SlideShare a Scribd company logo
How Time To First Byte (TTFB) Impacts Your Site’s Performance
Let’s all agree that your website speed
can be subject to many factors.
While it’s hard to break down all of
them, it’s still not impossible.
In this context, one of
the metrics you can try
to interpret and
improve is Time To
First Byte (TTFB).
Through this slide deck, we’ll
walk you through essential
information that will help you
assimilate the impact of this
metric on web performance.
So, what is Time To First Byte
(TTFB) anyways?
Time To First Byte (TTFB) is
the name given to the time
between the first request of
the end-user to a web server
and the response back of that
request from the webserver
to the end-user.
Resolving the address of the
site using a DNS and
retrieving the response of the
first request sent to the site
are the factors that cause this
time to occur.
In other words, this period will
occur in 3 steps and the
performance in these stages
will play an active role in the
duration of TTFB.
Step1
Submitting the first request to
the site address
Step1
Submitting the first request to
the site address
DNS response time (how fast is
the end-user’ side for resolving
DNS request)
Step1
Submitting the first request to
the site address
DNS response time (how fast is
the end-user’ side for resolving
DNS request)
The distance of the
website servers to the
end-user, shorter the
better.
Step1
Submitting the first request to
the site address
DNS response time (how fast is
the end-user’ side for resolving
DNS request)
The distance of the
website servers to the
end-user, shorter the
better.
Stability of
the network
Step2
Resolving this request by the
webserver
Step2
Resolving this request by the
webserver
Physical hardware response
time (how fast the web server
can resolve it)
Step2
Resolving this request by the
webserver
Physical hardware response
time (how fast the web server
can resolve it)
The existing workload
of the server
operations
Step2
Resolving this request by the
webserver
Physical hardware response
time (how fast the web server
can resolve it)
The existing workload
of the server
operations
Any
network-
related latency
within the
datacenter.
Step3
Sending the first response to
the end-user
Step3
Sending the first response to
the end-user
The end-users internet
speed
Step3
Sending the first response to
the end-user
The end-users internet
speed
Stability of the
connection.
What is
considered as
good TTFB and
what is not?
In general, many speed test
tools give ranges as following:
In general, many speed test
tools give ranges as following:
• 0 – 75ms Perfect
In general, many speed test
tools give ranges as following:
• 0 – 75ms Perfect
• 75 – 200ms Ideal
In general, many speed test
tools give ranges as following:
• 0 – 75ms Perfect
• 75 – 200ms Ideal
• 200 – 500ms Less than ideal but acceptable
In general, many speed test
tools give ranges as following:
• 0 – 75ms Perfect
• 75 – 200ms Ideal
• 200 – 500ms Less than ideal but acceptable
• > 500ms There is a problem
However, the fact that the site
is entirely composed of
dynamic data can increase
these ranges. Determining a
TTFB based on the content of
the site will always make more
sense.
What can make TTFB slow?
Network problems that exist in Steps
1 and 3 (mentioned earlier) will increase
TTFB.
Similarly, in step 2, during the
processing of the request, older
hardware and problems on Disk or
RAM may cause a slowdown.
Low I/O values on the disk will
prevent processing fast and will
queue too many requests.
Likewise, if the server hardware is not
enough for the operations or there is
an instant flood of requests, they will
end up with an increasing TTFB.
Additionally, poorly optimized code,
database, and web server
configuration may cause a delay in
meeting incoming requests.
Finally, high-level
software that meets and
processes requests
through the server will
also cause slowdowns.
So now that you can
determine the pain points,
let’s see how you can
increase TTFB speed
The initial response time
can be reduced by:
The initial response time
can be reduced by:
• Presenting static data at first boot
• Presenting static data at first boot
• Using a Content Delivery Network, i.e. having your site content closer to end-users.
The initial response time
can be reduced by:
• Presenting static data at first boot
• Using a Content Delivery Network, i.e. having your site content closer to end-users.
• Code Optimization: the settings of the software language used, and the performance of the
encoding can be improved to accelerate the first rendered page.
The initial response time
can be reduced by:
• Presenting static data at first boot
• Using a Content Delivery Network, i.e. having your site content closer to end-users.
• Code Optimization: the settings of the software language used, and the performance of the
encoding can be improved to accelerate the first rendered page.
• Database Query Optimization: database normalization and database operations must be
overhauled.
The initial response time
can be reduced by:
• Presenting static data at first boot
• Using a Content Delivery Network, i.e. having your site content closer to end-users.
• Code Optimization: the settings of the software language used, and the performance of the
encoding can be improved to accelerate the first rendered page.
• Database Query Optimization: database normalization and database operations must be
overhauled.
• Data Caching in Memory
The initial response time
can be reduced by:
• Presenting static data at first boot
• Using a Content Delivery Network, i.e. having your site content closer to end-users.
• Code Optimization: the settings of the software language used, and the performance of the
encoding can be improved to accelerate the first rendered page.
• Database Query Optimization: database normalization and database operations must be
overhauled.
• Data Caching in Memory
• Using the latest hardware as much as possible such as new CPUs, SSD or NVME for I/O.
The initial response time
can be reduced by:
The impact of CDN on TTFB
Basically, this is what you get
when you use a robust CDN
• All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS
server and speed up your web site.
Basically, this is what you get
when you use a robust CDN
• All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS
server and speed up your web site.
• Better Routing on the requests, i.e. every request is routed to the closest server.
Basically, this is what you get
when you use a robust CDN
• All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS
server and speed up your web site.
• Better Routing on the requests, i.e. every request is routed to the closest server.
• With static data transferred onto the CDN, origin servers will only be used for new content.
Your web site will not use origin hardware resources for CDN cached content or requests.
Basically, this is what you get
when you use a robust CDN
• All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS
server and speed up your web site.
• Better Routing on the requests, i.e. every request is routed to the closest server.
• With static data transferred onto the CDN, origin servers will only be used for new content.
Your web site will not use origin hardware resources for CDN cached content or requests.
• More hardware resources are available.
Basically, this is what you get
when you use a robust CDN
The impact of TTFB on web
performance
TTFB is indirectly related to the performance
of subsequent transactions since the first byte
is the return period.
Optimizations to reduce TTFB will ensure that
your site’s post-TTFB operations will be more
efficient. Operations that can be performed to
have a good TTFB pave the way for
subsequent I/O operations to be faster.
On the other side of the
spectrum, the user
experience will get much
better with higher
performance. Because your
site will load faster, and the
probability of losing the user
until the page opens
becomes considerably
reduced.
If you feel the need to reevaluate your site’s
performance, our team of experts has
elaborated a FREE assessment program to
help you measure the performance of your
website, by filling out this form, the detailed
report will be sent to you within 24 hours.
GET YOUR FREE ASSESSMENT

