Attribution guideMatch the cycleUse observed time-to-conversion instead of copying a default. Separate interactionsClick-through and view-through windows need different treatment. Compare consistentlyAlign event definitions and windows across every report used for budget decisions.
Choose attribution windows that match the buying cycle
An attribution window defines how long an ad interaction remains eligible to receive credit for a later conversion. The window changes reported results, so media buyers should choose it from the real buying cycle and keep it consistent when comparing channels.
Core controls
Measure the right thing before changing delivery
These three controls keep the analysis tied to real campaign decisions instead of isolated dashboard percentages.
Time eligibility
The window decides whether an interaction can receive conversion credit after a defined number of hours or days.
Model interaction
The attribution model decides how eligible interactions share credit. The window decides which interactions are eligible at all.
Sensitivity testing
Recalculate performance under shorter and longer windows to see whether a decision depends on the reporting rule.
Field note
Attribution window, conversion window and lookback window
Advertising platforms use related terms in different reporting areas. A conversion window is generally the period after an ad interaction during which a conversion can be recorded for that interaction. A lookback window controls how far back an attribution report searches for eligible interactions before the conversion. In everyday campaign work, both describe a time boundary around credit, but the exact setting and report behavior should be verified in the platform documentation.
The attribution model is a separate choice. A last-click model, data-driven model or multi-touch model determines how credit is distributed among the eligible interactions. The window determines which interactions enter the candidate set. Changing either setting can change campaign-level CPA or ROAS without changing the actual number of orders recorded by the business.
Document the event, interaction type, window and model together. A report labeled conversions is ambiguous unless the team knows whether it means a seven-day click window, a one-day view window, a data-driven model or another configuration.
Evidence gate
Before acting on attribution window, conversion window and lookback window, set the evidence threshold and review window in advance. Record the baseline, name the decision owner and define the maximum change that can be made in one cycle.
Field note
Click-through and view-through windows answer different questions
A click-through window starts from an ad click or another qualifying interaction. It is usually easier to defend because the user took a deliberate action. A view-through window starts from an impression without a click. It can capture delayed influence, but it also has a larger risk of claiming conversions that might have happened without the ad.
Keep view-through reporting separate from click-through reporting. Use a shorter view window, require a viewability definition where relevant, and inspect overlap with other channels. For direct-response optimization, many teams make click-attributed accepted conversions the primary operational signal and use view-through results as supporting evidence.
Engaged-view rules add another interaction type, often for video. They require more engagement than a simple impression but less than a click. Treat them as their own category and avoid blending them into click conversions without disclosure.
Reporting check
Use click-through and view-through windows answer different questions only after the reporting base is large enough to be credible. Keep the original control intact, document exclusions and schedule a follow-up check before expanding the decision.
Decision framework
A repeatable way to move from data to action
Use a controlled sequence that protects measurement quality before the campaign team changes delivery.
Field note
Choose the window from observed behavior
Start with the distribution of time from first qualified ad interaction to accepted conversion. Look at the median, the 75th percentile and the long tail. A low-consideration mobile utility may convert within hours, while a B2B software trial or finance lead can have a longer research cycle. The window should capture the meaningful buying process without extending so far that unrelated later activity receives credit.
Use the shortest window that captures the stable majority of valid outcomes for the decision at hand. A campaign used for daily source optimization may need a shorter operational window than a quarterly marketing-effectiveness report. Keep both definitions explicit so short-term buying controls are not confused with long-term business analysis.
When the underlying distribution changes, revisit the setting. A new GEO, device mix, checkout process or offer type can change conversion delay. Do not assume one global window fits every campaign.
Action boundary
Translate choose the window from observed behavior into one bounded action rather than several simultaneous changes. Assign an owner, preserve the comparison group and write down the condition that would reverse the action.
Field note
Align windows before comparing platforms
Cross-platform comparisons fail when one platform reports a long click window and another uses a short click window plus view-through credit. Align the conversion event, time zone, counting method and attribution window before ranking channels. Google Ads also recommends comparable conversion events and windows when using platform-comparable reporting.
Create a warehouse or analytics view with a common decision window. Keep the native platform reports for optimization, but add a normalized report for budget allocation. If exact alignment is not possible, show the differences instead of pretending the numbers are directly comparable.
