Content Strategy Applied 2014 
Don’t Fear the Sweeper: Scaling 
Content for Large Ecommerce Sites 
@heymatthobbs 
#sweeper14 
Matt Hobbs 
Senior User Experience Manager 
at National Builder Supply
Don’t Fear the Sweeper
A thoughtful approach to building a 
coherent ecommerce catalog… 
…that combines product data points, 
manufacturer-provided content, and 
original content. 
What you’ll hear today
It’s kinda like building a house with a team of experts….
Content 
Insanity in. Insanity out. 
Manufacturer A 
Product Info 
Content 
Manufacturer B 
Product Info 
Content 
Manufacturer C 
Product Info 
Content 
Product Info 
Content 
Product Info 
Product Info 
Content 
Manufacturer D 
Product Info 
Content Product Info
1. What content, data points drive consideration, 
conversion 
2. Translating fragmented incoming data into a 
pretty result 
3. Sweet tools to keep up with changes, errors 
4. Layering on original content to fill in gaps and 
snazz it up a bit 
Today’s agenda
First things first 
Defining product categories 
for all SKUs
Clear organization of products for users and spiders alike
Our most precious content 
Why we care so much about 
product pages
Product pages make up more than 
85% of our site’s content. 
Mmm… pie.
Product pages are your best shot at 
driving organic search traffic. 
Scale your army: programmatically updating 95K pages 
http://libertycitys.com/wp-content/uploads/2014/11/green-army-men.jpg
It’s where most of our users start their journeys
It’s where users make their most important decision: Buy or No
Welcome to the jungle, we’ve got alliteration 
Crucial content that causes 
consideration, conversion
Your most important content: Price, Image, Availability
Spec sheets, marketing copy, bullet points, product specs
And don’t forget product reviews, customer service Q&A’s
Whittling down result sets with faceted navigation
What do all these content 
types have in common? 
Good question, Morgan.
Most of this content is maintained in databases 
http://sqlbak.com/blog/wp-content/uploads/2014/01/SQL.png
Tactics to ensure content in databases drives clean UX 
Translating manufacturer data 
into content, navigation
A bit of history 
http://media.komonews.com/images/120630_johann_strauss_jr_660.jpg
Y’all, our old system was pretty rough. 
http://www.quickmeme.com/meme/35n1o9
Noisy data leads to sloppy product pages, ugly navigation 
http://www.polyvore.com/cgi/img-thing?.out=jpg&size=l&tid=8772030 
Bronze 
Classic Bronze 
Deep Bronze 
Stainless Steel 
Stainless steel 
Stainless_steel
Our new system: 
Human Decision Factors (HDFs) 
A clearly-defined set of data points and 
content designed to help users decide if a 
product is right for them. 
Focus on the content that helps users make decisions
Example HDFs for category “Toilets”
Example: Navigable HDFs for category “Toilets”
Insights from 
Sales and Support 
What’s Available 
from Manufacturers 
Competitive 
Assessment 
HDFs 
Define categories and curate clean list of HDFs for each
1. Is this HDF navigable or not? 
2. Priority of navigable decision factors (order 
in column) 
3. Setting consistent answers, syntax, format 
Building out HDFs for a category – more considerations
Database 
Developers 
Customers Manufacturers 
When you establish clear parameters, everyone’s happy!
!!! 
A win we weren’t counting on: Faster load times
Regular sweeps to keep things clean 
Ongoing optimization of 
catalog content
So how can we keep up?
Our old system: FRUSTRATION 
Find errors 
New products 
Brief eternity 
MASSIVE 
development queue
User interface that empowers business users to edit product pages in real time 
New System: CAT MAN
An “editable version of the product page”
Customer service reps Hot Night sweepers 
Who uses Cat Man? 
Interns 
Outsourced teams
Filling in the gaps 
Adding original content to 
make it awesome
Sometimes you’re given weak, irrelevant, or not-unique content.
6000 Elkay SKUs 
Programmatically write unique content to cover ground
TOTO is a huge fan of talking about their G-Max 
Flushing System, but sometimes they forget: 
Most shoppers have no idea 
what they’re talking about. 
Find the right mix of manufacturer Kool-aid and plain explanations
Detailed translations of cloudy manufacturer value props
Helpful info for customers and sales team alike
Sweeten the deal with clever supporting content
If you can validate your hunches with data-driven 
consumer insights, 
you can be confident your content will 
resonate with your audience. 
Hunch + Data = Great Supporting Content
Where our team finds consumer insights
Almost done here… 
Wrapping up
Don’t publish all the data you receive; make sure 
content and data are relevant to customer decisions 
(insights from sales team, keywords, competitors). 
Build sweet tools like Cat Man to streamline manual 
efforts for updating and editing product content. 
What we covered this morning (30,000 feet)
Programmatic writing is a great way to fill in gaps for 
copy blurbs, bullet points. 
Add in cool supporting content to clarify manufacturer 
Kool-aid or invent other cool uses for products. 
What we covered this morning (30,000 feet)
Come with me to Ecommerce Paradise
Thanks y’all! 
Any questions? 
Feel free to connect with me on Twitter: 
@HeyMattHobbs

