What Happens to Traffic in the 90 Days After a Website Rebuild
October 22, 2026 · Steph Walker · Clarity Web Design Studio
Across 16 client rebuilds, the typical site roughly doubled its organic search visits and roughly tripled its direct visits in the 90 days after launch. These are observational results from our own client sites, not a controlled study.
Key findings
- Of 15 website rebuilds with organic search data, all 15 had more organic visits in the 90 days after launch than in the 90 days before. The typical site roughly doubled (median 2.3x). Together, those sites went from 1,789 to 4,476 organic sessions, or 2.5x.
- Direct traffic grew more. All 16 rebuilds had more direct visits after launch. The typical site roughly tripled (median 2.9x), and the combined total went from 2,243 to 5,845 sessions, or 2.6x.
- Every site in the main results went up on both measures. The smallest organic gain was 1.22x, and the smallest direct gain was 1.03x.
- 4 sites started with fewer than 20 organic sessions in the 90 days before launch. Their percentage gains look huge and mean very little on their own. With those 4 removed, the typical site still grew 2.2x and the combined total 2.4x.
These are Clarity Web Design Studio client sites, measured 90 days before and 90 days after launch. This is not a controlled study. It shows what happened, not that the rebuild alone caused it. The section on limits below covers what the numbers can and cannot support.
What we measured and how
We compared each site's traffic in the 90 days before its rebuild launched with the 90 days after. Launch dates range from October 2021 to February 2025, and one site has no recorded date. We looked at 2 things: organic search sessions (visits that arrive from a search engine) and direct sessions (visits with no referring source, such as typing the address or using a bookmark).
Our tracker lists 33 rebuilds, all on WordPress. 16 of them have a clean 90-day before and after comparison and make up the main results. The other 17 are either reported separately or have no usable data, and the section on separate reporting says why for each group. A further 4 sites were refreshed or optimized on other platforms and are not counted as rebuilds.
We report 3 numbers for each channel, and we do not average per-site percentages.
- Median
- The middle site when all sites are lined up from smallest change to largest. A few huge results on tiny sites cannot move it.
- Combined total
- Every site's after sessions added together, divided by every site's before sessions. Larger sites count for more, which matches how much traffic they actually carry.
- Sites up and down
- How many sites rose and how many fell.
We show each change as a multiplier. A site that went from 100 to 250 sessions is 2.5x.
Sessions come from each site's own analytics. For 1 site, the before figures came from a different tool than the after figures, so those 2 counts are not strictly comparable. We kept it in and flagged it in the table. Where an export from a site's own analytics disagreed with our tracker, we used the export.
Organic search roughly doubled
Organic search visits roughly doubled for the typical site in the 90 days after launch (median 2.3x across 15 sites). The combined total rose from 1,789 to 4,476 sessions, or 2.5x. All 15 sites went up.
The range runs from 1.2x to 50.5x. The top of that range comes from sites that started with fewer than 20 organic sessions, which is why we lead with the median and the combined total.
4 of the 15 sites had fewer than 20 organic sessions before launch. With those removed, 11 sites remain. The median is 2.2x, the combined total goes from 1,771 to 4,243 (2.4x), and all 11 went up.
Direct traffic roughly tripled
Direct visits roughly tripled for the typical site (median 2.9x across 16 sites). The combined total rose from 2,243 to 5,845 sessions, or 2.6x. All 16 sites went up. Because direct traffic includes visits that analytics cannot trace to a source, treat it as a supporting signal rather than proof that more people knew the site.
Direct traffic is a mixed bucket. It includes people typing an address or using a bookmark, as well as visits that analytics could not trace to a source, such as links in emails or documents. Growth here can mean more people know the site, and it can also mean more of the traffic is being counted. We report it because it moved, and we do not claim to know which of those it was. Appendix A lists all 16 sites with their before and after sessions.
Growth came without new blog posts
10 to 15 years ago, the common advice was to launch a website, then publish blog posts to bring in search traffic, because search engines favour fresh content. Plenty of people followed it without a strategic or intentional plan, a habit sometimes called "publish and pray."
