In Google Ads, it is easy to find settings that remain unchanged for years simply because the system already assigned them a particular value. The conversion window is one of them. The account is set to 30 days, conversions are coming in, campaigns are running, so there seems to be no reason to revisit the setting.
The problem becomes visible when the actual customer journey does not fit within the selected period. One business receives most inquiries on the same day as the ad click, another a week later, while in B2B, months may pass between the first click and the action the company ultimately needs. If the same logic is applied to all of these scenarios, reporting begins to show a picture that is not quite suitable for making decisions.
That is why a conversion window is not just a formal setting and not the answer to the question “How long does a customer think before buying?” It defines the period during which a target action that happens after an ad interaction can still be attributed to that interaction.
This is exactly why universal recommendations should be avoided. The right conversion window is not determined by what is typically used in ecommerce, healthcare, or B2B, but by what your users actually do after interacting with an ad.
Imagine a simple situation. A person visits the website from an ad on September 1 but does not buy anything. They return a few days later and place an order on September 10.
If the conversion window for that action is 30 days, the purchase happens within the configured period and can be counted as a conversion after the ad interaction. If the window is 7 days, day ten is already outside it.
The customer’s behavior itself has not changed. They still saw the ad, visited the website, and purchased later. Only the rule used by the advertising system to count that result has changed.
That is why a short conversion window does not prove that the ad had no influence on a later purchase. Likewise, a long window does not prove that the ad caused the conversion. It is primarily a measurement rule, not a tool for establishing causality.
For this reason, the period should not be chosen according to the logic “the longer, the more we will see” or “the shorter, the cleaner the data.” The goal is to understand which time horizon actually fits the specific target action.
One of the most common mistakes is trying to define the conversion window only by the length of the sales cycle.
For example, the customer journey may look like this: a person clicks an ad, submits an inquiry five days later, attends a consultation two weeks later, and purchases the service on day thirty.
Which period should be used in Google Ads?
The answer depends on which action is actually being passed to the system as a conversion.
If the campaign is optimized for form submissions, the main concern for that conversion action is the time between the ad interaction and the inquiry. If a confirmed sale from the CRM is imported into Google Ads, then a much longer path becomes relevant.
The sales cycle describes the full process from first contact to sale. The conversion window is configured for a specific event being measured.
That is why the same business does not necessarily need the same period for every conversion action. A call, inquiry, booking, qualified lead, and actual sale may all happen at very different times after the first ad interaction.
Confusion often appears because several similar terms are used alongside each other in analytics.
A conversion window defines the time limit: how long after an ad interaction a subsequent target action may still be counted.
An attribution model answers a different question — how conversion value should be distributed across ad interactions when there were several of them along the customer journey.
A lookback window is also related to time, but it is used in attribution reporting: it defines how far back the system looks at previous ad interactions when analyzing the path to conversion.
Another term is conversion lag, or conversion delay. Unlike the conversion window, this is not a setting. Conversion lag shows the actual delay between an ad interaction and the target action.
This distinction matters in practice. If users in your business often convert after four or five days, that is their actual behavior. The conversion window determines whether the advertising system will count those delayed actions.
Conversion lag explains another situation that regularly appears when analyzing Google Ads.
Suppose historical data shows that a significant share of purchases occurs three to five days after a click. Today, we open yesterday’s statistics and see spend, clicks, and only some of the conversions.
The spend has already happened. The clicks have already been recorded. But some of the people who will eventually buy simply have not done so yet.
As a result, the most recent days may temporarily show a higher CPA or lower ROAS. A few days later, once delayed conversions are added, the picture changes.
That is why for businesses with noticeable conversion delay, evaluating performance only from the most recent period is risky. It can create the false impression that the campaign has suddenly deteriorated when, in reality, we are simply looking at incomplete data.
For many web conversions in Google Ads, 30 days is a common starting value. But default does not mean optimal.
In businesses with a short decision cycle, most target actions may happen within a few hours or days. For them, a long window does not necessarily add much useful information.
In other industries, the opposite is true. A person may spend several weeks comparing offers, reading reviews, returning to the website, and only then submitting an inquiry. If a meaningful share of conversions happens after day thirty, a shorter period will cut off part of the real journey.
