Google Ads Before a Website Redesign: 10 Things to Do

1. Why a Redesign Hurts Google Ads Before It Hurts SEO

Quick Answer: Rankings decay slowly after a redesign, so you get weeks to react. Google Ads fails the same afternoon. A dead final URL stops serving, a missing tag stops feeding Smart Bidding, and the algorithm starts optimising on nothing. Paid damage is immediate and billed by the hour.

Most Malaysian SMEs plan a rebuild around design, content and SEO. The usual website redesign checklist handles redirects and rankings well. Google Ads gets one line: update the links. That line hides the most expensive part of the launch.

The difference is speed. Google processes a permanent redirect over time, and 301 redirects do not cost you PageRank. Your ad account has no such patience. It checks the destination in real time, and Google's destination requirements mean a final URL landing on an error can stop an ad group dead.

We already have a guide on restoring tracking after a redesign. This is the opposite job: the work you do in the fortnight before launch, while it is all still fixable for free. For the wider picture, start at our Google Ads management service.

Key takeaway: SEO damage from a rebuild is recoverable over months; Google Ads damage is charged to your card the same day.

The walkthrough below covers the tracking side, and it is worth watching before you brief your developer.

2. The 10-Step Google Ads Website Redesign Checklist

Quick Answer: Ten jobs protect an account through a rebuild: export it, map final URLs, rebuild tags on staging, re-check assets, review Performance Max, freeze bid changes, agree a pause rule, brief the developer, roster launch day, and book the 48-hour audit.

Steps 1 to 6 happen before launch week. Steps 7 to 10 happen on the day and just after. Anyone with a tidy campaign structure can run this list.

1. Export the whole account. Every campaign, ad, asset and conversion action — your rollback copy.
2. Inventory every final URL. Ads, keywords, sitelinks, callouts, promotions and asset groups.
3. Get the URL map from the developer. Old URL to new, one row per page.
4. Rebuild conversion tracking on staging. Fire-test it while there is time to fix it.
5. Re-check every asset and extension. Sitelinks and promotions sit outside the ad editor and get missed.
6. Review Performance Max settings. Decide on final URL expansion and URL exclusions.
7. Freeze bidding and budgets. No target or strategy changes for seven days either side.
8. Agree a written pause rule. What gets paused, for how long, and who decides.
9. Roster launch day. One person on the account, one on the site, same chat.
10. Book the 48-hour audit. Review search terms, conversions and disapprovals.

Key takeaway: Six of the ten steps are preparation, not repair — the account that survives a relaunch did the export-and-map work early.

3. How Do You Map Final URLs So No Ad Points at a 404?

Quick Answer: Put every live final URL in one sheet, sit it beside the developer's old-to-new page map, and mark each row keep, change or retire. Apply the changes the moment the new URLs are confirmed, and never leave an ad relying on a redirect.

A redirect is a safety net, not a plan. Google Ads will usually follow one, but it slows the landing experience and a chain is worse. Google's guidance on site moves with URL changes asks you to point straight at the destination. Your ads should too.

Where stale URLs hide:

• Keyword-level final URLs. Older accounts often set these individually, and they override the ad.
• Sitelinks, promotions and callouts. Assets sit at account, campaign and ad group level, so one broken link can appear three times.
• Tracking templates and suffixes. A template pointing at the old domain builds a broken click URL even when the final URL is right. Google splits the two in its notes on final URLs and tracking templates.
• Shopping feed links. The product link column needs the same treatment as your Shopping feed titles and attributes.

Run the list against staging before the cut-over, then against the live site an hour after. Our website launch checklist catches site-side errors that look like ad problems, and 301 redirects done properly covers the mapping.

Key takeaway: Update final URLs at every level — keyword, ad, asset, template and feed — and point them at the real destination, not a redirect.

4. Which Google Ads Elements Break Most Often at Launch?

Quick Answer: Conversion tracking breaks most often, in roughly two-thirds of accounts handed to us after a relaunch. Final URLs and asset links follow close behind. The table ranks the six failures we find most in Malaysian SME accounts audited after a redesign.

Share of post-redesign Google Ads audits where each element was broken, Malaysia.

Element	Share of audits (%)	Share	Symptom
Conversion tags missing or duplicated		68	Conversions flatline or double count
Final URLs 404 or redirected		61	Ads disapproved or stop serving
Sitelink and callout links stale		54	Clicks land on error pages
Performance Max asset group URLs stale		47	Asset group serving drops
Remarketing tag and audience lists		39	Lists shrink below serving size
Call tracking number replaced		26	Calls stop being attributed

Source: ZenWeb client sample, post-redesign audits, 2024–2026. Licence.