More Related Content

PDF
What we can learn from CDNs about Web Development, Deployment, and Performance
PDF
Measuring CDN performance and why you're doing it wrong
PDF
Measuring CDN performance and why you're doing it wrong
PDF
NANOG32 - DNS Anomalies and Their Impacts on DNS Cache Servers
PDF
ICANN DNS Symposium 2021: Measuring Recursive Resolver Centrality
PPT
Request routing in CDN
PPTX
Network latency - measurement and improvement
PDF
RIPE 82: Measuring Recursive Resolver Centrality
What we can learn from CDNs about Web Development, Deployment, and Performance
Measuring CDN performance and why you're doing it wrong
Measuring CDN performance and why you're doing it wrong
NANOG32 - DNS Anomalies and Their Impacts on DNS Cache Servers
ICANN DNS Symposium 2021: Measuring Recursive Resolver Centrality
Request routing in CDN
Network latency - measurement and improvement
RIPE 82: Measuring Recursive Resolver Centrality

What's hot (20)

PDF
RIPE 82: An Update on Fragmentation Loss Rates in IPv6
PDF
Fighting Spam With A Perimeter Mail System 20071108 Sasag
PDF
RIPE 82: DNS Evolution
PPTX
DNS-SD Extentions
PDF
RIPE 76: Measuring ATR
PDF
RIPE 76: TCP and BBR
PPTX
Interative Traffic Engineering in Changing Internet Economics - Tom Daly at L...
PDF
Performance Evaluation of XMPP on the Web
PDF
Network
PDF
NANOG 82: DNS Evolution
PPTX
Nginx Scalable Stack
PDF
RIPE 78: IPv6 reliability measurements
PDF
Getting a Grip on CDN Performance - Why and How
PDF
Web Performance in the Age of HTTP/2 - FEDay Conference, Guangzhou, China 19/...
PDF
Covert Timing Channels using HTTP Cache Headers
PDF
Covert Timing Channels using HTTP Cache Headers
PDF
HTTP/2: What no one is telling you
PPTX
Covert timing channels using HTTP cache headers
PDF
Revisiting HTTP/2
PPTX
Web Performance Automation - NY Web Performance Meetup
RIPE 82: An Update on Fragmentation Loss Rates in IPv6
Fighting Spam With A Perimeter Mail System 20071108 Sasag
RIPE 82: DNS Evolution
DNS-SD Extentions
RIPE 76: Measuring ATR
RIPE 76: TCP and BBR
Interative Traffic Engineering in Changing Internet Economics - Tom Daly at L...
Performance Evaluation of XMPP on the Web
Network
NANOG 82: DNS Evolution
Nginx Scalable Stack
RIPE 78: IPv6 reliability measurements
Getting a Grip on CDN Performance - Why and How
Web Performance in the Age of HTTP/2 - FEDay Conference, Guangzhou, China 19/...
Covert Timing Channels using HTTP Cache Headers
Covert Timing Channels using HTTP Cache Headers
HTTP/2: What no one is telling you
Covert timing channels using HTTP cache headers
Revisiting HTTP/2
Web Performance Automation - NY Web Performance Meetup
Ad

Similar to How Time To First Byte (TTFB) Impacts Your Site’s Performance (20)

PDF
Web Server SEO: Make your TTFB faster!
PPTX
Web Server SEO: Making your TTFB faster!
PDF
Site Speed Fundamentals
PPTX
2015 03 06 lmtv wtf http webcast
PDF
DINR 2021 Virtual Workshop: Passive vs Active Measurements in the DNS
PPS
Pmw2 k3ni 1-3a
PPT
Performance engineering
PPT
PDF
Presentation on 'The Path to Resolverless DNS' by Geoff Huston
PDF
Tips to improve your website performance
PDF
Web performance mercadolibre - ECI 2013
PDF
Web performance optimization - MercadoLibre
PDF
How to Measure Your CDN’s Cache Hit Ratio and Increase Cache Hits
PDF
The Anatomy of Failure - Lessons from running systems to serve millions of pe...
PDF
Improve Load Times with a Website Speed Test Tool | Skillfloor
PPTX
Performance in business terms
PPT
Performance test
PPSX
Web server
PPTX
Cdn technology overview
Web Server SEO: Make your TTFB faster!
Web Server SEO: Making your TTFB faster!
Site Speed Fundamentals
2015 03 06 lmtv wtf http webcast
DINR 2021 Virtual Workshop: Passive vs Active Measurements in the DNS
Pmw2 k3ni 1-3a
Performance engineering
Presentation on 'The Path to Resolverless DNS' by Geoff Huston
Tips to improve your website performance
Web performance mercadolibre - ECI 2013
Web performance optimization - MercadoLibre
How to Measure Your CDN’s Cache Hit Ratio and Increase Cache Hits
The Anatomy of Failure - Lessons from running systems to serve millions of pe...
Improve Load Times with a Website Speed Test Tool | Skillfloor
Performance in business terms
Performance test
Web server
Cdn technology overview
Ad

More from Medianova (6)

PDF
This Is How We Test Our Performance With JMeter
PDF
What is Image Optimization and How Does it Make Websites Load Faster?
PDF
3 Computing Paradigms as Enablers of Smart Applications of The Future
PDF
The State Of Video In The Enterprise
PPTX
How can live streaming skyrocket your marketing strategy?
PDF
Why Use a Content Delivery Network (CDN)?
This Is How We Test Our Performance With JMeter
What is Image Optimization and How Does it Make Websites Load Faster?
3 Computing Paradigms as Enablers of Smart Applications of The Future
The State Of Video In The Enterprise
How can live streaming skyrocket your marketing strategy?
Why Use a Content Delivery Network (CDN)?

