Magdala Hotel: $87K Meta-attributed booking value, Jul-Sep 2026

From 2 Jul to 30 Sep 2026, the Meta campaign BrightStudios built and ran for Magdala Hotel recorded $87K in Meta-attributed booking value at 18.3x return on ad spend. Magdala is a 150-room design hotel on the Sea of Galilee (the Kinneret) that sells its stays on its own booking engine, and the campaign had two jobs: sell those stays to Israeli guests, and show which ads the bookings followed.

For the first four weeks, Meta attributed no bookings to the campaign at all. Then, on 29 and 30 Jul, it credited the first ones to a reel offering 20% off two nights in August. August went on to account for 70% of the booking value Meta attributed between 2 Jul and 30 Sep.

This was our final summer with Magdala. We worked with the hotel for five years, until October 2026. This case study covers the positioning brief we wrote in June 2026 and the Meta campaign that followed it.

Brand: Magdala Hotel

What they sell: Stays at a 150-room design hotel on the Sea of Galilee (the Kinneret), booked on the hotel's own booking engine. This campaign was aimed at Israeli guests.

Location: Northwest shore of the Kinneret, Israel, at the foot of Mount Arbel

Engagement dates: Five years working together, finished October 2026. This case study covers the June 2026 positioning brief and the Meta campaign launched on 2 Jul 2026. Results cover 2 Jul to 30 Sep 2026.

Channels: Meta ads (one Sales campaign optimizing for booked stays), including reels posted on the hotel's Instagram and promoted as ads

What we did: Positioning brief for Israeli guests, design and Hebrew copy for the four launch static ads, booking tracking (Meta dataset with Conversions API, plus the Purchase-event install spec), campaign build and media buying. New offer ads went in through the season, timed to the hotel's calendar.

Headline result: $87K in Meta-attributed booking value at 18.3x return on ad spend, 2 Jul to 30 Sep 2026

Measured by: Purchase event with booking value on the hotel's booking engine; Meta attribution 7-day click, 1-day view, 1-day engaged view

Campaign structure: One campaign with one ad set through 30 Sep 2026 (a second, Winter Promotions, went live on 9 Oct 2026 and is not included); twelve ads delivered between 2 Jul and 30 Sep 2026

All figures below are Meta-attributed, 2 Jul to 30 Sep 2026, converted from shekels to US dollars at Bank of Israel monthly averages and rounded down. ROAS is calculated in shekels.

$87K in Meta-attributed booking value

18.3x return on ad spend

70% of that value in August: $61K at 37.09x for the month

67% of that value followed three summer offer reels, promoted from 21 Jul, 27 Jul and 10 Aug

$28K in the strongest 7-day period, 20 to 26 Aug

Two things had to be in place before anyone could judge the ads: a booking event that records each stay with its value, and a campaign simple enough for a modest daily budget to learn from.

In June 2026 we wrote a positioning brief for Israeli guests. We audited the product, mapped the competing Kinneret hotels and defined three audience segments: couples on a romantic getaway, families in summer, and culture-and-calm guests aged 40 and up.

The brief set Hebrew language rules too. One of them: lead with the Kinneret (הכנרת), the name Israelis use, rather than Sea of Galilee. It positioned Magdala for domestic guests around the lake, the gardens and a short break from the center of the country. The positioning it worked from: a quiet design hotel on the Kinneret, the abroad-feeling break without the airport.

The brief also set the measure: booked stays with their value, recorded on the booking engine. We created the hotel's Meta dataset with the Conversions API enabled and wrote the install spec for the booking-engine developer. The spec asked for a Purchase event with the booking value in shekels, deduplicated between browser and server events. Every booking value and ROAS figure in this case study rests on that one event.

We built the campaign and launched it on 2 Jul 2026. The build sheet pooled the brief's three audiences into one Sales campaign with one broad ad set, and the campaign optimized for Purchase.

That kept a modest daily budget in one place, and every booking Meta recorded fed the same ad set's learning. Most new offers went in as new ads inside it.

Four statics went live on launch day. We designed them and wrote their Hebrew copy.

By 14 Jul, five early summer ads had gone live. On 21 Jul the first of three summer offer reels joined them. It carried a two-night family package for July to August, with one family attraction included.

On 26 and 27 Jul, one of the early ads was switched over to 20% off two nights in August. On 27 Jul a second reel went live with the same offer, so two ads now carried it.

The same day, three of the early ads were paused. One of them was the 2 Jul ad with the family package, the offer the 21 Jul reel also carried. Meta had not yet attributed a booking to any ad, and those three recorded none.

The wait lasted from 2 to 28 Jul 2026. On 29 and 30 Jul the first attributed bookings came in, and Meta credited every one of them to the reel promoted from 27 Jul.

July closed at $3,966 in attributed booking value and 2.62x. Four weeks of cost sat against two days of attributed bookings.