The items that break most are the ones nobody owns. The developer assumes the marketer holds the ad account; the marketer assumes the developer moved the tags. Both appear in our list of Google Ads mistakes that waste budget.

Key takeaway: Name an owner for tags, final URLs and assets before launch week. Most relaunch failures are ownership gaps, not technical ones.

5. Why Conversion Tracking Must Be Rebuilt, Not Re-Pointed

Quick Answer: A new site means new page structure, new form handlers and often a new theme. Re-pointing the old tag assumes all three stayed the same. Install it fresh on staging, fire every conversion once, and confirm each action reports before you go live.

Smart Bidding is only as good as the signal underneath it. Google's explanation of conversion measurement makes the dependency plain: nothing recorded means nothing for the model to learn from. Two silent days is enough to distort a target CPA campaign.

Test these on staging, one at a time:

• Every primary conversion action. Form submit, thank-you page, WhatsApp click, call click, purchase. Our GA4 and WhatsApp conversion setup guide covers the Malaysian lead flows.
• Duplicate firing. New themes ship with their own analytics field, so the tag lands twice. Two conversions per lead is worse than none, because the model believes it.
• Container versus hard-coded tags. Tags pasted into an old theme file are gone. Move to Google Tag Manager so the next rebuild is easier.
• Enhanced conversions. Field mapping breaks when form field names change, which is what a redesign does. See enhanced conversions not working.

Keep the old conversion actions rather than creating replacements. New actions reset the history your bidding strategy depends on.

Key takeaway: Reinstall and fire-test tracking on staging, reuse the existing conversion actions, and check for double firing.

6. What Does an Unprepared Launch Cost in Week One?

Quick Answer: Roughly 60% of the first week's budget is spent blind when tracking and URLs break. On RM 12,000 a month that is about RM 1,655 of misdirected spend and two dozen leads you never see, before any recovery work is billed.

Modelled first-week wasted spend and lost leads by monthly Google Ads budget, Malaysia.

Monthly budget	Week-one waste	Waste (RM)	Leads lost
RM 3,000		415	6
RM 6,000		830	12
RM 12,000		1,655	24
RM 25,000		3,450	49

Illustrative projection: 60% of week-one spend mis-served, RM 70 blended CPL, ZenWeb benchmarks 2024–2026. Licence.

That is not the whole bill. Add the hours spent diagnosing it and the fortnight of soft performance while bidding recovers. A day of preparation costs less than the smallest row above.

Key takeaway: The cost of an unprepared relaunch scales with your budget, so the bigger the spend, the earlier the prep starts.

7. What Happens to Performance Max When the Site Changes?

Quick Answer: Performance Max reads your whole domain, not only the URLs you entered. When page structure changes, asset group URLs go stale and final URL expansion starts sending traffic to pages you did not intend. Both need a decision before launch.

Performance Max is the campaign type most exposed to a rebuild, because it decides for itself where to send people. Google's notes on final URL expansion explain that it can substitute a more relevant page from your domain. On a fresh site, "more relevant" often means a thin page that is not ready to sell.

Three decisions to make before launch:

• Turn final URL expansion off for the switch. Keep it off for a week or two, then reintroduce it once the new site is stable.
• Refresh URL exclusions. Old rules point at old paths, so moving the blog or careers section leaves them matching nothing.
• Re-check asset group creative. A rebrand means new logos and images, and the asset group creative specs have to be met again for the group to serve.

If Performance Max already carries most of your volume, this is the highest-risk item on the list. Our review of whether Performance Max is worth it for SMEs explains that dependency.

Key takeaway: Switch final URL expansion off across the relaunch window, refresh URL exclusions, and re-verify asset group creative before the DNS change.

8. Should You Pause Google Ads on Launch Day?

Quick Answer: Pause narrowly, not completely. Stop the campaigns whose landing pages are changing for the two to four hours around the DNS switch, keep brand campaigns running, and restart the same day. A multi-day full pause costs more than the launch.

Full pauses feel safe and are usually the more expensive choice. Smart Bidding loses recent signal, competitors take the impression share, and remarketing lists start to age.

• Keep brand campaigns live. They point at the home page, the least likely URL to move, and they stop competitors bidding on your name.
• Pause only affected non-brand campaigns. Two hours before the switch until final URLs are verified live.
• Never change bid strategy while paused. That stacks two disruptions together, which our comparison of target CPA versus target ROAS warns against.
• Restart the same day. If the site cannot be verified within a day, the launch was not ready.

