Hero conference 2016 - advanced bidding
7 likes3,825 views
The document focuses on advanced bidding strategies and optimization methods in product marketing, highlighting key dynamics such as conversion success metrics and bidding inertia. It discusses patterns and strategies to improve efficiency, engagement, and conversion rates using various analytical tools and portfolio management techniques. Additionally, it emphasizes the importance of understanding the factors driving conversions while managing budget constraints and optimizing bids across different segments.
![Bid Modifiers
Programmatic
Display
Traditional
SEM
Enhanced
Campaigns
Audience
Keyword
Match Type
[exact]
+broad
Bidding
Geo
Device
Time Category
Frequency](https://image.slidesharecdn.com/heroconference-2016-advancedbidding-external-160425201208/85/Hero-conference-2016-advanced-bidding-49-320.jpg)
Hero conference 2016 - advanced bidding
- 1. Chris Haleua - Sr. Product Marketing Manager Advanced Bidding
- 2. 3x3 = 9 Segments CPA Conversions 155.48 1 270.22 1 CV CE CP NV NE NP IV IE IP Cost Conversions 250 0 190 0 Impressions Position 1,100 8.9 870 7.6
- 3. 3 1 | LAWS - Bidding Dynamics 2 | PATTERNS - Deadend Quadrants 3 | APPLICATION - Optimization Methods 4 | BALANCE - Dimensional Modeling 5 | ADVANTAGE Secret Tools Summary
- 4. Bidding Giants George Michie Brad Geddes Sid Shah Wijnand MeierHal VarianBen Vigneron tiny.cc/wijnandtiny.cc/halvariantiny.cc/vigneron tiny.cc/michie tiny.cc/sidshahtiny.cc/geddes
- 5. Marie Curie no fear of perfection; you'll never reach it. Nothing in life is to be feared; it is only to be understood
- 6. Rene Descartes solved became a rule, which serve afterwards to
- 7. Vilfredo Pareto many events, roughly 80% of the effects come from 20% of the
- 8. Isaac Newton I have seen further than others, it is by standing upon the shoulders of giants
- 9. Newtonian Laws AERODYNAMICS THERMODYNAMICS 1. Inertia 2. Acceleration 3. Equilibrium Transfer Entropy Absolute Zero F=ma A=0m/s2 FA=-FB -W dS 459.67° F
- 10. Newtonian Application MOTION ENERGY 1. Inertia:Things in motion keep going 2. Acceleration: Smaller things speed up faster 3. Equilibrium: Equal & opposite reaction Transfer: Energy becomes heat & work Entropy: Nothing is 100% efficient Absolute 0: No entropy for the coldest
- 11. Search Laws OPTIMIZATION ANALYSIS 1. Bidding Inertia 2. Learning Acceleration 3. Modifier Equilibrium Focus Transfer Volume Entropy Absolute Deadweight
- 12. Search Application OPTIMIZATION ANALYSIS 1. Inertia no contradiction in exclusive segments 2. Acceleration assist & micro > clustering 3. Equilibrium balance volume & efficiency Transfer - filter zero, sort next best Entropy averages lie Deadweight - archive not hoard
- 13. 13 Overcome bid inertia by reducing mass
- 14. 14 Extra focus on the critical few
- 15. 2 Dimensional Cartesian Plane
- 16. 1 | LAWS - Bidding Dynamics 2 | PATTERNS - Deadend Quadrants 3 | APPLICATION - Optimization Methods 4 | BALANCE - Dimensional Modeling 5 | ADVANTAGE Secret Tools
- 17. 3x3 Intersection 17 Conversion Success Non-Converting Spend Ignored Impressions Volume Leaders Efficiency Threats Visibility Struggles Conversions > 0 Conversions = 0 Cost > 0 Cost = 0 Impressions > 0 FILTERS Rev. (Conv.) Cost Imp. ROAS (CPA) Pos. CTR Pos. Q Score Pos. SORTS CV CE CP NV NE NP IV IE IP How are we defending our top terms? How are we avoiding waste? How are we capturing interest? How are we supporting growth? How are we balancing returns? How are we maintaining visibility? How are we qualifying prospects? How are we engaging visitors? How are we demonstrating relevance? How should we invest this extra budget? How should we project the potential? How did we grow 25% YOY in non-brand? Dead-end Quadrants
- 18. FILTERS Volume Leaders Efficiency Threats SORTS Conversion Success Non-Converting Spend Conversions > 0 Conversions = 0 Cost > 0 Revenue ROAS CPCCost Dead-end Quadrants 18 FILTERS
- 19. FILTERS Volume Leaders Efficiency Threats SORTS Non-Converting Spend Ignored Impressions Conversions = 0 Cost > 0 Cost = 0 Impressions > 0 CPCCost Impressions Quality Score Dead-end Quadrants 19 FILTERS Pick up where we left off
- 20. Filter zero, then sort the next best volume to identify the critical few leaders resulting in dead-end bottlenecks
- 21. Dead-ends You might know your total conversions ...but do you know the minority of kw driving the majority of $? You might know your efficiency threats... ...but do you know your deadends in the step before? You might know your loss leaders... ...but do you know the segments & pivots that caused them?