But even the industry itself does not provide a ready-made answer. Two ecommerce businesses may have completely different customer behavior: one sells inexpensive everyday goods, while the other sells equipment worth several thousand dollars. Both are formally ecommerce, but their decision cycles will be different.
That is why rules such as “use 7 days for ecommerce and 90 days for B2B” are convenient only because they are simple. For real configuration, they are far too broad.
Suppose historical data shows that a significant share of customers convert within 14–20 days, but the action is configured with only a 7-day window.
Purchases after day seven may still happen, but they will no longer fall within the selected conversion window.
As a result, reports will show fewer recorded conversions. CPA may appear higher, ROAS lower, and campaigns or keywords that attract users with a longer path to decision may look weaker.
This affects more than what the marketer sees. Smart Bidding uses the conversion data available to the system. If a meaningful share of desired actions regularly falls outside the configured period, the automated strategy also receives a different picture of performance.
This does not mean that all windows should immediately be increased. First, it is worth checking whether later conversions are actually a recurring part of the customer journey rather than isolated cases.
The opposite extreme is setting the longest available period to make sure nothing is missed.
At first glance, the logic seems reasonable: a longer window creates more opportunities to connect later conversions with advertising. But more recorded conversions do not always mean more accurate measurement.
The more time passes between the initial ad interaction and the purchase, the more other factors can appear along the customer journey. The user may find the company again through organic search, see an ad in another channel, receive an email, read a recommendation, or speak with a sales manager.
That is why the goal of a conversion window is not to maximize the number of conversions in the ad account. It should cover the period during which the connection with the ad interaction still makes practical sense for your measurement.
A reliable approach begins not with choosing between 7, 30, or 90 days, but with analyzing the data.
First, define exactly which conversion action is being configured. If it is a click-to-call action, the delay may be close to zero. If it is a form submission, it may be several days. If it is a qualified lead or actual sale, it may be much longer.
Then look at how long it actually takes for these conversions to happen. In Google Ads, Days to conversion data and time lag information can be used for this.
There is one important detail here: the average value is not always enough.
Suppose the average time to conversion is eight days. This may mean that nearly all users convert around day eight. But it may also describe a very different distribution: 60% convert within the first two days, 30% around two weeks later, and another 10% after a month or more.
For choosing the window, the second type of information is far more important.
That is why it is more useful to look at what share of conversions accumulates within 1, 7, 14, 30, 60, or 90 days. If the increase becomes minimal after a certain point, that provides a much stronger basis for a decision.
For lead generation, Google Ads data is often not enough because the advertising system only sees the conversion action that was passed to it.
For example, a user submits an inquiry two days after the click, but the deal closes only three weeks later.
If the business optimizes for inquiries, the first interval matters for that action. If the main goal is actual sales, the business needs to look much further.
This is where CRM shows what a standard advertising report does not: how long it takes to move from lead to sale, which inquiries become qualified, and which campaigns bring not just fast form submissions, but real revenue.
This is especially important for high-value services, B2B, and any business where several stages exist between the first inquiry and payment.
That is why the right conversion window for such projects often cannot be determined from Google Ads alone. The full customer journey needs to be considered.
In ecommerce, a user may click an ad today, return two days later, and purchase on day five. With a 7-day conversion window, that purchase still falls within the configured period. With a 3-day window, it does not.
In a medical clinic, the path may be longer. A person searches for a procedure, calls a few days later, books an appointment later still, and only then comes in for the visit. In this case, the team cannot choose the right window until it defines the primary conversion: the call, the booking, or the actual visit.
In B2B, a potential customer may submit a demo request ten days after the first click, receive a commercial proposal on day twenty-five, and sign the deal on day sixty. If the campaign optimizes for the demo request, one period matters. If a closed deal from the CRM is imported, a completely different one matters.
In all three cases, the correct answer depends not on the name of the industry, but on which event is being measured and how long users actually take to reach it.
The type of ad interaction should also be considered separately.
A click-through conversion means the user actively clicked the ad. Engaged-view is used in scenarios where the user did not necessarily click but engaged sufficiently with a video ad. View-through means the ad was shown and the user later converted without directly interacting with it.
These are signals of different strength.
