i)¸®i-1 sg(i) – p(i-1) ®i = CTR at rank i (think “volume”) p(i) = cost for rank i small i = low rank = high CTRLocally envy-free equilibriaLemma 1: A locally envy-free equilibrium of the GSP game corresponds to a stable assignment.Stable assignment: nobody wants to swap position and payment with anybody elseProof: No swap with positions below as we have an equilibrium: could just undercut advertiser to make this swap.Remains to show: no swap with positions (far) above.Locally envy-free equilibriaProof (ctd):Claim: resulting order is “assortative”, i.e. in the order of the sg(i):®i sg(i) – p(i)¸®i+1 sg(i) – p(i+1) (equilibrium)®i+1 sg(i+1) – p(i+1)¸®i sg(i+1) – p(i) (envy-free) Gives:(®i - ®i+1) sg(i)¸ (®i - ®i+1) sg(i+1)Locally envy-free equilibriaProof (ctd): Suppose i wants to go to m<i®i sg(i) – p(i)¸®i-1 sg(i) – p(i-1)®i-1 sg(i-1) – p(i-1)¸®i-2 sg(i-1) – p(i-2)…®m+1 sg(m+1) – p(m+1)¸®m sg(m+1) – p(m)Replace all sq(x) by sq(i) (using Claim and ®j > ®j+1). Then add and cancel. Get: ®i sg(i) – p(i)¸ ®m sg(i) – p(m)Locally envy-free equilibriaLemma 2: When there are more advertisers than slots, then any stable assignment corresponds to a locally envy free equilibrium of the GSP game.Could be an empty set …butTheorem: Bidding bj = pV,(j-1)/®j-1 gives a locally envy-free equilibrium with VCG payments. Here pV,(j-1) are VCG payments.Why is this of little practical relevance?So, still saw-tooth under GSP?At least GSP has equilibria, though not in dominant strategies.GSP is “reasonably stable”.Payment depends on position, not on bid directly.“Correct” generalization of SP:Vickrey-Clarke-Groves MechanismAssume “no ebay”: CTR depends only on slotAssign the slots in bid order … (again)Advertiser X has to pay for loss in (bid * clicks)(Sum of (bi¢CTRi) before X enters the game -sum of (bi¢CTRi) of other players after X enters) / CTRXExample: …. next slide …“Correct” generalization of SP:Vickrey-Clarke-Groves MechanismSame scenario again:3 advertisers: bids $20, $18, $10 (their valuations)Two slots: CTR 5%, CTR 4% [think: 5 clicks, 4 click]Slots go to bids $20 and $18 respectively.Corresponding payments?Advertiser 1:W/o adv. 1, sum over adv. 2 and 3$18*0.05 + $10*0.04 = $1.30W/ adv. 1, sum only over adv. 2$18*0.04 = $0.72Payment by advertiser 1:($1.30-$0.72)/0.05 = $11.6 (per click)Advertiser 2:Without adv. 2, sum over adv. 1 and 3$20*0.05 + $10*0.04 = $1.40With adv. 2, sum only over adv. 1$20*0.05 = $1.00Payment by advertiser 2:($1.40-$1.00)/0.04 = $10 (per click)“Correct” generalization of SP:Vickrey-Clarke-Groves MechanismTheorem: Bidding “truthfully” is a dominant strategy in this mechanism.Vickrey got Nobel prize in economics in ‘96(a few days before his death)VCG mechanism not used for web advertising!Still have ebay problem …Addressing the “ebay problem”Slot assignment by revenue orderHave weights for different advertisersMeasure probability of click (= quality of ad) ctrebay = 0.001, ctringmar = 0.01Assign slots in (decreasing) order of ctri ¢bi (~ revenue for search engine)Pay minimum bid needed to stay ahead: pi = ctri+1¢bi+1/ctriRevenue ordering vs. bid ordering30% more revenue per page impressionGSP in Practice
  • GSP with revenue ordering used by all major search engines
  • But with modifications …
  • minimum price (“reserve price”)
  • number of slots is variable
  • quality of landing page to avoid frustration
  • positional constraints
  • “Putting Nobel Prize-winning theories to work” ?Google’s unique auction model uses Nobel Prize-winning economic theory to eliminate the winner’s curse – that feeling that you’ve paid too much. While the auction model lets advertisers bid on keywords, the AdWords™ Discounter makes sure that they only pay what they need in order to stay ahead of their nearest competitor.http://www.google.com/adsense/afs.pdfKnowing the Click-Through Rates
  • How do we know the click-through rates?