- 22. Keep ratios in context where their volume building blocks are the closest available signal to the bottom line Avg. Pos CTR CPC CVR ROAS ROI
- 23. 23 1 | LAWS - Bidding Dynamics 2 | PATTERNS - Deadend Quadrants 3 | APPLICATION - Optimization Methods 4 | BALANCE - Dimensional Modeling 5 | ADVANTAGE Secret Tools
- 25. 25 3 reports hidden in every keyword report
- 26. Segment by Sorting & Filtering 1. Sort by conversions Identify the first conversion zero Filter out blank conversion data 2. Sort the rest by cost Identify the first cost zero Filter out blank traffic data 3. Sort the rest by Filter zero then sort by the next best metric Conversion Success Non-Converting Spend Ignored Impressions
- 27. 3x3 Segments in Rules ROAS or CPA Cost Impressions Filter zero then sort by the next best metric Segment Metric 3x3 Att Micro Weight Path
- 31. 3 Dimensional Cartesian Coordinates
- 32. Conversions (Last) Conversions (First) Conversions (Assists) 3x3 to 4x3 Keyword Segments ROAS (last) / CPA (last) Cost Impressions Converting Non-Converting Ignored ROAS (first) / CPA (first) Cost Per Assist Attribution Perspectives
- 33. Is this how your funnel feels?
- 34. Prioritize Micro-Conversions According to the Customer Path Traffic Engagement Attribution Conversions Revenue Profit Weight micro-conversions to refine portfolio objectives
- 35. Traffic Engagement Attribution Conversions Revenue Profit Filter zero then sort by the next best metric Brian Shindurling Use the Best Tool Available Until the Fuel is gone Blake Haleua In the absence of your preferred KPI, choose the metric closest to the bottom-line Cameron Cowan When a more precise metric is missing, fall back to the best derivative Chris Haleua Do what you can, with what you have, where you are. Theodore Roosevelt
- 36. Beyond the 3x3 36 Conversion Success Non-Converting Spend Ignored Impressions Volume Leaders Efficiency Threats Visibility Struggles CV CE CP NV NE NP IV IE IP Above: CPA COGS Profit Total On/Offline Between: Attribution Assists & Micro-Conversions Below: Long tail learning or Dead Weight Cleanup
- 38. Forecast Adjustments in Predictive Modeling 38 Massive scale Sparse data Time delays Conversion paths Complex objectives Micro-conversion weighting 𝑀𝑎𝑥 𝑖 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 𝑏𝑖 𝑠. 𝑡. 𝑖 𝐶𝑜𝑠𝑡 𝑏𝑖 ≤ 𝐵 + 𝑏𝑢𝑠𝑖𝑛𝑒𝑠𝑠 𝑐𝑜𝑛𝑠𝑡𝑟𝑎𝑖𝑛𝑡𝑠
- 39. Portfolio Modelling Summary Apply portfolio theory to make bid tradeoffs for maximum marginal return Maximize return for any budget through media mix modeling Build and show granular bid unit model tables for traffic & conversions Combine forecasts with dimensional algorithms for accurate simulations Build trust in forecasts through model accuracy trends by metric & day
- 40. Simulation Yield Curve 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 Average Daily Spend Average Daily Conversions 17% 59%
- 41. -20 -15 -10 -5 0 5 10 15 Average Cost accuracy Average of cost_accuracy_no_dow Average of cost_accuracy_w_dow
- 46. Tough Questions to Ask 3rd Party Tools Method Optimization types available Most Popular optimization types Intraday frequency Advantage Unique differentiators Competitor threats Innovation pace Budgeting Manual overrides Spend strategies Pacing visualizations Modifiers Report segments for traffic / conversions Geo, device, time, audience manually Fully automated dimensional modeling Attribution Intra-channel Cross-channel Cross-device Objectives Position Efficiency Multiple weights Assignment Cross-engine Cross-account Execution granularity Learning Clustering Engagement / Micro-conversions Prediction accuracy