Key takeaway: A short targeted pause protects you. A week-long blanket pause hands your impression share to competitors and restarts the learning you paid for.

9. Prepared vs Unprepared: What Changes in 30 Days?

Quick Answer: Prepared accounts are back to normal cost per conversion within about eight days. Unprepared accounts take roughly a month and are still 29% below their old conversion volume at day 30. The gap is preparation, not budget.

Google Ads outcomes 30 days after relaunch, prepared versus unprepared Malaysian SME accounts.

Measure (first 30 days)	Prepared	Unprepared
Days with no conversion data recorded	0–1	9
Bid strategies pushed back into learning	18%	74%
Cost per conversion vs pre-launch (day 30)	+4%	+38%
Conversion volume vs pre-launch (day 30)	98%	71%
Days to return to pre-launch cost per acquisition	8	31

Source: ZenWeb client tracking, Malaysian SME relaunches, 2024–2026. Licence.

The middle row drives the rest. Once three-quarters of your bid strategies re-enter learning at once, Smart Bidding rebuilds from scratch and every other number follows.

Key takeaway: Protecting conversion history is the highest-leverage pre-launch job, because it keeps bid strategies out of learning.

10. Your First 48 Hours After the New Site Goes Live

Quick Answer: Check disapprovals, click through every final URL yourself, confirm at least one real conversion has recorded, compare search terms against the previous week, and watch impression share. Two short sessions: one at four hours, one at 48.

At four hours after the switch:

• Ad and asset status. New disapprovals under destination policy are the fastest signal a URL is broken.
• Manual click test. Click your own ads on mobile and desktop, because staging and live mismatches surface here first.
• Redirect chains. Check no click bounces through two or three hops, since redirect loops burn budget silently.

At 48 hours:

• Conversions recorded. One of every primary action should have fired. Zero is a tag problem, not a demand problem.
• Search terms. Compare against the same weekday last week. A sudden shift means broad match or Performance Max found the new pages.
• Landing page experience. Slower pages drag Quality Score down within days; the fixes overlap with our landing page checklist.

Key takeaway: Two short check-ins beat one long post-mortem. Four hours catches broken destinations; 48 hours catches broken measurement.

11. How Long Does Google Ads Take to Recover?

Quick Answer: A prepared account dips about 6% in week one and clears baseline by week four. An unprepared account drops to 41% of baseline and is still short at week eight. The table indexes weekly conversions with the pre-launch week at 100.

Organic follows a similar curve. If rankings slipped too, read keeping rankings through a rebuild alongside the steps for traffic crashes after a site migration.

Key takeaway: Preparation turns an eight-week recovery into a four-week one, and the prepared curve usually finishes above where it started.

12. Put the Ad Account on the Launch Plan

Quick Answer: A Google Ads website redesign only goes badly when the ad account is treated as somebody else's problem. Put final URLs, conversion tags and Performance Max settings on the project plan with named owners, and the relaunch becomes a routine week.

Nobody sets out to break Google Ads. The account simply is not on the launch plan, so nobody checks it until the numbers look wrong. Add it as a work stream, run the ten steps, and you keep the conversion history you already paid for. If you would rather have it handled alongside the build, that is what our Google Ads management team does for Malaysian SMEs, and the ZenWeb home page shows how design and paid search fit together.

13. Frequently Asked Questions

1. Do I need to pause Google Ads during a website redesign?

Not entirely. Pause only the non-brand campaigns whose landing pages are changing, from about two hours before the DNS switch until you have clicked through every final URL on the live site. Keep brand campaigns running so competitors do not take your name, and restart the same day.

2. Will a website redesign reset my Performance Max learning?

It can. Learning resets when conversion actions are recreated or conversion data stops for several days. Reuse the existing conversion actions instead of making new ones, keep tracking live, and switch final URL expansion off for the relaunch window. Do that and Performance Max usually keeps its learning.

3. How far ahead should I start preparing the ad account?

Two weeks before launch for most SME accounts. You need a day to export the account and inventory final URLs, a few days waiting on the developer's URL map, and a full test cycle on staging for conversion tracking. Larger e-commerce accounts with feeds need three to four weeks.

4. Can I just update final URLs after the new site is live?

You can, but you pay for the gap. Every hour between the switch and the update is spend landing on error pages or redirects. Applying the changes takes minutes when the mapping sheet is ready in advance, so prepare it before launch and apply it within the hour.