SEO for large sites is completely different than SEO for smaller sites. Large sites have a strong (yet often overlooked!) lever that can boost rankings: internal linking! However, it can be challenging to understand which pages have the highest PageRank, so that you can tweak them to serve important pages better. That can only be determined when you combine internal and external PageRank. Join Kevin Indig as he presents an innovative approach that merges data from crawls, log files, and backlinks to solve the puzzle! You’ll learn how to:
• Combine crawls, log files, and backlinks to find weaknesses in your internal linking structure
• Analyze the impact of tweaking internal linking before you deploy the changes
• Understand how to tweak internal linking at scale
Report
Share
Report
Share
1 of 54
Download to read offline
More Related Content
TechSEO Boost 2018: Internal Link Optimization on Steroids
2. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Kevin Indig
Internal link building on steroids
… or how to flatten power curves
15. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Use tools to get recommendations
16. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Problem: Most internal link models are inaccurate!
17. Kevin Indig | @Kevin_Indig | #TechSEOBoost
PageRank exists between and within sites
18. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Internal PageRank is only half of the equation
Page A Page B
Page C
Page D
Site A
19. Kevin Indig | @Kevin_Indig | #TechSEOBoost
External PageRank is the other side of the equation
Page A Page B
Page C
Page D
Site A Site B
Page A Page B
Page C
Page D
20. Kevin Indig | @Kevin_Indig | #TechSEOBoost
What we need is a model that
combines internal and external PR
21. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Solution: the “True Internal PR” model (TIPR)
CheiRank Backlinks
Log files
TIPR
PageRank
22. Kevin Indig | @Kevin_Indig | #TechSEOBoost
What can you do with TIPR?
Calculate
“accurate” internal
PageRank
Identify technical
problems
Monitor
optimization
progress
23. Kevin Indig | @Kevin_Indig | #TechSEOBoost
The TIPR process
Analysis Recommendations Monitoring
24. Kevin Indig | @Kevin_Indig | #TechSEOBoost
TIPR – step by step
1. Crawl site
2. Calculate internal PR and CR
3. Add backlinks to get “true internal PR”
4. Add crawl rate from log files to understand impact of (internal + external)
links over time
5. Flatten power curve of pages high PR and low CR and vice versa
25. Kevin Indig | @Kevin_Indig | #TechSEOBoost
“Robinhood” principle: take from
the rich, give to the poor
26. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Let’s talk some results
27. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Dry testing the model at small scale
40. Kevin Indig | @Kevin_Indig | #TechSEOBoost
How do we flatten this curve?
41. Kevin Indig | @Kevin_Indig | #TechSEOBoost
So, what do we do with this information?
URL Crawl frequency Domain pop PageRank CheiRank
/URL1 200 300 0.0810 0.3555
/URL2 150 200 0.0300 0.3422
/URL3 300 100 0.0690 0.3000
/URL4 50 50 0.0220 0.2908
42. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Rank, take average, re-sort
URL Crawl frequency Domain pop PageRank CheiRank Average
/URL1 2 1 1 1 1.25
/URL2 3 2 3 2 2.5
/URL3 1 3 2 3 2.25
/URL4 4 4 4 4 4
43. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Rank, take average, re-sort
URL Crawl frequency Domain pop PageRank CheiRank Average
/URL1 2 1 1 1 1.25
/URL3 1 3 2 3 2.25
/URL2 3 2 3 2 2.5
/URL4 4 4 4 4 4
44. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Look for pattern in URLs and optimize accordingly
3.99000 0.03000 0.00972 0.02000 0.01000
0.03000 0.03000
0.07000
0.01000 0.00000
0.00000
0.05000
0.10000
0.15000
0.20000
0.25000
0.30000
0.35000
0.40000
0.45000
Categories Apps Add-ons Vendors Plugins
Average PageRank and CheiRank by directory
PageRank CheiRank
50. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Recap: TIPR
Crawl PR + CR Backlinks Log files
Power
curves
51. Kevin Indig | @Kevin_Indig | #TechSEOBoost
More lessons
Robots.txt XML sitemaps 404 errors
52. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Limitations of the model
• Way more ranking factors than PageRank
• Only suitable for a certain size of sites
• Just tested on a few sites (yet)
• Still trying to find the right weighting
53. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Taking the concept one step further
• Automating the model
• Predicting success with staging environments
54. Kevin Indig | @Kevin_Indig | #TechSEOBoost
Thanks for your attention
Thanks to Catalyst, Audisto, and Nozzle.
@Kevin_Indig
www.kevin-indig.com