  • Estimated from past performance
  • What if a new advertiser arrives?
  • If we show his ads, lose chance to show other good ads.
  • If we don’t show his ads, might not discover a new high-performing ad.
  • Solution: Explore-ExploitWhat is the problem?Multi-Armed Bandits$10$1$6$4Expect $8$3Expect $2Expect $6$4$10$2$8First, explore!Now, exploit!Multi-Armed Bandits
  • Set of k bandits, i.e. real distributions
  • B = {R1, …, RK}¹k = mean(Rk) ¹* = maxk {¹k}Game is played for H roundsRegret: ½(H) = H ¹* - t=1H rt where rt is the (random) reward at time tWant ½(H)/H ! 0 with probability 1 as H!1Suggestions?Multi-Armed BanditsEpsilon-greedy strategy:The currently best bandit is selected for a fraction of 1- ² of the rounds, and a bandit selected uniformly at random for a fraction of ².Restless Bandit Problem – distributions changeArm Acquiring Bandit – new bandits arrivePractical CTR Complications
  • CTR depends also presence/absence of other ads
  • And what the user has seen in the past
  • And on quality of search results
  • Should we show the worst search results so that users are “desperate” and click the ads?
  • Fraud
  • Click fraud
  • On opponent's paid search results (10%-20%)
  • On the contextual ads of your homepage
  • Impression fraud
  • Give your opponent a lower CTR
  • Lowers the amount you’ll have to bid
  • What should search engines do?
  • All search engines do not bill for fraudulent clicks
  • See case “Lane’s Gifts v. Google”
  • Other kinds?Does CPA Solve Fraud?Click fraud no longer works. Only get charged for “actions”, aka conversion.Now advertisers can cheat by underreporting conversions. Can Y!/G trust advertisers?Have to hand over monitoring to search engine. Can advertisers trust Y!/G?Very, very sparse data to derive estimates. Hard for Y!/G to make optimal decisions.End of story?Mobile Sponsored Search
  • Mobile devices offer more context
  • Location
  • More short-term needs -> more monetizable
  • More focused user attention
  • Can’t just open another tab while loading
  • More positive associations
  • People tend to feel “closer” to their mobile
  • Summary of Part 1
  • Search advertising is a multi-billion dollar business
  • Allows very targeted advertising
  • Fair payment model: you only pay for clicks (CPC)
  • How much you pay depends on
  • Your bid
  • Fraction of people clicking your ad (CTR)
  • Payment reasonably stable and “gaming” is difficult
  • Practical problems such as learning CTRs and avoiding click fraud
  • Exercise
  • 6 teams …
  • Part 2Display AdvertisingDisplay AdvertisingHistorical note: banners
  • Banners seem to be the oldest standard format in use
  • According to Wikipedia the first banner ad ever was sold in 1993 by Global Network Navigator (GNN) to Heller, Ehrman, White, & McAuliffe, a legal firm popular in Silicon Valley.
  • GNN was a popular pre-Yahoo! directory eventually sold to AOL in 1995
  • Heller Ehrman White & McAuliffe was started in 1890 and went bankrupt in 2008. In 1929 they negotiated the financing of the Bay Bridge.
