Your Reporting Window Is Shorter Than Your Customer's Decision

Most marketing teams are evaluating campaigns before the customer has finished deciding. That's not a measurement problem. It's a structural bias built into how reporting works, and it systematically punishes the campaigns that do the most long-term work.

The uncomfortable truth: your conversion lag is longer than your reporting window, and the gap between those two things is quietly corrupting your marketing decisions.

The Window Doesn't Match the Journey

Google published internal research in 2026 that should have changed how every performance marketing team runs its reporting. It didn't get enough attention.

Finding: the standard 30-day click and three-day engaged-view window captures approximately 70% of conversions for standard Google Ads campaigns. For Performance Max, that same window captures only 50%. For Demand Gen campaigns, it captures just 40%.

Think about what that means for Demand Gen specifically. Six in ten conversions from those campaigns fall outside the conventional reporting window. If you're evaluating Demand Gen performance at 30 days, you are, at best, working from an incomplete picture. At worst, you're cutting campaigns that haven't finished converting yet.

Google's data came from 5,000 standard Google Ads advertisers and 4,000 Performance Max or Demand Gen advertisers, with data collected between July and December 2025. It's proprietary and platform-produced, so treat it as directional rather than audited truth. But the direction is clear, and it aligns with what anyone running upper-funnel campaigns has quietly suspected for years.

Add to that a basic operational reality: recent conversion data can be understated by 24–48 hours or more due to reporting and processing delays. Even the last few days of your reporting window are probably undercounting.

The Window You Set Becomes the Reality You See

Here's the part that rarely gets discussed: a seven-day conversion window doesn't just measure the customer decision cycle. It defines it. Any conversion occurring after day seven is simply excluded. The reporting setup becomes self-fulfilling.

Set your window to seven days, and your data will show you a seven-day customer. It will tell you that decisions happen fast. It will reward campaigns that close fast. And it will steadily defund everything that works over longer timeframes, including brand campaigns, educational content, and anything targeting buyers who aren't yet in market.

At its core, marketing short-termism isn't about marketers being ignorant of long-term effects. It's that the measurement infrastructure actively discourages accounting for them.

Google itself recommends evaluating campaigns only after the account's average days-to-conversion have elapsed, and provides a "Days to conversion" report so advertisers can see the actual distribution. Almost no one uses it to set evaluation timelines. Almost everyone still runs weekly or monthly reviews and labels campaigns efficient or inefficient based on a window that may exclude a third to a majority of their eventual conversions.

B2B Makes This Worse by an Order of Magnitude

B2B marketers face a conversion lag problem that isn't just a platform quirk. It's a defining feature of how buying actually works.

A 2025 Dreamdata study, summarized by The Wise Marketer, found that the average B2B buying journey runs 211 days. The average number of touches before purchase: 76. The average number of stakeholders involved: 6.8. The average number of channels: 3.7. And the LinkedIn journey from first ad impression to closed revenue: 320 days.

Large companies show journeys that are 49% longer still. And more than three months can pass between a marketing-qualified lead and any sales involvement.

Every B2B marketer knows this intuitively: the lead is not the sale. It's not even close to the sale. A lead appearing in week two of a campaign means nothing about whether that campaign worked. The decision won't happen for another six months, involving six additional stakeholders across four channels, with 62 sessions along the way.

And yet most B2B reporting systems evaluate campaigns at the lead stage. Pipeline and revenue are tracked elsewhere, by a different team, with different tools. The attribution window closes before the deal opens.

Natural consequences follow: lead volume gets rewarded, and the campaigns that build the kind of awareness and credibility that make a 211-day journey end in a "yes" get cut for underperforming.

Chapter 21 Describes Exactly This Problem

Patrick Gilbert covers the structural version of this problem in Never Always, Never Never. Chapter 21 argues that measurement tools, including attribution, marketing mix modeling, and incrementality testing, are film room tools, not scoreboards. They're designed for learning and analysis, not for rendering verdicts.

When you use a 7-day or 30-day attribution window to decide whether a campaign succeeded or failed, you're treating the film room like a final score. The problem is that the film doesn't end when the reporting window closes. The customer is still deciding.

The book's measurement framework draws a sharp distinction between evaluating campaigns (the film room, where you're learning what worked and why) and evaluating teams or business outcomes (the scoreboard, where you're measuring revenue, market share, and long-term growth). Conversion lag sits squarely in that film room category. It's information about timing and path. Treating it as a verdict produces the kind of short-termism that undermines brand-building investment and rewards whatever is easiest to close fast.