Recently uploaded (20)

PDF
MIND Revenue Release Quarter 2 2025 Press Release
PDF
Agricultural_Statistics_at_a_Glance_2022_0.pdf
PPTX
1. Introduction to Computer Programming.pptx
PDF
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
PPT
Teaching material agriculture food technology
PDF
Blue Purple Modern Animated Computer Science Presentation.pdf.pdf
PDF
Mobile App Security Testing_ A Comprehensive Guide.pdf
PDF
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
PPTX
MYSQL Presentation for SQL database connectivity
PPTX
Group 1 Presentation -Planning and Decision Making .pptx
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
PDF
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
PDF
A comparative analysis of optical character recognition models for extracting...
PDF
Unlocking AI with Model Context Protocol (MCP)
PDF
gpt5_lecture_notes_comprehensive_20250812015547.pdf
PPTX
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx
PDF
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
PPTX
SOPHOS-XG Firewall Administrator PPT.pptx
PPTX
Programs and apps: productivity, graphics, security and other tools
PDF
Assigned Numbers - 2025 - Bluetooth® Document
MIND Revenue Release Quarter 2 2025 Press Release
Agricultural_Statistics_at_a_Glance_2022_0.pdf
1. Introduction to Computer Programming.pptx
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Teaching material agriculture food technology
Blue Purple Modern Animated Computer Science Presentation.pdf.pdf
Mobile App Security Testing_ A Comprehensive Guide.pdf
Video forgery: An extensive analysis of inter-and intra-frame manipulation al...
MYSQL Presentation for SQL database connectivity
Group 1 Presentation -Planning and Decision Making .pptx
Reach Out and Touch Someone: Haptics and Empathic Computing
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
A comparative analysis of optical character recognition models for extracting...
Unlocking AI with Model Context Protocol (MCP)
gpt5_lecture_notes_comprehensive_20250812015547.pdf
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
SOPHOS-XG Firewall Administrator PPT.pptx
Programs and apps: productivity, graphics, security and other tools
Assigned Numbers - 2025 - Bluetooth® Document

