Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Victor Pan - When is cloaking a good idea?

Victor Pan - When is cloaking a good idea?

Tech SEO Connect

October 23, 2024
Tweet

More Decks by Tech SEO Connect

Other Decks in Marketing & SEO

Transcript

  1. Trivia Time! *Side-effects of cloaking include lost website rankings, organic

    traffic, a manual penalty, difficulty sleeping at night, nightmares , anxiety, depression, and spying on how other people search on Google. Which two countries allow for medical ads? Cloaking Improves SEO*
  2. Important Cloaking Disclaimer* *This deck was meant for an audience

    of SEO’s with 10+ years of experience Cloaking Sanity Check Is this a good idea? I have a long-term technical team Yes/Maybe The SEO & admins are part-time No If you document changes others in your company will care Yes/Maybe You don’t know how to handle, maintain, and test IP-cloaking to search engines No I am aware of the risks of cloaking includes losing traffic and revenue that could put people I care about at risk. Maybe
  3. User-agent:* Disallow: / Victor Pan (He/Him/先生) • Does weird things

    at a B2B SaaS company that tracks traffic to $ • Cares about making the web a better place • My favorite human condition is awkward tension
  4. Cloaking un-agenda Disclaimer & intro GDPR & the cookiepocolypse How

    adding parameters within internal links broke Google When cloaking becomes a solution When is cloaking a good idea? 04 MIN 05 MIN 06 MIN 07 MIN 08 MIN
  5. E.g. Creating an account doesn’t count as consent. Didn’t accept

    cookie consent? You’re a direct conversion.
  6. How much GDPR exposure does your website have from EU

    citizens? Austria Belgium Bulgaria Croatia Cyprus Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Latvia Lithuania Luxembourg Malta The Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden *United Kingdom I’ll save you some time. Don’t use GSC Search Analytics to quantify this. *Passed GDPR before leaving the EU
  7. Apple introduces Intelligent Tracking Protection for iOS. 2017 2018 2019

    2020 2021 2022 2023 2024 Firefox Enhanced Tracking Protection Chrome declares they’ll be phasing out 3rd party cookies Chrome introduces Privacy Sandbox breakthroughs Google shifts stance on fully killing 3rd party cookies in July Appendix: Cookiepocolypse Timeline (Most internet users use Chrome) Don’t sit on this. It’s coming.
  8. Appendix: What could you do different? What could go wrong?

    • Use # (hash, anchors) provided it doesn’t conflict with your analytics backend • Unify to one parameter where event combinations are encoded • Server-side/hybrid analytics using first-party data only • …and much more.
  9. The business of B2B SaaS SEO 1 Paying Customers $$$

    (Customer Lifetime Value) 3 Channel Conversion Rate (hubs_signup-cta) 5 Search Total Addressable Market (Topics -> Products) 2 Leads/Demos/Free Sign-ups CVR (estimated $) 4 Feasible Market Share by CTR (hubs_signup-url) You need numbers to build or expand an SEO team
  10. Mathing influence 1 If we assume $1000 value for a

    customer, $1M* 3 Channel Conversion Rate (1%)* 5 Search Total Addressable Market (100M/M)* 2 Book Demo->Paid Customer CVR (10%)* 4 Feasible Market Share by CTR (1%)* *Numbers/channels simplified, but this is how headcount/a bonus/promotion could math out.
  11. What could go wrong with adding parameters to internal links?

    I’ll believe Gary when it’s in the help docs
  12. What could go wrong? More crawl = better rankings right?

    Cctlds were also impacted negatively
  13. What could go wrong? Part II Our kafka swimlane may

    have been shared with customers
  14. What could go wrong? Part III Sites that are still

    “polluted” with duplicate content from our parameters
  15. Fixing things (short term) because we wanted to maximize Page

    Rank We get around 300+ parameters. Based on log files, the parameter handling tool was not effective
  16. The bigger fix - block javascript and parameters that don’t

    impact render iplookup is used for personalization and returns an X-robots-tag: none
  17. Duplicates - From 200:1 to 20:1 Big win on technical

    SEO debt, but still room for improvement…
  18. Small difference, but this is still technically cloaking Person sees

    Bots see No, we did not get a manual penalty from doing this
  19. Why stop here? *Let’s chat if you’ve already done this

    • Users don’t need schema, OG, and other metadata for bots • Bots don’t execute non-rendering scripts (ad tracking, heatmaps, tag manager) • 301 redirect parameterizations to canonical via CDN dynamic redirect rules*
  20. Appendix: Dumb Cloudflare Dynamic Redirect Example Example Cloudflare dynamic redirect

    rule that would 301 redirect all parameterizations non-param. This would ruin a site where ?page=N is used for pagination. Do your own SEO due diligence.
  21. Cloaking may be good idea when* *Disclaimer: Caveat Emptor •

    What you’re serving users is essentially the same as search engines (in spirit) • You can get some credit for reducing bytes on the wire ◦ this could be reduced server costs ◦ or a sustainability effort where you work • It’s not cloaking, it’s personalization
  22. The more data Googlebot is downloading, the more room for

    improvement?* *Screenshot intentionally given without context. What comes after T? Q/S/G?
  23. Fuck yeah. Tech SEO solving for global warming with make-believe

    trees h/t Arnout. Crawls X bytes not rendered X trees saved = tree hugging tech SEO math Websitecarbon.com
  24. Why have browsers resize images when you can serve it

    on the edge? h/t Jono I’m a huge fan of brotli compression and not resizing images
  25. Call-to-Action - gotta’ have one *If you’re cloaking for good,

    I’d love to hear from you. Let’s make robots.txt cloaking official for starters. fucking great idea!!! *