Same Ad Budget, Different Window: How Date Range Changes ROAS

The comparison problem: equal budgets do not mean equal timing

A local business spends the same amount on Google Ads and Meta Ads. The dashboard shows Google producing the stronger return, so the owner moves budget away from Meta.

That decision may be right. It may also be an artifact of the selected date range.

ROAS is calculated as:

Attributed revenue ÷ ad spend

The spend appears immediately. Revenue does not. A prospect might click today, request an estimate next week, approve the work three weeks later, and pay after completion. If the report is viewed before that process finishes, it contains the full cost but only part of the eventual revenue.

The distortion can differ by channel:

• Google Search often captures people already looking for a provider.
• Meta often introduces or reminds people of a service before they are ready to inquire.
• Neither pattern is universal. Brand strength, service urgency, offer, targeting and sales follow-up can change the timeline.

This is why a seven-day cross-channel comparison can confuse speed of conversion with quality of investment.

Illustrative model: one budget viewed three times

The following is a labeled hypothetical model, not a Nika Spark client result or an industry benchmark.

Assume a high-consideration local service runs one acquisition cohort on each platform:

• Google Ads spend: $10,000
• Meta Ads spend: $10,000
• Google’s cohort eventually generates $40,000 in CRM-confirmed revenue
• Meta’s cohort eventually generates $50,000 in CRM-confirmed revenue
• Revenue is measured using the same business definition on both channels

Now assume Google customers tend to close sooner, while Meta customers need more time:

| Revenue observed by cutoff | Google Ads | Meta Ads | |---|---:|---:| | 7 days | $28,000 | $15,000 | | 30 days | $36,000 | $35,000 | | 90 days | $40,000 | $50,000 |

The resulting apparent ROAS is:

| Reporting cutoff | Google ROAS | Meta ROAS | Apparent winner | |---|---:|---:|---| | 7 days | 2.8x | 1.5x | Google | | 30 days | 3.6x | 3.5x | Nearly even | | 90 days | 4.0x | 5.0x | Meta |

Nothing about the original spend changed. The final revenue did not change either. Only the amount of time allowed for revenue to become observable changed.

At seven days, reallocating budget toward Google looks obvious. At 90 days, the same decision would move money away from the higher-return cohort.

Why Google and Meta can distort at different rates

A platform comparison becomes unreliable when it mixes channel behavior, conversion lag and platform attribution into one number.

First, the channels may enter the buying journey at different points. A Google search for an urgent service can occur near purchase. A Meta impression may create demand well before a prospect contacts the business.

Second, platform dashboards do not necessarily assign revenue the way the CRM does. View-through conversions, click-based attribution, modeled conversions and cross-device behavior can create different platform totals. Our related article, Platform ROAS vs. CRM Revenue: Why They Differ, explains why dashboard revenue should not automatically be treated as booked revenue.

Third, operational delay matters. Slow lead handling pushes revenue recognition farther from the original ad interaction. Lead Response Time: The Monthly Cost of Delay provides a framework for diagnosing that separate sales-process problem.

Finally, audience saturation can raise acquisition costs even when the reporting window is sound. For that issue, see Ad Frequency vs. CPL: A Local Meta Ads Model.

Date range is therefore one diagnostic layer—not a universal explanation for weak performance.

Which window gives the least-distorted budget signal?

For the illustrative business above, the 90-day view gives the least-distorted ROAS comparison because it captures the modeled sales cycle more completely. That does not mean every local business should default to 90 days.

Use this decision process:

1. Measure actual conversion lag. In the CRM, calculate the time from first attributable interaction or lead creation to booked revenue. 2. Find the maturity point. Identify when most revenue from a lead cohort has historically appeared. Avoid choosing a universal number without checking your data. 3. Use mature cohorts for ROAS. If jobs commonly close several weeks after inquiry, exclude recent cohorts from final efficiency judgments. 4. Apply one revenue definition. Compare channels using booked, collected or otherwise consistently qualified CRM revenue—not one platform’s reported value against another platform’s value. 5. Check volume. A mature window can still be noisy if it contains only a few sales. 6. Account for business changes. A very long window may blend old pricing, offers, targeting or seasonality with the current strategy.

The correct window is usually the shortest period that captures a mature share of revenue while remaining representative of current conditions. In a long-lag service, that may be closer to 90 days. In an urgent, short-cycle category, 30 days—or less—may be adequate.

Use two views instead of forcing one window to do everything

A single dashboard should not be asked to measure both current momentum and final profitability.

Build two views:

1. Mature efficiency view

Use sufficiently aged cohorts and CRM-confirmed revenue to compare ROAS by channel. This is the primary view for major budget reallocations.

2. Recent operating view

Use the latest seven or 30 days to monitor spend, lead volume, qualified-lead rate, frequency, search terms, landing-page behavior and sales follow-up. Treat these as early indicators, not completed revenue outcomes.

Before moving budget, ask:

• Are both channels being measured through the same cutoff date?
• Have their lead cohorts had enough time to close?
• Is revenue coming from the CRM or only from platform attribution?
• Did one large sale dominate the period?
• Has seasonality or a campaign change made the older data less relevant?

A longer window is not automatically more accurate, but an immature window is predictably incomplete. If your seven-, 30- and 90-day views tell different stories, investigate conversion lag before declaring a winner.

Need help building a channel comparison around mature revenue rather than dashboard timing? Book a call with Nika Spark to map the reporting window to your actual sales cycle.