How Time To First Byte (TTFB) Impacts Your Site’s Performance

  • 2. Let’s all agree that your website speed can be subject to many factors. While it’s hard to break down all of them, it’s still not impossible.
  • 3. In this context, one of the metrics you can try to interpret and improve is Time To First Byte (TTFB).
  • 4. Through this slide deck, we’ll walk you through essential information that will help you assimilate the impact of this metric on web performance.
  • 5. So, what is Time To First Byte (TTFB) anyways?
  • 6. Time To First Byte (TTFB) is the name given to the time between the first request of the end-user to a web server and the response back of that request from the webserver to the end-user.
  • 7. Resolving the address of the site using a DNS and retrieving the response of the first request sent to the site are the factors that cause this time to occur.
  • 8. In other words, this period will occur in 3 steps and the performance in these stages will play an active role in the duration of TTFB.
  • 9. Step1 Submitting the first request to the site address
  • 10. Step1 Submitting the first request to the site address DNS response time (how fast is the end-user’ side for resolving DNS request)
  • 11. Step1 Submitting the first request to the site address DNS response time (how fast is the end-user’ side for resolving DNS request) The distance of the website servers to the end-user, shorter the better.
  • 12. Step1 Submitting the first request to the site address DNS response time (how fast is the end-user’ side for resolving DNS request) The distance of the website servers to the end-user, shorter the better. Stability of the network
  • 13. Step2 Resolving this request by the webserver
  • 14. Step2 Resolving this request by the webserver Physical hardware response time (how fast the web server can resolve it)
  • 15. Step2 Resolving this request by the webserver Physical hardware response time (how fast the web server can resolve it) The existing workload of the server operations
  • 16. Step2 Resolving this request by the webserver Physical hardware response time (how fast the web server can resolve it) The existing workload of the server operations Any network- related latency within the datacenter.
  • 17. Step3 Sending the first response to the end-user
  • 18. Step3 Sending the first response to the end-user The end-users internet speed
  • 19. Step3 Sending the first response to the end-user The end-users internet speed Stability of the connection.
  • 20. What is considered as good TTFB and what is not?
  • 21. In general, many speed test tools give ranges as following:
  • 22. In general, many speed test tools give ranges as following: • 0 – 75ms Perfect
  • 23. In general, many speed test tools give ranges as following: • 0 – 75ms Perfect • 75 – 200ms Ideal
  • 24. In general, many speed test tools give ranges as following: • 0 – 75ms Perfect • 75 – 200ms Ideal • 200 – 500ms Less than ideal but acceptable
  • 25. In general, many speed test tools give ranges as following: • 0 – 75ms Perfect • 75 – 200ms Ideal • 200 – 500ms Less than ideal but acceptable • > 500ms There is a problem
  • 26. However, the fact that the site is entirely composed of dynamic data can increase these ranges. Determining a TTFB based on the content of the site will always make more sense.
  • 27. What can make TTFB slow?
  • 28. Network problems that exist in Steps 1 and 3 (mentioned earlier) will increase TTFB. Similarly, in step 2, during the processing of the request, older hardware and problems on Disk or RAM may cause a slowdown. Low I/O values on the disk will prevent processing fast and will queue too many requests.
  • 29. Likewise, if the server hardware is not enough for the operations or there is an instant flood of requests, they will end up with an increasing TTFB. Additionally, poorly optimized code, database, and web server configuration may cause a delay in meeting incoming requests.
  • 30. Finally, high-level software that meets and processes requests through the server will also cause slowdowns.
  • 31. So now that you can determine the pain points, let’s see how you can increase TTFB speed
  • 32. The initial response time can be reduced by:
  • 33. The initial response time can be reduced by: • Presenting static data at first boot