More Related Content

PDF
Power-Up Your PPC - Get Started With Audiences - BrightonSEO April 2017 - Ari...
PPTX
Onsite SEO Webinar
PDF
#UKHUG 2016: Common Technical SEO Issues - LexisClick and bandv
PPTX
Actionable insights with Google Analytics - Edinburgh Chamber of Commerce
PPTX
Ecommerce dev for business needs
PDF
Kidlandia 2011 Web Strategy
PDF
Nosto (2017) The Anatomy of a Successful Ecommerce Store.pdf
PDF
Essence of Retail e-Commerce and its Optimization Webinar
Power-Up Your PPC - Get Started With Audiences - BrightonSEO April 2017 - Ari...
Onsite SEO Webinar
#UKHUG 2016: Common Technical SEO Issues - LexisClick and bandv
Actionable insights with Google Analytics - Edinburgh Chamber of Commerce
Ecommerce dev for business needs
Kidlandia 2011 Web Strategy
Nosto (2017) The Anatomy of a Successful Ecommerce Store.pdf
Essence of Retail e-Commerce and its Optimization Webinar

Similar to eBay Content Strategy Applied 2014 (20)

PDF
Jane_clark_pres2016_FUP_web
PPTX
Fantastic Feeds & How to Mind Them
PDF
MivaCon Seattle - On-Page SEO & Miva Merchant 2015: Properly Structuring Your...
PPTX
Website upgrade strategy
PDF
Expert Panel Session - SEO for Ecommerce - ionSearch 2012
PPTX
The On-page of SEO for Ecommerce - Adam Audette - SearchFest 2013
PDF
Advertising with Linked Data in Web Content
PPTX
Product Creation & Managment Webinar Deck
PDF
Goodrelations Presentation from SemTech 2010
PDF
Internet Retailer Web Design 2014 - Product Page
PDF
Dominating DTC Product Page SEO
PDF
Mid-2017 Product Update Webinar
PPTX
ionSearch Ecommerce SEO Presentation
PPTX
MivaCon 2013 - The New Normal
PPT
Top 10 website needs for manufacturers- Webinar held January 30, 2014
PDF
Conversion Hotel 2014: Karl Gilis (BE) keynote
PPTX
TFM - Using Google Tag Manager for ecom
PPT
Catalog Design Revolution The Consumer Demand for a New Medium
PPT
SMX West 2012 - Duplication, Aggregation, Syndication, Affiliates, Scraping &...
PDF
Building Device Agnostic UX systems - GeekGirl, London 28 Oct 2014
Jane_clark_pres2016_FUP_web
Fantastic Feeds & How to Mind Them
MivaCon Seattle - On-Page SEO & Miva Merchant 2015: Properly Structuring Your...
Website upgrade strategy
Expert Panel Session - SEO for Ecommerce - ionSearch 2012
The On-page of SEO for Ecommerce - Adam Audette - SearchFest 2013
Advertising with Linked Data in Web Content
Product Creation & Managment Webinar Deck
Goodrelations Presentation from SemTech 2010
Internet Retailer Web Design 2014 - Product Page
Dominating DTC Product Page SEO
Mid-2017 Product Update Webinar
ionSearch Ecommerce SEO Presentation
MivaCon 2013 - The New Normal
Top 10 website needs for manufacturers- Webinar held January 30, 2014
Conversion Hotel 2014: Karl Gilis (BE) keynote
TFM - Using Google Tag Manager for ecom
Catalog Design Revolution The Consumer Demand for a New Medium
SMX West 2012 - Duplication, Aggregation, Syndication, Affiliates, Scraping &...
Building Device Agnostic UX systems - GeekGirl, London 28 Oct 2014
Ad