So we checked whether each site published new blog posts after launch, because new content is a common alternative explanation for traffic growth. 15 of the 16 sites in the main results had no blog or published no new posts in the 90 days after launch.
Site A had occasional blog posts, and our notes do not say whether any fell inside the 90 days. Leaving it out, the other 14 sites with organic data had a median organic gain of 2.4x, and their combined total went from 1,609 to 4,257 sessions (2.6x). That is close to the full group's 2.3x and 2.5x. Site O also sends a newsletter.
So the typical gain here did not come from blogging. The 2 largest organic gains, Site O (50.5x) and Site N (31.0x), came from a site with a paused blog and a site with no blog. Both started with fewer than 20 organic sessions, so those multipliers carry little weight on their own.
This does not show that blog posts would not have added more. We have no comparison group, and we do not have enough sites that posted regularly to compare. The 3 sites that did post regularly after launch are all reported separately: the large, complex site, the site with an analytics gap, and the site that was expanded after launch. Each has its own reason its traffic cannot be read as a clean rebuild result.
Sites reported separately
The rule for the main results is simple. A site is in if it was rebuilt on WordPress and has 90 days of traffic data before launch and 90 days after, from comparable sources. Every other site is reported here, whatever it showed. We did not set them aside because of the results. 1 site is here for a different reason: it was changed heavily after launch, and its row says so. Counting the figure shown for each site, organic search rose for 5 of the 7 sites that do not fit the main comparison, stayed flat at 1.0x for the expanded site, and fell for the large, complex site. Direct traffic rose for 5 and stayed flat at 1.0x for the expanded site and the site with the analytics gap. Appendix B lists each one with its window and why it is reported separately.
4 more rebuilds have no usable traffic data in our tracker, so they appear nowhere in these results.
4 more sites were refreshed or optimized on other platforms, not rebuilt on WordPress. We call these Site Refresh. They are not counted as rebuilds, and Appendix B lists them with their results.
What these numbers do not tell you
This is a record of what happened on our own client sites. It is not a controlled study, and it does not prove the rebuild caused the change.
- No comparison group. We did not track similar sites that were not rebuilt over the same period, such as sites whose owners put a rebuild off.
- Several changes at once. A rebuild usually changes the copy, the page structure, and the site speed together. Only 4 of the 33 rebuilds included a visual rebrand, meaning new graphic design work such as a new logo. 2 of those 4 are in the main results. We did not separate which change drove which result.
- Small sites. The typical site had about 45 organic sessions in the 90 days before launch. A gain of 50 sessions on a base that small is a big multiplier and a small number of people.
- Our own clients. These are businesses that chose to work with us, most by referral, so they may not be typical of all small business websites.
- Seasons and tools. Launch dates differ, so seasonal swings are inside the numbers, and we did not adjust for them. Analytics tools also count visits differently, which is why we flagged the 2 sites that switched tools.
We think the rebuild played a real part in these results. That is our opinion, and the data supports it without proving it.
What this means if you own a website
More traffic is realistic, a different business overnight is not
A realistic expectation after a rebuild is more traffic, not a different business overnight. In our group, the typical site roughly doubled its search visits and roughly tripled its direct visits within 90 days. About half the sites did less than that, and the smallest organic gain was 1.2x.
Ask for the median and the number of sites
If a web designer quotes you an average growth figure, ask for the median and the number of sites behind it. Averages of percentages are easy to inflate with one or two tiny sites, and ours used to be too. Our own tracker produced a much bigger average than the typical site delivered, which is why this page reports the median and the combined totals instead.
Record your before numbers first
Get your before numbers recorded before the work starts. 12 of the 33 rebuilds in our tracker have no usable before figures. In one case, nearly every visit before launch landed on a 404 page, and in another the old site had no analytics installed, so there was nothing meaningful to compare against. Your web host's stats are not a substitute. They count every request to the server, bots included, while Google Analytics counts visits from real browsers, so setting one against the other is comparing apples to oranges. Without a before, no one can say what changed.
About this data
Clarity Web Design Studio rebuilt the 33 WordPress websites in this tracker. A further 4 sites were refreshed on other platforms and are reported separately. Steph Walker, founder and CPACC-certified accessibility professional, compiled the figures from each site's own analytics. Client names are removed. We update this page every 6 months and add any new rebuilds that meet the same rules, whatever their results.
Last updated October 23, 2026.
To cite this page: Clarity Web Design Studio, "What Happens to Traffic in the 90 Days After a Website Rebuild," October 2026.
Appendix A: Results by site
The 16 sites in the main results, anonymized. Sessions are visits in the 90 days before and the 90 days after launch. Sites with an i icon have a note: point to it, tap it or tab to it to read it.
| Site | Direct before | Direct after | Direct change | Organic before | Organic after | Organic change |
|---|---|---|---|---|---|---|
| Site A | 303 | 370 | 1.2x | 180 | 219 | 1.2x |
| Site B | 71 | 89 | 1.3x | 134 | 172 | 1.3x |
| Site C | 41 | 194 | 4.7x | 58 | 84 | 1.4x |
| Site D | 179 | 201 | 1.1x | 71 | 108 | 1.5x |
| Site E | 47 | 60 | 1.3x | 6 | 10 | 1.7x |
| Site F | 99 | 144 | 1.5x | 76 | 145 | 1.9x |
| Site G | 194 | 200 | 1.0x | 33 | 71 | 2.2x |
| Site H | 30 | 105 | 3.5x | 30 | 70 | 2.3x |
| Site I | 487 | 1,859 | 3.8x | 1,022 | 2,530 | 2.5x |
| Site J | 50 | 250 | 5.0x | 45 | 130 | 2.9x |
| Site K | 16 | 43 | 2.7x | 7 | 29 | 4.1x |
| Site L | 106 | 136 | 1.3x | 77 | 409 | 5.3x |
| Site M | 42 | 274 | 6.5x | 45 | 305 | 6.8x |
| Site N | 55 | 232 | 4.2x | 3 | 93 | 31.0x |
| Site O | 171 | 596 | 3.5x | 2 | 101 | 50.5x |
| Site P | 352 | 1,092 | 3.1x | – | – | – |
Other ways to summarize the results
The median is the headline number because a few very large gains on small sites cannot move it. The table shows other common summaries for comparison. The geometric mean is the standard average for ratios (Fleming and Wallace, 1986). The middle half is the range holding the central 50% of sites, from the 25th to the 75th percentile, calculated with linear interpolation, the same method as the PERCENTILE function in Excel and Google Sheets. The simple mean is shown only to illustrate how far the largest gains pull an ordinary average.
| Measure | Organic search (15 sites) | Direct traffic (16 sites) |
|---|---|---|
| Median (the headline) | 2.3x | 2.9x |
| Geometric mean | 3.4x | 2.4x |
| Middle half of sites | 1.6x to 4.7x | 1.3x to 3.9x |
| Simple mean | 7.8x | 2.9x |
These are calculated from each site's before and after session counts. The simple mean is pulled up by Site O and Site N, which both started with fewer than 20 organic sessions. We did not run significance tests or confidence intervals, because there is no comparison group.
Appendix B: Sites reported separately
7 sites that do not fit the main comparison
Large, complex site
- Window 90 days (and a cleaner February to March comparison)
- Direct change 2.4x (0.9x)
- Organic change 0.5x (1.0x)
Organic search fell by about half, the biggest drop in the set. The 2 comparison windows disagree on direct traffic, so no single number is reliable. Within 60 days after launch, the site changed a lot. Pages were added, some content was hidden for the first few months and later republished, and plugins and code were changed. We cannot separate the effect of those changes from the rebuild. Tracking also ran in two places for a short time, so the figures carry some uncertainty.
Launch-month comparison A
- Window First 30 days after launch vs the next 30
- Direct change 1.5x
- Organic change 1.4x
No before period. We lost analytics access after 60 days.
Site with an analytics gap after launch
- Window 90 days before launch vs days 91 to 180 after
- Direct change 1.0x (0.2x in the first 90 days)
- Organic change 1.4x (0.1x in the first 90 days)
Analytics were not capturing all traffic for the first 90 days after launch, so those figures are incomplete and could not be verified. The later window starts 3 months after launch, so it is not a like-for-like comparison.
Post-launch comparison A
- Window 60 days after launch vs the next 60
- Direct change 8.5x
- Organic change 1.3x
No before data. The old site had no analytics installed. This compares 2 periods after launch, so it cannot show what the rebuild changed. Job postings were added to the site during these months, likely driving direct traffic.
Post-launch comparison B
- Window 60 days after launch vs the next 60
- Direct change 1.7x
- Organic change 1.1x
No before data was recorded. This compares 2 periods after launch, so it cannot show what the rebuild changed.
Post-launch comparison C
- Window 60 days after launch vs the next 60
- Direct change 26.5x
- Organic change 17.0x
The old site was on Wix and averaged about 1 visit a month, so there is no meaningful before. This compares 2 periods after launch. Directional only.
Site expanded after launch
- Window 90 days before and 90 days after
- Direct change 1.0x
- Organic change 1.0x
The before and after windows are clean, but the site after launch had grown well beyond what we built, with dozens of pages added and other changes to the setup. Traffic in the 90 days after launch cannot be tied to the rebuild. With this site counted in the main results, the combined organic total is 1.9x, and the combined direct total is 2.1x. The medians barely move (2.2x organic, 2.7x direct).
Rebuilds with a shorter window or no clean before period
Short window A
- Window 30 days
- Direct change 3.2x
- Organic change 2.0x
30-day window instead of 90. A seasonal site that launched at the start of its season, so the gain cannot be separated from the seasonal rise.
Short window B
- Window 30 days, compared with the same dates a year earlier
- Direct change 1.5x
- Organic change 2.0x
30-day window instead of 90.
Broken before data
- Window Days 30 to 60 after launch
- Direct change No figure
- Organic change 1.5x
1,229 of 1,260 visits before launch were to a 404 error page, so the before figure is unusable.
Launch-month comparison B
- Window First 30 days after launch vs the next 30
- Direct change 6.1x
- Organic change 5.2x
No before period. This compares 2 months after launch, so it cannot show what the rebuild changed.
No before period
- Window None
- Direct change No figure
- Organic change No figure
No usable before data.
After only, no before tracking
- Window 90 days after launch only
- Direct change No figure
- Organic change No figure
The old site had no tracking, so there is no before. In the 90 days after launch the site had 954 organic and 290 direct sessions. This rebuild included a visual rebrand.
Site Refresh: 4 sites not rebuilt on WordPress
These 4 sites were refreshed or optimized on a platform other than WordPress, so they are not counted as full rebuilds. Their results are shown here for completeness.
Wix-to-Square redesign
- Window 90 days before and 90 days after
- Direct change 4.6x
- Organic change 1.7x
Redesigned on Square, not rebuilt on WordPress, so it is not counted as a full rebuild.
Wix-to-Wix redesign
- Window 30 days in summer 2025 vs 30 days around launch, both from Wix
- Direct change Up, from 4 to 35 sessions
- Organic change Down, from 15 to 6 sessions
This was a redesign within Wix. Our rebuilds are on WordPress, so this site is not counted as a full rebuild. The before window ended about 3.5 months before launch, and the counts are very small. Most of the new traffic came from Facebook, not search. We show session counts instead of multipliers because a multiplier on counts this small means very little.
Online store A
- Window 30 days
- Direct change 0.7x
- Organic change 1.5x
Shopify store, not rebuilt on WordPress, so not counted as a full rebuild. Traffic is not the main success measure for a store.
Online store B
- Window 30 days, no earlier data
- Direct change 0.7x
- Organic change 1.1x
Shopify store, not rebuilt on WordPress, so not counted as a full rebuild. Short window and nothing earlier to check it against. Sales from Google search rose 14% in the same period.