Blog /

Marketing Strategy

Hotel Time-Decay Attribution vs. Last-Click: Booking Credit

Compare hotel time-decay vs last-click attribution: how credit splits change, what to check, and why incrementality matters.

By Shahar Rubin, Co-founder

•

Updated Oct 29, 2026

I use last-click to identify the closing click and time-decay to divide credit across tracked touches, with more weight on recent ones. Before comparing them, I keep booking values, tracking rules, and attribution windows the same.

For the article’s $1,500 booking, last-click gives branded paid search 100% of the credit. Time-decay gives it 53.33%, assigning the remaining $700 to earlier touches. That’s a change in reporting - not proof that those touches caused the booking.

Quick Comparison

What I compare	Time-decay	Last-click
Credit split	Shared across eligible touches, weighted by recency	All credit goes to the final qualifying click
Earlier touches	Receive a share within the window	Receive none if a later click qualifies
Data needed	Linked touches and timestamps	Final qualifying click
Reporting use	Review the tracked booking path	Identify the closing touch
Native GA4 support	No; requires a separate calculation	Paid-and-organic last click is available

Before moving budget, I check the window from interaction to booking, direct-traffic rules, duplicate-click handling, cancellations, and net room revenue. If time-decay is available, I also check its half-life.

My bottom line: <u>credit is not added bookings</u>. I use an incrementality test, such as a geo-holdout, to check whether a campaign generated bookings that would not have happened otherwise.

One Hotel Booking Journey, Two Attribution Models

Hotel Time-Decay vs. Last-Click: $1,500 Booking Credit

For this $1,500 booking, the time-decay model uses a seven-day half-life and normalized shares. Both models evaluate the same journey; only the credit split changes. The same booking can produce very different credit splits.
Touchpoint Timing
Channel
Time-decay credit
Last-click credit
22 days before booking
Paid social
6.67% ($100)
0% ($0)
15 days before booking
Unpaid search
13.33% ($200)
0% ($0)
8 days before booking
Paid metasearch
26.67% ($400)
0% ($0)
1 day before booking
Branded paid search
53.33% ($800)
100% ($1,500)
Each interaction is seven days closer to booking than the previous one, so its time-decay weight doubles. The weights are 1, 2, 4, and 8. Normalize them to split the $1,500 booking. Percentages are rounded for display; dollar credits use the unrounded shares.
Time-decay shifts $700 of credit to the earlier touches; last-click shifts that $700 to the final branded search.

Touchpoint Timing	Channel	Time-decay credit	Last-click credit
22 days before booking	Paid social	6.67% ($100)	0% ($0)
15 days before booking	Unpaid search	13.33% ($200)	0% ($0)
8 days before booking	Paid metasearch	26.67% ($400)	0% ($0)
1 day before booking	Branded paid search	53.33% ($800)	100% ($1,500)

Use the Same Booking and Tracking Rules

Keep the booking value, attribution window, and tracking rules the same for both models. Apply the same direct-visit rule, too. In last-click reporting, a direct return can still get credit if no earlier qualifying campaign interaction falls within the lookback window.

Time-Decay vs. Last-Click: Key Differences

The booking stays the same. What changes is how each model divides credit across the path to that booking.

Comparison	Time-decay	Last-click
Credit distribution	Divides credit across eligible interactions, giving more weight to recent ones	Gives 100% to the final qualifying click
Early discovery	Gets a share if tracked and within the attribution window	Gets no credit when a later click qualifies
Final conversion touch	Receives the largest share in the journey	Gets all credit
Data requirements	Linked interactions and timestamps across the journey	The final qualifying click
Primary use	Full-path analysis	Closing-touch baseline

How Time-Decay Weights Earlier Interactions

Time-decay gives more credit to interactions closer to the booking. Earlier discovery still gets a share if it falls within the attribution window. A longer window can keep older touches in the calculation, though their share may be small. Time-decay reflects recency - not proven influence.

Why Last-Click Leaves Earlier Interactions Uncredited

Last-click gives all credit to the final qualifying click, often a branded search or metasearch click. Earlier touches get zero credit, even if they helped create demand.

Use last-click to identify the closing touch and time-decay to examine the full path. Then check whether your provider uses the same attribution window and direct-traffic rules for both models. These rules determine which earlier interactions get credit and which get none.

Attribution Windows and Reporting Support

Once the credit split is clear, check whether the reporting rules follow the same booking path.

Set the attribution window based on the time between an interaction and a booking - not the time between booking and arrival.

Check Direct-Traffic Rules and Available Models

Confirm whether your provider lets direct sessions receive credit.

Available models also depend on the reporting platform. In November 2023, GA4 removed first-click, linear, time-decay, and position-based attribution. It retained paid-and-organic last click and data-driven attribution. Treat time-decay as a historical comparison or a separate calculation, not a native GA4 model.

Ask Providers How They Credit Bookings

For the same booking path in Sail reporting, confirm which models it supports, the attribution window, and the time-decay half-life, if applicable. Also check the click-deduplication rules used to link interactions to reservations. Use these settings to compare credit on the same revenue basis.

Before shifting spend, keep booking value and cancellation treatment consistent. Define whether revenue reflects reservations or completed stays, and compare net room revenue excluding taxes and add-ons.

Conclusion: Match the Model to the Reporting Question

After comparing the same booking path, choose the model that fits your reporting question. Use last-click to credit the closing touch, or time-decay to spread credit across the full booking path. Time-decay still gives more weight to recent touches than to discovery.

Keep booking definitions, windows, and channel rules identical. Otherwise, differences in those rules can distort the credit split.

Attribution assigns credit; it doesn’t prove incremental booking lift. Before shifting budgets, run an incrementality test to check whether the campaign added bookings.

A geo-holdout is one such test.

FAQs

How do I choose a half-life for hotel bookings?

Choose a time-decay half-life that matches how long guests take to book after their first ad interaction. Use past booking lead times and platform attribution windows to keep expected conversions within your measurement window, before you reconcile stays in your booking engine/PMS.

This helps prevent giving interactions credit too soon or missing late bookings. Check your results by reconciling attributed bookings with net room revenue from completed stays.

How do tracking gaps affect attribution credit?

Tracking gaps break the link between ad clicks and booking revenue. Without cross-domain tracking - for example, when a booking engine uses a different domain - a returning visitor may appear as a new user at checkout. That disconnects the booking from the original ad.

Redirects can also break the link if they remove parameters like _gl. Attribution models then can’t accurately assign credit because the booking appears to come from an untracked source.

Should I change my budget when attribution credit shifts?

Change budgets only when a shift in attribution credit also reflects a change in completed-stay revenue after costs - not just attributed bookings. Time-decay can credit earlier interactions that last-click misses. Treat the attribution window as a reporting lens.

Apply rules consistently, match ad-reported bookings to reservations, and account for cancellations and no-shows. If completed-stay revenue per ad dollar holds steady, don’t move budget just because attribution credit shifted.