Recently uploaded (20)

PDF
WOOl fibre morphology and structure.pdf for textiles
PDF
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PDF
Enhancing emotion recognition model for a student engagement use case through...
PPTX
observCloud-Native Containerability and monitoring.pptx
PDF
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
DOCX
search engine optimization ppt fir known well about this
PDF
Assigned Numbers - 2025 - Bluetooth® Document
PDF
Zenith AI: Advanced Artificial Intelligence
PDF
Unlock new opportunities with location data.pdf
PDF
Getting started with AI Agents and Multi-Agent Systems
PDF
sustainability-14-14877-v2.pddhzftheheeeee
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PPTX
Tartificialntelligence_presentation.pptx
PDF
Architecture types and enterprise applications.pdf
PDF
Getting Started with Data Integration: FME Form 101
PPTX
Benefits of Physical activity for teenagers.pptx
PDF
August Patch Tuesday
PDF
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
PDF
Five Habits of High-Impact Board Members
WOOl fibre morphology and structure.pdf for textiles
DASA ADMISSION 2024_FirstRound_FirstRank_LastRank.pdf
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
Enhancing emotion recognition model for a student engagement use case through...
observCloud-Native Containerability and monitoring.pptx
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
search engine optimization ppt fir known well about this
Assigned Numbers - 2025 - Bluetooth® Document
Zenith AI: Advanced Artificial Intelligence
Unlock new opportunities with location data.pdf
Getting started with AI Agents and Multi-Agent Systems
sustainability-14-14877-v2.pddhzftheheeeee
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
Tartificialntelligence_presentation.pptx
Architecture types and enterprise applications.pdf
Getting Started with Data Integration: FME Form 101
Benefits of Physical activity for teenagers.pptx
August Patch Tuesday
Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor
Five Habits of High-Impact Board Members
Ad