The same rule applies to FroggyAds source reporting. Keep the click timestamp, conversion timestamp and accepted conversion status so the team can recalculate results under a chosen window.
Cross-metric check
Evaluate align windows before comparing platforms beside conversion quality, cost and delivery context. A single favorable percentage is not enough; require consistent evidence across the chosen segment and time window.
Field note
Run a sensitivity analysis before a large budget move
Calculate CPA, conversion rate and revenue under at least three windows: short, working and long. If the same sources remain strong across all three, the decision is robust. If rankings change dramatically, the campaign may depend on delayed or weakly attributable conversions.
Break the analysis down by format, device, GEO, new versus returning customer and conversion type. A one-day window may be adequate for an app install but too short for a high-consideration purchase. View-through sensitivity should be reported separately.
Use the result to set guardrails. A source that looks profitable only under the longest possible window should not receive the same scaling confidence as a source that performs under a conservative window.
Rollback check
Treat run a sensitivity analysis before a large budget move as a decision input, not a standalone verdict. Keep a test cell, watch for measurement gaps and stop the change when the downstream quality signal moves in the wrong direction.
Comparison model
Read the signal in context
Use a consistent table so every buyer, analyst and campaign owner interprets the same metric in the same way.
| Campaign pattern | Starting analysis window | Why | Validation step |
|---|---|---|---|
| Immediate lead or app action | Hours to a few days | The user can complete the action quickly | Inspect the actual delay distribution and accepted lead rate |
| Ecommerce purchase | Several days to a few weeks | Research and return visits may matter | Compare new and returning customers separately |
| Subscription or software trial | Long enough to cover evaluation and signup | Multiple sessions may precede the trial | Separate trial start from paid activation |
| B2B or high-consideration lead | Potentially longer | The buying process includes research and approval | Use CRM stages and offline conversion dates |
| View-through reporting | Shorter than the click window | Impression-only credit has more uncertainty | Report separately and test incrementality |
Workflow
Build the control loop step by step
Complete the measurement and validation steps before a source, bid or budget decision becomes permanent.
Define the accepted event
Choose the exact business outcome that will receive attribution credit.
Measure conversion delay
Calculate the time between qualified interaction and accepted outcome.
Separate interaction types
Create distinct click, view and engaged-view reporting.
Choose a working window
Select the shortest period that captures the meaningful buying cycle.
Normalize comparisons
Align event definitions, time zones and windows across platforms.
Run sensitivity scenarios
Recalculate results under short and long alternatives.
Set operational rules
Use a conservative window for source decisions and document any broader reporting view.
Review after material changes
Reassess when the GEO, offer, checkout, format or conversion definition changes.
Eligibility rule: interaction timestamp ≤ conversion timestamp ≤ interaction timestamp + selected window Sensitivity check: compare CPA and accepted conversion volume under short, working and long windows.
Pre-launch check
Quality gate before the campaign depends on the data
Use the checklist as a release gate. A missing identity, inconsistent window or broken redirect can invalidate later optimization.
Accepted conversion event documented
Click and view windows reported separately
Time-to-conversion distribution measured
Platform time zones aligned
Counting method aligned
Sensitivity analysis completed
Native and normalized reports labeled clearly
Window reviewed after major campaign changes
Standards and primary documentation
Worked scenarios
How the decision changes in real campaign conditions
Use these examples to separate the metric from the action. The same headline number can require a different response when the objective, data quality or business outcome changes.
Fast mobile action
A mobile utility campaign records most accepted installs within the first day after a click. A long click window adds little useful volume and increases the chance that unrelated later activity receives credit. Use a short operational window for source optimization, then check a longer reporting window to confirm that the ranking of sources does not change materially. Keep post-install events separate because their delay distribution can be longer than the install itself.
High-consideration B2B lead
A B2B campaign produces form submissions after several research visits, and accepted opportunities appear later in the CRM. A one-day window understates the campaign, while a very long window can overclaim brand-driven demand. Measure the interaction-to-lead and lead-to-qualified-opportunity delays. Use a documented working window for campaign reporting and a separate CRM analysis for pipeline impact. The media team should not optimize from raw form submissions when the business decision depends on qualified outcomes.