That is why an equally long window for a click and a simple view does not necessarily have the same analytical meaning. For each type, it is important to understand what behavior the system is attempting to measure.
Automated Google Ads strategies do not operate in a vacuum. They use the conversion data available to the system.
That means the window setting affects more than whether the numbers in reports look better or worse. If relevant conversions consistently fall outside a period that is too short, the algorithm also does not see them in the same context.
But here too, the rule “longer is better” does not apply. Smart Bidding does not need the largest possible numbers; it needs data that correctly represents the business action that matters.
That is why high-quality conversion tracking begins not with choosing a bidding strategy, but with defining what counts as a conversion and understanding how that action behaves over time.
The conversion window also helps explain why data from different systems may not always match.
Google Ads, GA4, and CRM may use different attribution logic, different time limits, and different ways of assigning an event to a date.
For example, in Google Ads, a conversion may be associated with the date of the ad interaction, while another system may display it on the actual date of purchase.
Conversion delay, different channel sets, and attribution settings add further differences.
That is why a discrepancy between numbers does not automatically mean tracking is broken. First, make sure you are comparing the same event, the same period, and compatible measurement logic.
There is no reason to change it every month simply according to a schedule.
If the funnel and customer behavior remain stable, the setting may stay unchanged for a long time. It is worth reviewing after significant changes: launching a new conversion action, changing the product or price, entering a new market, integrating CRM or offline conversions, changing the funnel, or seeing a noticeable seasonal shift in the decision cycle.
New Days to conversion data may also be a reason to revisit the setting if it shows that a significant share of target actions now happens earlier or later.
In other words, the conversion window is not a “set it once and forget it” setting, but it is also not something that should constantly be adjusted just to improve the numbers.
The most common mistake is leaving the default 30 days and never checking whether it matches the actual data.
The second is choosing a period based on assumptions: “our sales are fast, so seven days is enough” or “this is B2B, so we should immediately use 90.”
It is equally problematic to use the same window for all conversion actions, confuse it with the attribution model or conversion lag, evaluate the most recent days without accounting for delay, or look only at the average time to conversion.
Another mistake is increasing the period simply because reports may then show more conversions. The purpose of the setting is not to improve how the statistics look, but to make the statistics better reflect the real user journey.
Finally, it is worth remembering that changing the window does not automatically rewrite all historical data according to the new rules. That is why experimenting with this setting without a clear reason is not advisable.
First, define which conversion action needs to be measured. Then review the actual Days to conversion and conversion lag — not only the average, but the full distribution.
Next, check how long it takes for most target actions to accumulate and compare this with the actual sales cycle. If there is a significant gap between inquiry and sale, include CRM data.
Evaluate click-through, engaged-view, and view-through conversions separately because they represent different types of interaction with advertising.
After that, you can set a window that covers the real path to the required action without unnecessary extension and review the result again after a sufficient period.
The logic should therefore look like this:
conversion action → actual time lag → conversion distribution → sales cycle → CRM → attribution settings → specific number of days.
Not the other way around.
The right conversion window does not begin with the number 7, 30, or 90. It begins with understanding what the customer actually does after interacting with an ad and at which point that journey becomes a conversion for your business.
A conversion window is the period after an ad interaction during which a subsequent target action by the user can still be attributed to that interaction.
For many web conversions, 30 days is a common starting value, but the available range depends on the source and type of conversion action.
There is no universally better option. If a significant share of users convert after day seven, a 7-day window will be too short. The decision should be based on actual Days to conversion data.
For some conversion sources, this period is available. However, the maximum value does not automatically mean the most accurate measurement.
A conversion window is a system setting. Conversion lag is the actual time between an ad interaction and completion of the target action.
A conversion window defines the time limit for counting a conversion, while an attribution model determines how its value is distributed across ad interactions.
Yes, that can be completely reasonable. An inquiry, booking, qualified lead, and actual sale often have different delays.
Because of conversion lag. Advertising spend has already been recorded, while some users who will eventually convert have not yet completed the target action.
Yes. Automated strategies use the conversion data available in Google Ads, so a window that is too short may leave some relevant delayed conversions outside the data used for optimization.
No. It should be reviewed when the funnel, conversion type, decision-making cycle, market, or data transfer method changes — not simply according to a calendar.