  • 34. • Presenting static data at first boot • Using a Content Delivery Network, i.e. having your site content closer to end-users. The initial response time can be reduced by:
  • 35. • Presenting static data at first boot • Using a Content Delivery Network, i.e. having your site content closer to end-users. • Code Optimization: the settings of the software language used, and the performance of the encoding can be improved to accelerate the first rendered page. The initial response time can be reduced by:
  • 36. • Presenting static data at first boot • Using a Content Delivery Network, i.e. having your site content closer to end-users. • Code Optimization: the settings of the software language used, and the performance of the encoding can be improved to accelerate the first rendered page. • Database Query Optimization: database normalization and database operations must be overhauled. The initial response time can be reduced by:
  • 37. • Presenting static data at first boot • Using a Content Delivery Network, i.e. having your site content closer to end-users. • Code Optimization: the settings of the software language used, and the performance of the encoding can be improved to accelerate the first rendered page. • Database Query Optimization: database normalization and database operations must be overhauled. • Data Caching in Memory The initial response time can be reduced by:
  • 38. • Presenting static data at first boot • Using a Content Delivery Network, i.e. having your site content closer to end-users. • Code Optimization: the settings of the software language used, and the performance of the encoding can be improved to accelerate the first rendered page. • Database Query Optimization: database normalization and database operations must be overhauled. • Data Caching in Memory • Using the latest hardware as much as possible such as new CPUs, SSD or NVME for I/O. The initial response time can be reduced by:
  • 39. The impact of CDN on TTFB
  • 40. Basically, this is what you get when you use a robust CDN
  • 41. • All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS server and speed up your web site. Basically, this is what you get when you use a robust CDN
  • 42. • All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS server and speed up your web site. • Better Routing on the requests, i.e. every request is routed to the closest server. Basically, this is what you get when you use a robust CDN
  • 43. • All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS server and speed up your web site. • Better Routing on the requests, i.e. every request is routed to the closest server. • With static data transferred onto the CDN, origin servers will only be used for new content. Your web site will not use origin hardware resources for CDN cached content or requests. Basically, this is what you get when you use a robust CDN
  • 44. • All queries start with a DNS request. So a CDN using Anycast DNS will use the closest DNS server and speed up your web site. • Better Routing on the requests, i.e. every request is routed to the closest server. • With static data transferred onto the CDN, origin servers will only be used for new content. Your web site will not use origin hardware resources for CDN cached content or requests. • More hardware resources are available. Basically, this is what you get when you use a robust CDN
  • 45. The impact of TTFB on web performance
  • 46. TTFB is indirectly related to the performance of subsequent transactions since the first byte is the return period. Optimizations to reduce TTFB will ensure that your site’s post-TTFB operations will be more efficient. Operations that can be performed to have a good TTFB pave the way for subsequent I/O operations to be faster.
  • 47. On the other side of the spectrum, the user experience will get much better with higher performance. Because your site will load faster, and the probability of losing the user until the page opens becomes considerably reduced.
  • 48. If you feel the need to reevaluate your site’s performance, our team of experts has elaborated a FREE assessment program to help you measure the performance of your website, by filling out this form, the detailed report will be sent to you within 24 hours. GET YOUR FREE ASSESSMENT