Moderator: Matt Van Wagner, President, Find Me Faster Q&A Moderator: Barry Smyth, Director, Search Strategies
Kevin Lee, Executive Chairman & Co-Founder, Did-It is up first.
Does Bid / Campaign Management Matter? - Maximize the campaign's opportunity - Manage Complexity - Reduce HR time required to manage campaigns - Provide actionable reporting while taking its own - Provide a platform for hypothesis
Auction pressure creates a winner's curse. By that he means, half the top there is a brilliant marketer and the other half the time you are the idiot. You may be competing against marketers who are acting irrationally.
Bid Management Overviews: 1996 - 2001: SEO evolves, paid bid alerts, did it, go to, google missed rest...
APIs were originally used for just managing bids. Overtime, the engines made commitments to make the features of the GUI available directly in the api.
If bid management is going to make an intelligent decision on your behalf it has to predict the best move. The objective is to figure out which clicks are the most valuable. You can do personalized click routing, profiling and so on.
Bid management technology has to react to real time markets. Bid landscape is changing, inventory fluctuations, changes in elasticity, delta between bid and billed CPC. changes in conversion due to a variety of causes, changes in conversion value/quality, inventory levels, and budget changes.
How do these technologies evolve? - Engine level tools, self service, full service - High volume data and low data volumes - Levels of control - Sensitivity adjustments - Utility to use blended success metrics - Testing suites / click routing and variable appending
Never-ending optimization concept. The feedback loops and constant testing starting at the high volume segments.
You need to decide which metrics you will use. Think also about segmentation levers such as geographic, dayparting, day of week, network source, demographics and so on.
Chris Zaharias, VP Search Sales, Omniture is next up. He first shows some cool equations of bid management examples. Point is, he as a marketer doesn't understand all the crazy math...
Bid management is aggregating data, data modeling that data, optimization based on it and bid changes. Behind this are the user interfaces with forecasts. The reporting interfaces have a way to go. Good bid management systems should show you a forecast of what you need to do and spend. What would you get from position one versus position two or three and so on.
What data signals do bid management tools have? - Impressions - CTR - Clicks - Conversins - Orders - ROAS - CPA
What we are missing in this includes: - Micro metrics, micro conversions such as - Product Views - Carts - Cart Adds - Cart Removals - Checkouts - Average order value - Cost Per Step Metrics - Step By Step Conversion Rates
Data Limitations / Uncertainties - Traffic Estimation -- 30-50% off, 70% of the time -- Solution is to leverage your data - Revenue Forecasting -- Impressions clicks - Recency
Types of Bid Management - Rules based lets you control things but lack of certainty and time issues - Portfolio - Algorithmic
Don't buy on price alone. Know what model you want before going on. He mentions the Yahoo thing today and Google's auto-matching.
George Michie, Principal, Search Marketing, Rimm-Kaufman Group is next up.
Requirement #1: - Proprietary, Sound Statistical Foundation -- Bid based on observation value, not position -- Bid as granularly as statistically possible -- Variable balance btw recency and volume -- Smart mechanism for folding in ancient history -- Treats low CR/high AOV terms differently than the converse
Requirement #2: - Robust and Flexible Tracking -- Redirect or JavaScript -- Item level tracking -- Sales, Margin, Leads or Hybrids -- Variable cookie windows -- Support internal analytics systems -- Smart Brand, Non-Brand Allocation
Requirement #3: - Atomic bidding
Requirement #4: - Back-feed capability -- Frauds and Cancels -- Margin -- New to file vs house file -- Business vs Consumer vs Government -- Order Allocation -- Account funding / lead valuation
Requirement #5: - API Integration
Analytics Imperatives #1 - Right balance between human and machines -- Anticipatory bidding
Analytics Imperatives #2 - Day parting with brains
Analytics Imperatives #3 - Match type, syndication and geo targeting -- bidding needs to handle differences in traffic quality
Analytics Imperatives #4 - Flexible keyword attribute definitions
Analytics Imperatives #5 - Deep, Flexible Filtering Field - Granular level - Very deep as possible
David Rodnitzky, CEO and Founder, PPCAdBuying.com is last up.
Bid management is one piece of the pie. There is ad text optimization, there is keyword selection, landing page optimization and much more. Bid management software is also not magic.
There are lot of considerations around bid management software. Am I qualified to evaluate these software? What do I want to do internally, what do I want my bid management company to handle? Can I absorb the additional cost of bid management and still meet my business objectives. Do I have engineering resources available to integrate the software? How will I define success? Will I be able to benchmark success.
Questions to Ask: How difficult is integration and disintegration? What do you offer beyond bid management? Do you offer a free trial? Are there additional feeds? Who pays if your system/team makes an error? How do you handle returns/offline transactions/other special needs? Who will be my day to day contact? Does your system support my specific business objectives?
More questions: Can you get a short contract our out clause? Keep up to date on vendors? Where you wowed by a powerpoint? Do you really need to vie your head keywords? Have you considered performance pricing alternatives? How reliable are vendor references?
