An open, pre-registered study

Booking friction: the measured comparison

Most hotel websites lose bookings in the last step, where a guest leaves your domain and meets a form built for the system behind it, not for the person in front of it. Here, Shopify checkout, the Plekify app and Apaleo took 4 clicks and 0 hand-offs to the payment page, with a friction index of 10.9, the lowest measured (tested 2026-06/07, n=7), and 19 seconds to the payment page; every figure is tagged MEASURED, PROVISIONAL or DERIVED, so you can see how firm each one is. The methodology is public at github.com/Matthew-Swart/plekify-booking-friction-study, and this page is the evidence behind every number on plekify.com.

See the live demo Book a 20-minute call View the GitHub repo ↗

Recipe

Shopify checkout, the Plekify app, Apaleo: the recipe that scored lowest.

Shopify checkout: the conversion layer

250M+ buyers already use Shop Pay. Shop Pay converts up to 50% higher than guest checkout. Checkout is one page, hosted on the hotel's own domain, secured to PCI-DSS Level 1. This is the layer where bookings are won or lost.

Plekify: the bridge

Plekify connects PMS inventory to a Shopify storefront on the hotel's own domain. The path took 4 clicks and 0 hand-offs to the payment page, in 19 seconds (tested 2026-06/07, n=7). Friction index 10.9, the lowest of 13 systems measured in this study.

Apaleo: the system of record

Rates, availability and operations stay in the PMS, exactly where hoteliers expect them. Plekify is the booking engine; Apaleo is the system of record, run on an open API. Read more about how Apaleo works.

For the Apaleo integration story behind these numbers, see how Plekify connects PMS inventory to Shopify.

Together, Shopify checkout, Plekify and Apaleo form the highest-conversion recipe measured in this study.

Findings

Three headline results

Shopify checkout, the Plekify app and Apaleo took 4 clicks and 0 hand-offs to reach the payment page, with a friction index of 10.9 (tested 2026-06/07, n=7). Time to the payment page: 19 seconds. Shop Pay's 250M+ buyers, and its up to 50% higher conversion than guest checkout, complete the recipe.

Off-domain booking engines, NightsBridge's included, carry a handoff penalty: 1 handoff, 16 form fields, friction index 22.9 vs 10.9. Industry data puts direct bookings more than 60 percent above OTA value ($519 vs $320), with cancellation rates of 10.6 percent vs 21.8 percent.

For hotels running NightsBridge today, read what the handoff penalty means in practice.

Most systems score at or below 2.0; agents get blocked at the cart (Cloudbeds) or at identity walls (OTAs). Only Plekify scores Open, 3.0 — the agent-readiness story continues at /pages/mcp.

To see how this same inventory reaches booking agents, read how Plekify surfaces availability through MCP.

The table

All 13 Systems, Side by Side

Friction index: lower is better. Agentic-readiness score: higher is better. Form fields, handoffs, outcome, and payment method are reported as measured, no adjustment.

System	Friction (LFI)	Clicks	Fields	Domain hand-offs	Outcome	Payment	ARS	Apaleo IBE †	Tested
Plekify	10.9	4	0 (Shop Pay) / 22 (guest)	0	reached payment · 19s	Shop Pay · Apple Pay · Google Pay · card	3.0	†	2026-06/07
Mews	16.6	—	—	1	reached payment · 67%	card (Mews Payments)	1.9	†	2026-06/07
NightsBridge	22.9	7	16	1	reached payment	card · EFT	1.35	†	2026-06/07
SiteMinder	29.4	4.3	14	1	reached payment	card (property gateway)	1.5	†	2026-06/07
Cloudbeds	—	—	—	—	agent-blocked at cart (isTrusted guard)	—	2.0	†	2026-06/07
RoomRaccoon	—	—	—	—	agent-blocked (kept as exhibit)	—	1.4	†	2026-06/07
Booking.com	—	17	—	—	API gated · bot-defended	—	1.8	†	2026-06/07 · 10-07
Airbnb	—	—	—	—	login-walled at checkout advance	—	1.25	†	2026-06/07 · 10-07
Expedia	—	—	—	—	hard bot-wall (4/4 sessions)	—	0.6	†	2026-06/07 · 10-07
Travelstart	30.2	12	7	2	reached payment · 118s	card (Stripe)	1.4	†	2026-10-07
LekkeSlaap	63	26	42	0	payment-gated behind Reserve (Turnstile)	unreached (gated)	—	†	2026-10-08
Stayntouch	—	—	—	—	—	—	1.6	†	—
OPERA-ecosystem	—	—	—	—	—	—	1.2	†	—
Apaleo IBE	—	—	—	0 (API-driven, same domain)	no public demo IBE to walk	card via Apaleo Pay (Adyen Drop-in)	—	—	—

† Apaleo IBE not scored. Apaleo is the system of record in this study; Plekify is the booking engine tested.

Provisional: 3-run morning sample. Full-day sampling to follow.

Checkout was tested in sandbox mode. The payment methods listed are each platform's publicly available methods in US and EU target markets, not those enabled on the test store.

Every figure carries the date it was tested; systems re-tested on 2026-10-07 or 2026-10-08 are marked with those dates, all others reflect the 2026-06/07 dataset.

METHOD

Methodology

Friction index: lower is better. Agentic-readiness score: higher is better. Form fields, handoffs, outcome, and payment method are reported as measured, no adjustment.

01Environment

Each system tested in a live browser session, from the public booking page.

