Google Ads Data Retention Limits: What You Can Still See, and What Disappears

Google Ads keeps most of your account history, but not every report lets you reach it, and a few tools only cover a limited window. If you need old data for an audit, a client handover, or a year-over-year comparison, the question is rarely "does Google still have it?" It's "which report will still show me this, and at what level of detail?" Google changes these windows from time to time, so confirm specifics in Google Ads Help (support.google.com/google-ads) before you rely on them.

This article sorts out the three ideas that get blurred together under "google ads data retention limits," points to the data types that behave differently, and gives you an archiving routine so you stop depending on what the interface happens to show later.

What Google Ads Data Retention Limits Actually Mean

People use "retention limit" to describe three separate things, and mixing them up leads to bad assumptions.

• Data retention: how long Google stores a given type of data at all.
• Date-range limits: how far back a specific report, view, or segment lets you pull data, even if the underlying numbers still exist somewhere.
• Lookback or attribution windows: how long after a click or interaction a conversion can still be credited to it. This is a measurement setting, not a storage rule.

The biggest misconception is that a "limit" means your campaign history has been deleted. For core performance stats such as impressions, clicks, cost, and conversions at the campaign level, history generally remains available well beyond what most audits need. What differs is the report type and the level of detail. A granular view, like a search-term-level breakdown or a heavily segmented report, can be restricted in ways a simple campaign summary is not.

That's why "I can't find it" doesn't always mean "it's gone." Sometimes you're in the wrong report, the date range is set wrong, or a filter hides removed campaigns. Check those three before you assume a loss.

One more caution: these windows are not fixed. Google updates reports, retires tools, and adjusts what's available. Everything below should be read as "as of October 2026," and each specific figure should be checked against the current Google Ads Help page for that report. This article deliberately avoids quoting exact day or month counts from memory, because a wrong number in an audit plan is worse than no number.

Which Reports and Data Types Have Time Limits

Not every part of the account follows the same rules. These are the ones worth checking individually.

• Standard performance stats: campaign, ad group, and keyword metrics. These are the most durable and usually the safest to rely on for long-range comparisons.
• Search Terms Report: shows the queries that triggered your ads, but only a subset, and it behaves differently from keyword data. More on this in the next section.
• Segmented reports: adding segments such as device, time, or network can change what's available, and some combinations are limited to specific periods.
• Change history: the log of who changed what and when. Its window is typically shorter than the window for core metrics, so it's the one most likely to have aged out by the time you need it.
• Auction Insights: competitive data that depends on report availability and on enough data existing for a given period.

The pattern is that summary numbers last, while diagnostic and explanatory data (what changed, who you competed with, which queries triggered ads) tends to be more time-sensitive. Those are exactly the things you want when asking "why did performance drop in spring?"

Here's a quick reference you can fill in once you've verified each item on Google's current Help pages:

1. Core performance stats. Check: the available date range in the reports interface. Why it matters: it determines how far back year-over-year comparisons can go.
2. Search Terms Report. Check: the date range you can select, plus how Google describes query visibility. Why it matters: it's your evidence for negative keyword and new keyword decisions.
3. Segmented reports. Check: any segment-specific date restrictions. Why it matters: a breakdown you need later may not exist for older periods.
4. Change history. Check: how far back the log goes. Why it matters: it's often the only record of what changed before a performance shift.
5. Auction Insights. Check: available date range and data thresholds. Why it matters: it's how you explain competitive pressure over time.

Fill in the actual values only after you've confirmed them, and put the verification date next to each one. Google's rules can shift, and a dated note tells the next person how fresh your information is.

Search Terms Data: Why It Is the Most Fragile

A keyword is what you bid on. A search term is what someone actually typed. Your keyword might be "running shoes" (phrase match), while the search terms it matched could be "best running shoes for flat feet," "running shoes repair," or "free running shoe catalog." The first is something you control. The second is something you discover.

Google also doesn't show you every search term. Queries are reported only when they meet its visibility criteria, so a portion of your traffic is never itemized, even inside the date range you selected. Check Google Ads Help for its current explanation of search term visibility, because it has changed over the years and may change again.

The practical problem is timing. Search terms drive two different decisions: which irrelevant queries to block as negative keywords, and which good queries to add as positive keywords with a suitable match type. Both decisions rely on context, such as what the query looked like, what it cost, and when it started appearing. Leave the review for months and you get a backlog where you can't tell a one-off oddity from a trend.

Here's an illustration, not a real case. Imagine an agency takes over an account and notices a pile of irrelevant queries eating budget. They pull a recent date range, see the junk, and add negatives. But they can't work out when it started, because the earlier period was filtered out or no longer reported at the detail they need. Was it a broad match expansion? A new campaign setting? A seasonal shift? Without the earlier data, they're fixing the symptom with no idea of the cause, and the client asks questions they can't answer.