On 10 Aug the third summer reel went live, with one midweek night in August at a special price. It finished narrowly ahead as the top ad: $20K in Meta-attributed booking value at 42.55x for that ad, 10 Aug to 30 Sep 2026.

The other two reels sat just behind it. Over 2 Jul to 30 Sep 2026, the 21 Jul family package reel recorded $20K at 26.44x, and the 27 Jul reel, 20% off two nights in August, recorded $18K at 35.08x. Together the three accounted for about two-thirds of the period's attributed booking value and 89% of August's.

August closed at $61K in attributed booking value and 37.09x, 70% of the 2 Jul to 30 Sep total. The strongest 7-day period, 20 to 26 Aug, added $28K.

Two of the early ads recorded Meta-attributed booking value as well, all of it in August: $5,366 between them, 6% of the 2 Jul to 30 Sep total. Of that, $4,135 went to the early ad switched to 20% off two nights in August.

On 26 Aug, with the peak month still under way, the first holiday ads went live: a Rosh Hashana reel (a holiday stay with a holiday meal and family activities) and a Sukkot reel.

On 3 Sep the summer reels were paused, and two more Sukkot ads followed. A reel on 5 Sep carried a family holiday stay with activities for the whole family. An image on 10 Sep listed the holiday activity program: live music, wine workshops and tastings, a circus and juggling workshop, wooden games, prize bingo and a saxophone show.

With the summer reels paused, the four holiday ads accounted for all of September's attributed booking value: $21K at 13.79x for the month. The Sukkot image, the last of the twelve ads to go live, was the month's top ad: $13K at 15.46x, 61% of September's value. Over 2 Jul to 30 Sep, the four holiday ads recorded $22K in Meta-attributed booking value, 26% of the total.

In October we added a second ad set, Winter Promotions, for winter offers. It went live on 9 Oct 2026, after the period this case study measures, so its results are not included.

Our work with Magdala finished that same month, five years in.

Booking value comes from the Purchase event on the hotel's booking engine, attributed by Meta (7-day click, 1-day view, 1-day engaged view). Attribution credits an ad with a booking that followed it. It does not show how many of those bookings would have happened without the ads. Values have not been reconciled against cancellations.

Shekels are converted to US dollars at Bank of Israel monthly average rates and rounded down. ROAS is attributed booking value divided by ad spend, calculated in shekels, for the period, the month or the ad named.

All twelve ads ran in one ad set, so Meta's delivery choices shape how the value splits between ads.

One possible duplicate booking on 27 Sep 2026 is excluded.

The July weeks before the first attributed bookings on 29 Jul 2026 still count as cost inside the 18.3x.

All dates 2026.

June: positioning brief for Israeli guests; Meta dataset created with the Conversions API; Purchase-event install spec written for the booking-engine developer

2 Jul: campaign launched (one Sales campaign, one ad set, optimizing for Purchase); the four launch statics live

2 and 14 Jul: five early summer ads go live

21 Jul: first summer offer reel (a two-night family package with one family attraction)

26 and 27 Jul: one early ad switched to 20% off two nights in August

27 Jul: second summer offer reel (20% off two nights in August); three early ads paused

29 and 30 Jul: first Meta-attributed bookings, all credited to the 27 Jul reel

10 Aug: third summer offer reel (one midweek night in August), the top ad by attributed booking value

20 to 26 Aug: strongest 7-day period, $28K in Meta-attributed booking value

26 Aug: Rosh Hashana reel and Sukkot reel go live

3 Sep: summer reels paused

5 Sep: Sukkot family reel goes live

10 Sep: Sukkot activity-program image goes live, September's top ad

30 Sep: results period closes at $87K in Meta-attributed booking value and 18.3x

9 Oct: second ad set, Winter Promotions, goes live (results not included)

October: our work with Magdala finishes, after five years together

For a hotel, or any business that sells bookings through its own engine:

Confirm the booking event is recording before you spend. Put a test booking through the engine and check that it reaches Meta with its value before the first ad goes live. The campaign optimizes for that event, and every booking value and ROAS figure you read later comes from it.

Keep a modest budget in one place. One campaign, one ad set, new offers added as ads. Split three ways, the same budget gives each ad set less booking data to learn from.

Know which offer each ad carried before you read the split. Magdala's 20% off two nights in August ran in two ads. The reel promoted from 27 Jul recorded $18K in Meta-attributed booking value, 2 Jul to 30 Sep 2026, and the early ad switched to the same offer recorded $4,135. The family package ran in two ads as well. The one from 2 Jul recorded nothing, but it was paused on 27 Jul, before Meta attributed a booking to any ad. Meta chooses which ad to show inside one ad set, so read the split as a record of where bookings followed.

Plan offers around your own calendar. Get the next holiday's ads live while the current season is still booking. Magdala's first Rosh Hashana and Sukkot ads went live on 26 Aug, inside the peak month, eight days before the summer reels were paused on 3 Sep.

Tell us about your project: https://bybrightstudios.com/start