02The systems under test

We tested 13 systems spanning four categories: hotel IBEs, PMS booking engines, channel managers and OTAs. Each system completed the same booking task on the same property, dates and party size. The set was chosen to represent how hoteliers actually sell today, not to isolate a single competitor. Full system names and configurations are listed in the public repo alongside the raw run data.

03The friction formula

Friction is scored as `F = C + 6.6H + Fld_excess + P + 9.8I − 3.7A − 5.4Acc`. Each letter corresponds to a measurable step in a real checkout: clicks, domain handoffs, excess form fields, page-load time, interactive interruptions, address autocomplete and accelerated checkout. The formula is pre-registered and frozen. We disclose a known discrepancy in panel 4 rather than silently reconcile it.

F = C + 6.6H + Fld_excess + P + 9.8I − 3.7A − 5.4Acc

04The weights

Weights: C clicks 1.0 · H domain handoffs 6.6 · Fld_excess form fields beyond 8, 1.0 · P page-load seconds 1.7 · I interactive interruption (CAPTCHA or challenge) 9.8 · A address autocomplete −3.7 · Acc accelerated checkout (Shop Pay, Apple Pay, Google Pay) −5.4. Weights are calibrated to the Form-Field Unit, where one field equals 4.1% relative conversion loss. Disclosure: the displayed formula string shows P unweighted, while the pre-registered protocol carries P at 1.7. The string is frozen as originally published; the discrepancy is disclosed here rather than silently reconciled.

Signal	Weight
C — clicks	1.0
H — domain hand-offs	6.6
Fld_excess — form fields beyond 8	1.0
P — page-load seconds	1.7
I — interactive interruption (CAPTCHA/challenge)	9.8
A — address autocomplete	−3.7
Acc — accelerated checkout	−5.4

05The agentic-readiness rubric

Agentic readiness is scored as `ARS = 0.20·SD + 0.15·BM + 0.20·CW + 0.15·AP + 0.20·API + 0.10·PA`, on a 0–3 scale. Signals: SD structured data, BM robots/bot posture, CW CAPTCHA/WAF friction, AP express payments, API public API, PA protocol adherence. "Agent-ready" means an autonomous agent can discover and complete a booking on the open web without a private commercial agreement. This is separate from the friction score and measures a different property of each system.

Signal	Weight
SD — structured data	0.20
BM — robots/bot posture	0.15
CW — CAPTCHA/WAF friction	0.20
AP — express payments	0.15
API — public API	0.20
PA — protocol adherence	0.10

06Raw runs

Every run is published, including slow ones.

07Standardized test case

2 adults, 1 room, 2 nights, lowest published standard double, standard flexible rate.

08The statistics

Friction times are modeled as lognormal. Confidence intervals use cluster bootstrap with bias-corrected acceleration and 10,000 resamples. We also ran a sensitivity analysis on the un-recalibrated click weight to check whether results depend on that calibration choice. Full statistical code is in the public repo alongside the raw runs.

09Ethics

No real bookings were completed during any run. No CAPTCHA solving was performed. No terms-of-service violations occurred in the course of testing. Where a system presented a CAPTCHA or similar challenge, that run was scored as an interactive interruption and stopped there.

10Limitations and conflict of interest

This study covers one test property and one market; results may not generalize beyond that scope. The Mews result is provisional, based on a 3-run morning sample. Plekify sponsored this study and topped the table. We mitigate this conflict with a pre-registered protocol, a public repo of raw runs, and an open invitation to replicate or challenge the results.

11Reproducible scripts

Scripts are in the public repository.

Sources

Every Figure, Sourced

Each number on this page links to an inline popover citing its source. The full study, methodology, and raw runs sit in a public GitHub repo, open for inspection and replication.

This study covers one test property in one market. Plekify sponsored the study and topped the table. The protocol was pre-registered before testing began, raw runs are public, and we invite replication by any hotelier, technologist, or competitor.

Corroboration

Five Outside Studies, Graded by Source

These five studies were run by others, not by Plekify, and each is graded by source type: independent research, peer-reviewed, or vendor-disclosed. Their magnitudes sit alongside our own; none of them is a replication of our protocol.

Independent research Baymard Institute — checkout usability research
The average large ecommerce site can gain a 35.26% conversion-rate increase through better checkout design alone — within a point of the 36% platform-commissioned figure cited above.
Peer-reviewed Unal & Park — “Fewer Clicks, More Purchases”, Management Science 69(12)
35-month quasi-experiment: registering for one-click checkout increased customer spending by ~28.5% and site engagement by 7%.
Vendor-disclosed Stripe — testing the conversion impact of 50+ payment methods
Offering Apple Pay lifted conversion 22.3% on average across Stripe’s network; surfacing wallets early roughly doubled it versus payment-step-only placement.
Vendor engineering record Cloudbeds — iFrame deprecation and migration overview
Cloudbeds retired its iframe booking engine for performance, security and analytics reasons; the replacement is a web component that runs on the property’s own domain.
Agency benchmark 80 DAYS — average hotel website conversion rate
Independent hotels: 0.73% of sitewide sessions convert vs 3.28% of booking-engine sessions (chains: 1.90% / 6.80%; 2019 data).

Trust

Pre-registered protocol, raw runs public

PCI-DSS Level 1 certified

99.9% uptime

250M+ Shop Pay buyers

See it on your own inventory

Plekify is the live two-way bridge between your Apaleo PMS and Shopify — your rooms, your rates, your checkout, on your own domain.

See the live demo Book a 20-minute intro Read the repo ↗ No sign-up required, runs in your browser