Platform totals disagree
Two advertising platforms both claim the same purchase. One uses a longer click window and includes view-through credit, while the other uses a shorter click-only window. Do not add the totals together. Build a normalized report with one accepted purchase event, one time zone and comparable windows. Keep native reports for platform optimization, but use the normalized view for budget allocation. Run an incrementality test when overlapping credit remains material.
Limits
What this measurement cannot prove by itself
No attribution window can determine causality on its own. It changes eligibility for credit inside a reporting model. Incrementality testing is required when the question is whether advertising created net-new outcomes.
A window chosen from historical behavior can become outdated after changes in price, offer, device mix, geography or checkout flow. Review the distribution rather than treating the setting as permanent.
Evidence gate
Before acting on how the decision changes in real campaign conditions, set the evidence threshold and review window in advance. Record the baseline, name the decision owner and define the maximum change that can be made in one cycle.
Operations
Build an attribution-window policy
A useful policy names the owner, the accepted conversion event, the default click window, the view-through rule, the time zone and the exception process. It also states which report is used for source optimization and which report is used for broader budget allocation. Without that separation, a media buyer may pause a source from a conservative click-only view while a finance report uses a longer blended window, creating two apparently correct but incompatible decisions.
Keep the policy close to the data. Include the current setting in campaign briefs, tracker configuration and dashboard notes. When a window changes, record the date and avoid comparing the before and after periods as though the measurement rule stayed constant. For long-running campaigns, save a normalized view using a fixed historical window so trend analysis is not rewritten every time a platform default changes.
Finally, connect the policy to conversion maturity. A campaign should not be judged before enough of the chosen window has elapsed. Use a provisional report for early pacing and a matured report for final source decisions. This prevents recent sources from looking artificially weak simply because their conversions have not had time to arrive.
A mature reporting process also distinguishes event time from report time. The conversion can occur today but be credited to a campaign interaction from several days earlier. Daily campaign totals can therefore continue to change as the window matures. Mark recent periods as provisional, and avoid comparing a fully matured week with a current week that still has conversions arriving. For finance and executive reporting, freeze the reporting date and explain the maturity rule so later backfill is expected rather than treated as an unexplained discrepancy.
Use a consistent naming convention for every report export. Include the event, interaction type, window, model and extraction date in the file name or dashboard subtitle. This small governance step prevents an old report from being compared with a newer report that uses different attribution settings.
Reporting check
Use build an attribution-window policy only after the reporting base is large enough to be credible. Keep the original control intact, document exclusions and schedule a follow-up check before expanding the decision.
Questions
Choose attribution windows that match the buying cycle: FAQ
Practical answers for media buyers, analysts and campaign operators.
What is an attribution window?
An attribution window is the period after an eligible ad interaction during which a later conversion can receive credit for that interaction.
Is an attribution window the same as an attribution model?
No. The window determines which interactions are eligible. The model determines how credit is assigned among the eligible interactions.
What is a click-through conversion window?
It is the time after an ad click or qualifying interaction during which a conversion may be attributed to that interaction.
What is a view-through conversion window?
It is the time after an ad impression during which a conversion may receive impression-based credit even though the user did not click.
Should the view-through window be shorter?
Often yes, because impression-only credit has more uncertainty. Keep it separate and validate it with broader measurement or incrementality testing.
How do I choose the right window?
Measure the real time-to-conversion distribution and use the shortest window that captures the meaningful buying cycle for the decision.
Why do platforms report different conversion totals?
They can use different windows, models, time zones, event definitions, counting rules and identity methods.
Can I use one window for every campaign?
A single governance rule can simplify reporting, but different offers and conversion types may need different windows. Document every exception.
What is sensitivity analysis?
It is the practice of recalculating performance under shorter and longer windows to see whether the decision is stable.
When should I change an attribution window?
Review it when the offer, GEO, device mix, checkout process, format or measured conversion event changes materially.
Continue the workflow
Related FroggyAds resources
Connect this guide to campaign setup, source controls, tracking and optimization.
Launch with a measurable plan
Use FroggyAds to test formats, GEOs, devices and sources with clear tracking, budget limits and source-level reporting. Results depend on the offer, creative, destination, bid and optimization process.