The fix is boring and effective: review search terms on a regular schedule, so each decision is made while the context is fresh. If you do this inside the Search Terms Report, Keywordme can speed it up. It's a Chrome extension that lets you mark junk terms as negatives, add good ones as keywords, and apply match types without leaving Google Ads. You still make each call yourself; the tool just removes the spreadsheet round trip that makes people put the job off.

Conversion Windows and Attribution Lookback Limits

This is the third meaning of "limit," and it has nothing to do with storage. When you set up a conversion action in Google Ads, you choose how long after an ad interaction a conversion can still be credited to it. The click-through conversion window is the main one, and there are others for different interaction types. Google Ads Help lists the allowed ranges and the defaults for each, and those are what you should cite rather than a figure remembered from an old tutorial.

Why it matters for old data: the window shapes the numbers you see. A short window credits fewer late conversions. A longer one credits more. If your sales cycle runs weeks, a short window will undercount, and the account will look worse than it is.

Before you change a window, read what Google says about how the change applies. Treat it as affecting how conversions are counted going forward, and check the documentation for exactly what happens to historical reporting. Don't assume past numbers will quietly rewrite themselves, or that they won't.

Two mistakes come up repeatedly:

• Comparing periods without checking settings. If the window or attribution model changed between two date ranges, the difference in conversions may be partly or wholly a measurement artifact. Look at the conversion action's settings history, and your own change log, before drawing conclusions.
• Judging recent campaigns too early. Conversions can arrive days or weeks after the click. Last week's numbers are usually incomplete, so a campaign launched recently can look weak simply because its conversions haven't landed yet.

The habit that helps: when you compare two periods, note the conversion window and attribution model that applied to each, and leave recent days out of any performance verdict until they've had time to mature.

How Limits Hurt Audits, Reporting, and Year-over-Year Comparisons

Retention problems rarely show up on a normal day. They show up when someone asks a backward-looking question.

Take a few common cases. A quarterly review wants to explain why cost per conversion rose, but the change history that would show a bidding switch has aged out. A seasonal comparison needs last November's search-term-level detail to see which queries drove Black Friday, and only campaign totals remain. A client handover lands on your desk with a request to "keep doing what worked," and nobody can say what that was because the granular evidence is no longer in the interface.

Account changes make this worse even when the data technically exists:

• Restructures: splitting or merging campaigns means last year's totals no longer map cleanly onto this year's campaigns.
• Removed campaigns: they can drop out of default views, so totals look smaller than they were unless you adjust the status filter.
• Merged or moved accounts: history can end up split across accounts, with no single place to see it whole.

In each case the numbers aren't lost; they're hard to read. Without notes on what changed and when, you're guessing at comparability.

Finally, don't assume connected tools mirror Google Ads. Google Analytics and BigQuery data transfers have their own retention settings and rules, set and documented separately. Verify each one on its own Help page. A longer retention period in one tool doesn't extend what Google Ads itself will show you, and the reverse is true as well.

How to Preserve Your Data Before It Ages Out

You don't need a data team. You need a routine that runs whether or not anyone remembers it.

Set a monthly export cadence

Once a month, export the reports you'd hate to lose:

• Search terms, at the most detailed level available.
• Campaign (and, if useful, ad group) performance.
• Conversion performance by conversion action.

Name files consistently, for example clientname_searchterms_2026-09, and always export full calendar months so periods line up later. Store them in a shared folder rather than someone's desktop.

Choose your method

Google Ads scheduled reports can email or deliver recurring exports. Google Sheets and Looker Studio connections can pull data on a schedule, and a data warehouse suits larger operations. Check Google's current documentation to confirm each option is still available and how it's configured, since features and connectors change.

Keep a change log

A simple dated document does a lot of work. Record budget changes, bidding strategy switches, structure changes, conversion window edits, and major negative keyword additions. Six months on, this note explains your numbers better than the raw export does, and it covers you if change history has expired.

Standardize across accounts

If you run an agency, use the same folder structure, naming, cadence, and change log template for every client. Then a handover or a team member leaving doesn't mean reconstructing history from memory. Put someone's name against the monthly task, and a backup name too.

Check the Docs, Export the Rest, and Review Search Terms Now

Retention limits aren't one rule. They vary by data type, by report, and by connected tool, and Google adjusts them over time. So the habit that holds up is simple: look up the current limit on Google's Help pages before you promise anyone a historical comparison, and export anything you can't afford to lose on a regular schedule.

Search terms are the easiest place to start, because they're both the most time-sensitive data and the data you can act on today. Open the report, clear out what's wasting spend, and add what's working.

Optimize Google Ads Campaigns 10X Faster. Without Leaving Your Account. Keywordme lets you remove junk search terms, build high-intent keyword lists, and apply match types instantly, right inside Google Ads. No spreadsheets, no switching tabs, just quick, seamless optimization. Start your free 7-day trial (then just $12/month) and take your Google Ads game to the next level.