- 47. Method What types of bid strategies are available? Method What is the most commonly-used bid management feature in your platform, and why? Method Can you set automated bidding rules for different segments of keywords? Method Does the system leverage any type of portfolio optimizations? Method For the automated bid changes, is there capability to receive forecasted results from changes prior to implementing? Method Would you share methodology around automated bidding services? Advantage What would you consider your competitive advantage to be over your competitive set? Advantage What are your plans to maintain your competitive advantage and stay at the forefront of innovation? Advantage How often do you innovate and refine your algorithms? Budgeting What budget options exist in the system? (Monthly/Daily Caps) Budgeting What budget pacing tools exist? (can I pull a pacing report, set a monthly campaign cap, and let the system adjust budgets? Budgeting Are there any budget pacing alerts capabilities in the UI? Modifiers Can you assign time/days of week to apply bid multipliers through the system? Modifiers How can device, geo, time, and audience modifiers be automated? Modifiers Does the system have ability to implement any type of day-parting or geo optimization? (ex. increase bids by X% during specific hours) Attribution Please describe how attribution is done across channels (ex. tiered system (if so, is this customized), custom weighting, etc.). Attribution What insight can the platform give into a client's customers path to conversion? Attribution Please describe any attribution settings within search (ex. last click/first click, even distribution across all in path, custom distribution, etc.). Attribution Do you offer custom attribution models (first click, "prefer last," dynamic, etc.)? If yes, describe how you help each team to begin to find the best fit. Manual What tools are in place to more efficiently change bids in bulk (ex. campaign or group setting to up all keyword level bids within group by X%.) Manual Can you change bids, by keyword, across accounts via one bulk sheet? If not, please detail the process needed for cross account optimizations. Manual How do manual bidding & automated bidding co-exist? Assignment Can you assign automated bidding for predefined segments/groupings cross account? Assignment Do you have cross engine bidding strategy? One massive portfolio or engine first? Assignment At what granularity is bid optimization executed? (at the group level, keyword level or both)? Objectives What options exist for bidding to position (ability to target a specific position and let the system adjust bids accordingly) Objectives What options exist for conversion efficiency thresholds? (CPA/ROI goal targeting by a custom formula or set of metrics) Objectives Can you specify specific actions or weighting to count toward the goal? Can I state my goals to match how my business really works?
- 48. 1 | LAWS - Bidding Dynamics 2 | PATTERNS - Deadend Quadrants 3 | APPLICATION - Optimization Methods 4 | BALANCE - Dimensional Modeling 5 | ADVANTAGE Secret Tools
- 49. Bid Modifiers Programmatic Display Traditional SEM Enhanced Campaigns Audience Keyword Match Type [exact] +broad Bidding Geo Device Time Category Frequency
- 50. Difficult Decisions 50 First PPC Platforms Enhanced Campaigns AdWords Facebook Ads Real-Time Bidding Shopping Campaigns Complexity
- 52. Device
- 53. Geo
- 54. 54 1 | LAWS - Bidding Dynamics 2 | PATTERNS - Deadend Quadrants 3 | APPLICATION - Optimization Methods 4 | BALANCE - Dimensional Modeling 5 | ADVANTAGE Secret Tools
- 55. FREE STUFF! Bid Rule Playbook: tiny.cc/rule Bid Rule Library: tiny.cc/templaterule 3x3 Template: tiny.cc/3x3template 3x3 Macro: tiny.cc/3x3macro I hear and I forget. I see and I remember. I do and I understand.
- 57. Why 57
- 58. What 58
- 59. How 59
- 60. Key Takeaways 60 Performance Segments Matched Metrics 3x3 Pattern Micro- Conversions Attribution Perspectives Weighted Objectives Conversion Paths Excel Template Filters Saved Views Sorts Multi-Purpose Rules If Zero, Sort the Next Best Metric Look Beyond Last for Influencers Refine Portfolio Objectives Multi-Metric Simulations
- 61. Certified Knowledge Audit Series on Bidding tiny.cc/wijnand
- 62. QUESTIONS?