  • Display Advertising
  • Usually sold on a CPM basis
  • Guaranteed delivery (GD): deliver 30 million impressions on finance.yahoo.com in Feb ’11
  • Typically large, “premium” campaigns
  • Non-guaranteed delivery (NGD): sold in auctions on the spot market at varying prices
  • Typically smaller, ad-hoc campaigns
  • How much does it cost?Components of a GD system1. Forecast supply and demandHow many users will visit a page in a certain period?2. Forecast NGD pricing How much could we get on the spot market?3. Admission control & pricing 30m impressions in July 2011 on sports.yahoo.com Should we accept the contract? Can we meet the guarantee? What price should we charge? How are other contracts impacted?4. “Optimal” allocation of impressions to active contracts What is the objective function? Cannot re-run after every impression due to scalability. 5. Ad servingDemand (long term) depends on quality of allocation!“females, 30-50, high income” more valuable than “teenager drop-outs”Cannot only use low value impressions to satisfy contract“Simple” (stochastic) packing problem?Optimal Allocation
  • Optimal allocation
  • Maximize a stated objective function subject to supply and demand constraints
  • What objective?
  • Value of the remaining inventory? - Good for publisher
  • Maximize quality? - Good for advertiser
  • Need to balance utilities: publisher, advertiser, user, & network!
  • Representative Allocations A. Ghosh & al., “Randomized Bidding for Maximally Representative Allocation”, Yahoo! Research Technical Report 2008-003
  • Unless the targeting is very fine-grained there is a wide spectrum of quality of impressions matching a typical contract
  • Contract says: Male, US, auto interests. What should be supply to this contract?
  • Is it OK to supply 100% 15 year-old males, daydreaming about cars, weekly allowances $25 ?
  • Advertiser probably wants/expects a representative sample of car-buying US male population
  • Publisher’s potential strategiesAssume publisher has just one GD contract
  • Suboptimal strategy:
  • Deliver first all impressions to the contract
  • Only after the contract is met, sell in spot market
  • Bad for the publisher because some of the GD pageviews may fetch lot more money on the spot than the contract value
  • Better strategy
  • Put up every pageview on auction (as a seller)
  • Also place a bid on it for the contract (as a buyer)
  • Value determined by probability & penalty of not fulfilling the contract
  • Why suboptimal?Publisher-optimal bid strategy
  • If target is 30 million, place the smallest constant bid in each round so that exactly 30 million pageviews are won
  • All excess inventory will be sold to someone else (not the GD contract) at a higher price.
  • “Unfair” to the GD contract
  • All impressions delivered are of low value
  • 2 a.m. viewers
  • viewers from poor neighborhoods
  • basically, viewers nobody wanted!
  • Volume vs. price of winning bids on spot marketVolume = number of impressions sold at p ~ price densityPrice on sport market used as proxy for “quality” of impressionPrice pPublisher-Optimal VolumeFind position for the arrow such that area before the arrow = d (GD Advertiser gets the cheapest stuff)PriceAdvertiser-Optimal VolumeFind position for the arrow such that area after the arrow = d (GD Advertiser gets the most expensive stuff)PriceCompromises
  • The GD contract could get half of the bottom stuff and half of the top stuff
  • More fine-grained:
  • Of the supply selling at every price, give d/s fraction to the GD contract.
  • Then, price distribution in GD mirrors the intrinsic distribution in the total supply.
  • Objective function must penalize deviation from this ideal.
  • Problem setting
  • Assume the publisher knows the distribution of the external winning bid on the spot market
  • Notation
  • p = price (winning bid)
  • f(p) = price density = the highest bid is drawn i.i.d. from f
  • s = total supply (inventory) of impressions
  • d = demand (GD volume) for the contract
  • t = target spend per impression (budget)
  • d/s is the fraction of the total supply that needs to be delivered to the (unique!) contract
  • Find an allocation a(p)
  • a(p)/s = fractional allocation to GD at price p, that is:
  • There are s*f(p)*dp impressions available at price p (or rather in interval [p,p+dp)
  • The GD contract gets
  • a(p)/s * s*f(p)*dp = a(p)*f(p)*dpimpressions at price p
  • Ideal: a(p)/s = d/s for all p
  • Objective: close to this ideal
  • u measures distance
  • Allocation Constraints
  • a() is not assumed continuou
  • Related Search
    Related Documents
    We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks