A 90-day plan to make your website faster

Key takeaways

A website speed improvement plan works best in three 30-day stages: measure and take the easy wins, fix templates and scripts, then protect the gains and prove the result. Ninety days is deliberate. Real-visitor data takes 28 days to reflect a change, so each stage ends with evidence instead of hope. Most small business sites can move from failing to passing on mobile within that period.

• Days 1 to 30: baseline, images, caching and unused scripts.
• Days 31 to 60: first-screen templates, forms and third-party tools.
• Days 61 to 90: budget, monitoring and a before-and-after report.
• Allow 28 days after each change for field data to catch up.

Why 90 days?

Ninety days fits how speed is measured. The field data Google uses covers a rolling 28-day window, so a fix made today is only fully visible a month later. Three stages of about a month each let you make changes, see their effect in real-visitor data and decide the next step on evidence.

It is also long enough to deal with the cause of most slowdowns, which is not one fault but habits: unsized images, tools added without review and nobody watching.

• How website speed affects customer experience and conversions
• Lab data vs field data: which speed numbers to believe

Days 1 to 30: what do you measure and fix first?

Start by recording where you are, then make the fixes that need little or no development.

• 1. Score the site on the GBL Speed Experience Scorecard.
• 2. Record field and lab results for three key pages, with enquiries by device.
• 3. Resize and convert the images on those pages.
• 4. Switch on page and browser caching.
• 5. Delete unused plugins and pause tags for inactive platforms.
• 6. Replace mobile sliders and video headers with one image.

• The website speed scorecard: score your site in an hour
• Website speed checklist: twenty checks before you spend
• Optimise website images for speed and search
• Website caching explained for business owners

Days 31 to 60: what do you fix in the templates?

The second month is developer work on the templates behind your key pages and the steps where customers act.

• 1. Make the first-screen text, image and button load before anything else.
• 2. Delay chat, reviews, pop-ups and non-essential scripts.
• 3. Reserve space for images, banners and embeds.
• 4. Cut fonts to the weights in use and show text immediately.
• 5. Make the enquiry form confirm at once.
• 6. Move booking or other heavy widgets behind a tap.

• Third-party scripts: the hidden reason your site is slow
• Why your website jumps around while it loads
• Slow enquiry forms: the wait after a visitor clicks Send
• Slow booking widgets: where appointment bookings are lost

Days 61 to 90: how do you protect and prove it?

The final month makes the gains stick and shows what they were worth.

• 1. Write a one-page performance budget and name its owner.
• 2. Set up uptime alerts and a monthly check of three pages.
• 3. Re-score the site and compare with day one.
• 4. Compare mobile engagement and enquiry rate with the baseline.
• 5. Report the result on one page, with any remaining work.

• Performance budgets: stop your website slowing down again
• Website speed monitoring: catch slowdowns before customers
• A website speed report directors will actually read

What will it cost and who does the work?

The first month is largely the business's own time, with some help on caching. The second is where a developer is needed. Indicative UK market ranges for focused speed work on an existing site are £500 to £3,000 in total, depending on the number of templates and the tools involved.

Name one person on your side to own the plan. Without that, the stages drift and the gains are not defended.

• Website speed optimisation costs in the UK
• Who should fix a slow website: developer, host or agency?

When will this plan not be enough?

The plan will not be enough when the platform is the cause. If the site still fails after the scripts are trimmed and the first screen is light, the theme or page builder is probably shipping more code than can be tuned away. At that point the evidence from the first 60 days makes the case for a redesign of templates or a rebuild, and tells the next supplier what to avoid.

It is also the wrong plan if speed is not your problem. If pages pass on day one, put the 90 days into conversion work instead.

• Slow website: fix it or rebuild it?
• A 90-day plan to turn a good website into a converting one

What does this look like in practice?

A pattern we see when speed work is done as a single burst: a developer spends a week on it, the score improves, and nobody checks whether customers noticed or whether it lasted. Six months on, the site is slow again and the business doubts speed work is worth paying for.

Staging the work, measuring at each step and ending with a budget and an owner turns a one-off fix into a lasting one. Across client work at Global Bridge Labs (GBL), we have seen a 2.4x enquiry rate from conversion changes like these.

• See client work
• Website Design & Development for UK businesses

90-day speed plan checklist

• Name the owner and the three key pages.
• Record the baseline before changing anything.
• Month one: images, caching, unused plugins and tags.
• Month two: templates, scripts, fonts and forms.
• Month three: budget, monitoring and the report.
• Decide on evidence whether more work is justified.

• Core Web Vitals for business websites
• What a slow website really costs a UK business

Next step

Three months from now your site can be measurably faster, with someone responsible for keeping it that way. If you would like help with any stage, let's discuss your requirements.