Leveraging the Crowd for SEO SMX Advanced June 8, 2011 Natala Menezes Sr Product Manager Amazon Mechanical Turk
What is Crowdsourcing? Distributed problem solving Mechanical Turk makes crowdsourcing easy.
Mechanical Turk is a marketplace for work. Mechanical Turk gives businesses and developers access to an on-demand, scalable workforce. Flexibility: Scale your workforce up and down quickly Accuracy: Get high quality, efficient and cost effective results. Price: Pay only when you are satisfied with the results. Speed: Start receiving results in minutes
How it works Micro payments through Amazon Flexible Payments Huge workforce: Over 500,000 Workers, 190 countries Platform can handle millions of tasks, via Web or API
Business Use Cases Categorization Classification Tagging Sentiment Analysis Data Management Data Verification Data Entry & Collection Data De-duplication Algorithm Training Content & Media Moderate Photos & Content Content Creation & Editing Transcription Business Services Search Relevancy Product Usability Testing Research
Search Experts <3 Big Data Search Marketing: Keyword expansion Adding attribute tags Pulling lists from publicly available tools De-duplication / data cleanup and organization Search query research “How would you search for X?” Web Research Competitive research – What is the tagline for this company? What is the SEOMoz Score for this website? Search Results Analysis Of the two results shown – which is the best result?
Content Creation Article writing & editing Write a caption for a photo Summarize the benefits of a product Write a story about a specific topic Edit the grammar & style of an article Photo selection Find a creative commons photo for an article Transcription Transcribe video or audio content Tip: Always collect content in your HIT, never have Workers post directly to a 3rd party website.
Content Management Content Moderation Is this photo appropriate given our guidelines? Is this a good question or article for our site? Content Discovery Do these photos have useful meta tags? What are the keywords used during this video? Content Analysis What is the sentiment of this tweet? Is this tweet positive or negative?
Use Case: Lead Collection The Problem: Needed to create a list of local business contacts that met certain criteria: Review rating above 3 stars Vertical focus – family friendly restaurants only Within a certain distance of downtown The Solution Posted tasks to Mechanical Turk to gather information about potential contacts. Used this data to prioritize their list, then took the smaller subset for additional data augmentation
Case Study: Magnum Photos “An image sent out through the new system will come back, having been keyworded by up to eight people, in less than a minute. After piloting small trials last summer, images are now being sent out in batches of 20,000.” "You can keyword an entire archive within weeks.“ -- Meagan Young, Magnum's Web Content ManagerQuickly removed their backlog through access to on-demand workforce 10 x more cost efficient than an in-house solution
Things you can’t do: Disrupting or degrading the operation of any website or internet service Ex. Generate "referred" site visits or click through traffic Go to this website and click on the most valuable ad Ex. Ask Workers to take action to manipulate a website’s behavior or results Go to X search engine and search for “Danny Sullivan is Awesome” Violate the terms and conditions of an activity or website (for instance asking Workers to vote for something) or that directly or indirectly promote a site, service, or opinion or ask Workers to solicit third parties Ex. Ask a Worker to perform a marketing activity on your behalf – for instance voting for something “Tweet This” or “Like” Ex. Ask a Worker to write a review on a 3rd party site Go to Yelp and write a review for this restaurant Capturing PII Ex. requiring disclosure of the Worker's identity or e-mail address, either directly or indirectly Ex. requiring registration at another website or group https://requester.mturk.com/help/faq#restrictions_use_mturk
Applications & Solution Providers Mechanical Turk supports a robust ecosystem of providers that have built or can build solutions to meet your needs.
Framework What is crowdsourcing What is mturk What can you do w/ MTurk: big data tasks Keyword expansion Competitive research / web research Content Creation How to access Direct or partners A note on policy
Content Discovery Search discovery requires attributes – keywords that users can use to find content that relates to them. Adding meta data – such as detailed descriptions, attribute tags and categories can improve discoverability. How it works:
Search Enhancement / Relevance Search discovery requires categorization and expansion or classification of attributes – keywords that users can use to find content that relates to them. Adding meta data – such as detailed descriptions, attribute tags and categories can improve discoverability. How it works:
How it works:. “Validate, Pay & Go” “Design & Publish”