eBay Content Strategy Applied 2014

  • 1. Content Strategy Applied 2014 Don’t Fear the Sweeper: Scaling Content for Large Ecommerce Sites @heymatthobbs #sweeper14 Matt Hobbs Senior User Experience Manager at National Builder Supply
  • 3. A thoughtful approach to building a coherent ecommerce catalog… …that combines product data points, manufacturer-provided content, and original content. What you’ll hear today
  • 4. It’s kinda like building a house with a team of experts….
  • 5. Content Insanity in. Insanity out. Manufacturer A Product Info Content Manufacturer B Product Info Content Manufacturer C Product Info Content Product Info Content Product Info Product Info Content Manufacturer D Product Info Content Product Info
  • 6. 1. What content, data points drive consideration, conversion 2. Translating fragmented incoming data into a pretty result 3. Sweet tools to keep up with changes, errors 4. Layering on original content to fill in gaps and snazz it up a bit Today’s agenda
  • 7. First things first Defining product categories for all SKUs
  • 8. Clear organization of products for users and spiders alike
  • 9. Our most precious content Why we care so much about product pages
  • 10. Product pages make up more than 85% of our site’s content. Mmm… pie.
  • 11. Product pages are your best shot at driving organic search traffic. Scale your army: programmatically updating 95K pages http://libertycitys.com/wp-content/uploads/2014/11/green-army-men.jpg
  • 12. It’s where most of our users start their journeys
  • 13. It’s where users make their most important decision: Buy or No
  • 14. Welcome to the jungle, we’ve got alliteration Crucial content that causes consideration, conversion
  • 15. Your most important content: Price, Image, Availability
  • 16. Spec sheets, marketing copy, bullet points, product specs
  • 17. And don’t forget product reviews, customer service Q&A’s
  • 18. Whittling down result sets with faceted navigation
  • 19. What do all these content types have in common? Good question, Morgan.
  • 20. Most of this content is maintained in databases http://sqlbak.com/blog/wp-content/uploads/2014/01/SQL.png
  • 21. Tactics to ensure content in databases drives clean UX Translating manufacturer data into content, navigation
  • 22. A bit of history http://media.komonews.com/images/120630_johann_strauss_jr_660.jpg
  • 23. Y’all, our old system was pretty rough. http://www.quickmeme.com/meme/35n1o9
  • 24. Noisy data leads to sloppy product pages, ugly navigation http://www.polyvore.com/cgi/img-thing?.out=jpg&size=l&tid=8772030 Bronze Classic Bronze Deep Bronze Stainless Steel Stainless steel Stainless_steel
  • 25. Our new system: Human Decision Factors (HDFs) A clearly-defined set of data points and content designed to help users decide if a product is right for them. Focus on the content that helps users make decisions
  • 26. Example HDFs for category “Toilets”
  • 27. Example: Navigable HDFs for category “Toilets”
  • 28. Insights from Sales and Support What’s Available from Manufacturers Competitive Assessment HDFs Define categories and curate clean list of HDFs for each
  • 29. 1. Is this HDF navigable or not? 2. Priority of navigable decision factors (order in column) 3. Setting consistent answers, syntax, format Building out HDFs for a category – more considerations
  • 30. Database Developers Customers Manufacturers When you establish clear parameters, everyone’s happy!
  • 31. !!! A win we weren’t counting on: Faster load times
  • 32. Regular sweeps to keep things clean Ongoing optimization of catalog content
  • 33. So how can we keep up?
  • 34. Our old system: FRUSTRATION Find errors New products Brief eternity MASSIVE development queue
  • 35. User interface that empowers business users to edit product pages in real time New System: CAT MAN
  • 36. An “editable version of the product page”
  • 37. Customer service reps Hot Night sweepers Who uses Cat Man? Interns Outsourced teams
  • 38. Filling in the gaps Adding original content to make it awesome
  • 39. Sometimes you’re given weak, irrelevant, or not-unique content.
  • 40. 6000 Elkay SKUs Programmatically write unique content to cover ground
  • 41. TOTO is a huge fan of talking about their G-Max Flushing System, but sometimes they forget: Most shoppers have no idea what they’re talking about. Find the right mix of manufacturer Kool-aid and plain explanations
  • 42. Detailed translations of cloudy manufacturer value props
  • 43. Helpful info for customers and sales team alike
  • 44. Sweeten the deal with clever supporting content
  • 45. If you can validate your hunches with data-driven consumer insights, you can be confident your content will resonate with your audience. Hunch + Data = Great Supporting Content
  • 46. Where our team finds consumer insights
  • 47. Almost done here… Wrapping up
  • 48. Don’t publish all the data you receive; make sure content and data are relevant to customer decisions (insights from sales team, keywords, competitors). Build sweet tools like Cat Man to streamline manual efforts for updating and editing product content. What we covered this morning (30,000 feet)
  • 49. Programmatic writing is a great way to fill in gaps for copy blurbs, bullet points. Add in cool supporting content to clarify manufacturer Kool-aid or invent other cool uses for products. What we covered this morning (30,000 feet)
  • 50. Come with me to Ecommerce Paradise
  • 51. Thanks y’all! Any questions? Feel free to connect with me on Twitter: @HeyMattHobbs