Also relevant from the book: the Wilt Chamberlain Effect. Marketers often know their reporting windows are too short. They know that Demand Gen campaigns need longer evaluation cycles. They know that B2B pipelines don't close in 30 days. And they evaluate campaigns at 30 days anyway, because changing the system is uncomfortable and the short-term numbers feel more concrete. The Wilt Chamberlain effect is abandoning what works because it feels awkward, not because it stopped working.

Demand Capture vs. Demand Creation Is the Real Fault Line

Not every campaign has a long conversion lag. That nuance is worth preserving.

Standard Search campaigns targeting existing intent tend to produce shorter, more visible conversion paths. Google's data shows roughly 70% of conversions from those campaigns appearing within the standard 30-day window. If someone searches "buy project management software" and converts within two weeks, the attribution is probably telling you something real.

Short windows aren't wrong for demand capture. They're not. Applying demand-capture measurement logic to demand generation campaigns is where the problem starts. Performance Max shows only 50% of conversions within the standard window. Demand Gen shows 40%. These are campaigns designed to introduce, educate, and influence buyers who weren't yet looking. The conversion path is structurally longer. Measuring them on a demand-capture timeline produces a systematic false negative.

This is a measurement version of a brand vs. performance budget problem. Campaigns that create future demand get measured on current-demand timelines, underperform against those timelines by design, and get defunded. Budget then concentrates in channels where conversions appear fastest, often because those campaigns are capturing demand that upper-funnel work already created. The halo effect from brand and awareness activity goes uncredited, and the performance channels that benefit from it look more efficient than they actually are.

At AdVenture Media, this dynamic comes up consistently when connecting upper-funnel spend to downstream revenue. The lag between influence and conversion is real, and the standard reporting setup doesn't account for it.

A Practical Framework for Measuring What's Actually Happening

Fixing this doesn't mean ignoring short-term indicators. Early signals, clicks, qualified sessions, pipeline, and leads, are genuinely useful for learning and optimization. Separate evaluation timing from measurement frequency.

You can report weekly. You should not evaluate weekly if your average days-to-conversion is 60. Those are different activities.

A more defensible approach:

• Pull the actual distribution. Use Google's "Days to conversion" report to see where your conversions actually land. Set your evaluation window based on that data, not platform defaults.
• Separate campaign types. Apply a shorter evaluation window to Search campaigns targeting existing intent. Apply a longer one to Demand Gen, Performance Max, and anything upper-funnel.
• Connect to CRM revenue. In B2B especially, platform-reported conversions at the lead stage are early indicators, not outcomes. Track opportunity creation and closed revenue by cohort, not by campaign window.
• Run incrementality tests before cutting. If a campaign appears to underperform at 30 days, a holdout test will tell you whether pulling it changes downstream outcomes. Platform attribution often can't answer that question on its own.
• Allow conversion lag to fully expire before reassessment. A campaign that ended last week hasn't finished converting. Treat recent data as incomplete until the lag window closes.

None of this is complicated in theory. The complication is organizational. Weekly reporting cycles create weekly pressure to explain weekly numbers. If the campaign is in its lag window, those numbers don't yet mean what the reporting implies they mean. That's a conversation worth having with leadership, explicitly, before the evaluation happens.

The Campaign May Be Finished. The Customer Isn't.

Short reporting windows aren't always wrong. They're often incomplete. Incomplete data interpreted as final verdicts produces systematic errors in a consistent direction: away from long-term influence, toward short-term capture, and against the campaigns that build the mental availability that makes all future marketing more efficient.

Consider the 95-5 rule. At any given moment, roughly 95% of your addressable market is not in an active buying cycle. Demand creation campaigns reach that 95%. Their conversions won't show up in your 30-day window. They'll show up six months from now, attributed to something else, when the buyer finally searches and clicks. That's not a failure of the demand creation campaign. That's how memory and decision-making actually work.

Measurement systems that can't see that effect will defund the work that causes it. That's not a data problem. It's a framing problem, and fixing it starts with acknowledging that your reporting window was never designed to match your customer's decision.

To explore how to structure a measurement system that handles both short and long timeframes, the scoreboards vs. film room framework is the right place to start.