Editor's Notes

  • #2: Matt Hobbs, Content Strategy Lead at National Builder Supply down in Atlanta, GA. National Builder Supply is… Before NBS, spent time working for a digital agency in Midtown Atlanta, doing research (technographics, user behavior trends, etc), SEO oversight, channel strategy for a major web builds and campaigns.
  • #3: Title slide – big picture of blue oyster cult and a broom (plug how I like musical puns – musical improviser)
  • #4: This is what happens when a UX / Content strategy guy gets involved with revamping a database driven catalog – this is just a different way of thinking when it comes to managing content – but a crucial one! Because you can’t let developers run this shit. That’s how we started. This year was a lot of time working with them to retroactively make this work. A pain, as you could imagine.
  • #5: 100,000 SKUs from more than 80 manufacturers It’ll be ok – they’ll provide content and product data. it’ll be easy! They even have videos (which they’re super proud of). It’s like asking your cousins, aunts, and uncles help you come build a house. Each can help with a few tasks, but you ALWAYS end up having to sweep through after them and make it better. Sure they helped you put together each room, but if each room turns out completely different – that’s a loss
  • #9: Search engines put a lot of emphasis on nesting of ideas and hierarchy Assign products/articles anything to a node in a primary path.
  • #12: Google loves product pages – when was the last time you searched for a product and saw an amazon result set? It was a product page. Google PLAs? Product pages (better data here helps them too) Maybe poll the audience – how many people just go straight to amazon or ecommerce site to search for products? (and not start with google or bing?)
  • #13: Product pages are primary landing pages for the bulk of our paid channels, organic search channels, and direct links.
  • #16: Price, image, availability – two of three most important pieces of content are controlled programmatically
  • #18: Dynamic content that changes a lot – user generated. Do I even need this slide? – not relevant to catalog
  • #19: Price, image, availability – two of three most important pieces of content are controlled programmatically
  • #20: They’re managed by databases. And require programmatic updates. Maybe a – wait a minute matt (a la morgan freeman slide – all these are managed by databases? A little different animal than writing articles?)
  • #21: OREOS! They’re managed by databases. And require programmatic updates. Maybe a – wait a minute matt (a la morgan freeman slide – all these are managed by databases? A little different animal than writing articles?)
  • #23: strauss
  • #24: Our old system: give us everything – we want any and all data about products Result: noisy product information and noisy navigation for filtering result sets (database-driven navigation)
  • #25: Our old system: give us everything – we want any and all data about products Result: noisy product information and noisy navigation for filtering result sets (database-driven navigation)
  • #26: We used to call them “questions” or “Facets” – they still are that, but we needed to align them moreso with our organization’s service focus Result: what matters to customers Examples! Also – gets rid of noise and bullshit – like light bulbs for faucets, etc Human benefits – focus on pro and non pro audience Search crawlers, seo
  • #27: EVERYTHING THAT’s HERE is here for a reason – no more haphazard stuff Also needs to have an image, etc Not all of these are navigable
  • #29: Also search volume
  • #30: HDF Section Step 1 – take an example product category Step 2 – list out a bunch of possible descriptors / data points Step 3 – classify each one direction by audience as “pro-worthy” “retail-worthy” “Universal” or “wtf” – where we validate our decisions – inputs include what data points do pretty much all manufacturers give us? (if everyone provides it, it might be important) what do our competitors use? What do our customers (pro and retail) ask for on the telephone – example – 1 piece vs 2 piece toilets is ALWAYS step one in helping a customer whittle down a decision. Step 4 – classify each another direction by navigable or non-navigable GOAL: we’re a service focused organization – my area is technology tools – to replicate the service and logical thought processes our trained product experts give customers on the telephone. Other notes: setting conventions to follow – measurements… length is always left to right, etc – mention phase three – extra boxes like Best Buy or tool tips like Build.com
  • #31: Database developers: cleaner parameters for which data is required (easier automation) Customers: they don’t have to sift through nonsense to find their personal HDFs Manufacturers: easier to anticipate our data and content needs
  • #32: Our new system: Human Decision Factors (HDFs) Added benefit – because we didn’t have to look through all facets for each category, page load times dropped, organic search indexing rise. SCORE! Show graph.
  • #34: New products, discontinues, updates, we find a way to layer on new hdfs to sell widgets better as we learn more about customer preferences (how can we make their lives easier) – plus adding new brands
  • #35: Finding a lot of opportunities, waiting forever for nothing, ugly navigation. And that guy is me (the product owner) and the customer. We
  • #36: Our new system: Cat Man User interface for non-developers Rolling out hot night
  • #37: Our new system: Cat Man User interface for non-developers Rolling out hot night
  • #40: Why elkay? A main brand of ours, TONs of other etailers have the same copy and don’t do much extra to sell them, a popular brand. So based on REVENUE impact to our organization, we move on elkay because we think we can win. Covering ground for 6000 products – same elkay blurb for every water fountain – present on every other ecommerce site…
  • #41: That’s why categorization is key – you rely on that system to prgrammatically bucket out how to write this stuff
  • #42: Use this when writing those marketing copy blurbs – of there’s more to explain… (next slide) for supporting buyer guide content
  • #43: Buyer Guides for customers Appropriate for each audience – some customers need to understand the difference between a widespread or centerset faucet – a plumber however doesn’t have much interest in reading about what makes pvc different from cpvc and when it’s used… he already knows. And just wants to see your price.
  • #44: Training guides for sales and support teams Example – rough plumbing 101 guide Taking a team of people who knew enough about finish plumbing (most people know what a faucet is) – and making them plumbing supplies experts? How we learned – trips to warehouse, walking aisles at home depot (believe it or not)
  • #45: And then – less buyer guide explaination articles and more open topics But THEN – if you have the products in the catalog and the audience there, why not try to make worlds collide? “Uses for pvc” article – we sell that shit! Social media currency, etc
  • #47: Also Pinterest trends? These insights are huge – also google adwords keyword planner toolsfor general keyword trends – align with topics your users are demanding content about