Above the clouds: How SkiBig3 turned scattered data into a smarter way to run the business
In travel and tourism, “windows” are short. The ability for brands to have full visibility into their customer behavior is the difference between a mediocre and a blockbuster season.
“Ad platforms change their rules every year. A clear view of your own customers is the one strategy that holds up no matter what changes next”
— MIKE TYLER, CEO, WAR ROOM
For most destination tourism businesses, the buying pattern is on repeat. Guests research for months, book across multiple channels, and pull the trigger on their purchase inside a relatively short “activation” window. The data trail scatters across the stack: CRM knows one thing, website analytics knows another, the booking system knows a third, and every ad platform in the media mix has its own “partial” view. None of these systems talk to each other.
Every real business question; “Who are our best guests, what are they worth to us over time, which segments are we losing and to whom?” comes back as a manual data pull, a spreadsheet, and a pivot that arrives after the season has moved on.
SkiBig3 was living the pattern. Three of Canada’s most iconic mountain resorts (Sunshine, Lake Louise, and Mt. Norquay), a global audience, and years of first-party data that couldn’t get to the systems making the daily decisions.
Passes like Ikon and Mountain Collective meant a skier could decide to come to Banff almost anytime. The problem was that SkiBig3 often didn’t know they were coming until they showed up. Understanding guest behaviour and intent earlier changes that: the team can reach these skiers before they arrive, shape the trip, and improve the experience once they’re on the mountain.
Enter War Room, and the power of Kedet Intelligence.
8x
Longer customer memory
11k+
Previously invisible guests
Daily
Automated audience refresh across the paid media stack
Kedet Intelligence, War Room’s data unification layer, was built to solve exactly this. It sits on top of any modern data warehouse and turns scattered customer signals from CRM, website, booking, and ad platforms into one activation-ready view of the customer. War Room installs it. War Room runs it. And when a business can see its customer at that level of clarity, better decisions start compounding across media, retention, acquisition, and how the next season gets planned.
Here’s how we did it
Objectives:
Picking the route
Before touching a single system, SkiBig3 and War Room had to be on the same page about what a working solution actually had to deliver:
Match every known customer back to platform data. Close the gap between what SkiBig3’s CRM already knows about a guest and what the ad platforms can actually see. Every conversion counted. Every audience built from real customer signals.
Extend the useful life of customer data past platform expiry. GA4 and the ad platforms discard signals at the 18-month mark. Real business decisions need a longer memory than that.
Unlock audiences too small for platforms to build on their own. A multi-year lookback makes narrow, high-value segments (checkout abandoners, families and groups, lapsed premium) large enough to actually activate.
Optimize for the customers worth having, not the cheapest clicks. Lifetime value is now in the driver’s seat.
Tie media spend to real revenue. Including the offline conversions no ad platform will ever report.
Give the marketing team direct answers. Custom reporting on demand, not manual pulls and delayed pivots.
Fragmented data ownership. Customer, booking, web, and ad-platform data lived in different systems, with different schemas, retention windows, and access rules.
Platform data expiry. GA4 and ad platforms discard signals at the 18-month mark. Multi-year customer behaviour, the exact thing SkiBig3 needed to see, was falling off the cliff of every platform report.
There are no “standard” funnels in this journey. Mountain guests research for months, book across channels, and often convert close to travel. Last-click optimization paints a misleading picture.
Matching CRM customers back to platform data. SkiBig3 always knew who their customers were in the CRM. The ad platforms didn’t. Closing that match rate required a secure identity bridge between the two systems, built so no raw customer data ever leaves SkiBig3’s environment.
Under-reported platform tracking. Meta and other platforms under-report attribution. Platform numbers alone aren’t a reliable source of truth for a business making revenue decisions.
Challenges:
Reading the terrain
We came in with clear objectives, but the landscape made this more adventurous than a typical data warehouse integration:
Strategy:
Dropping the line
Select a step to explore
One customer, one identity, everywhere it matters
SkiBig3’s CRM already knew the customer. The ad platforms didn’t. Every Google Ads click, Meta impression, and website session was a stranger to the customer record it should have belonged to. Before building anything on top, we built the seam underneath: a unified and anonymized customer ID that connects SkiBig3’s CRM to every platform they run media on. That match is what lets every other layer of the strategy actually work.
Only the minimum data required for activation and measurement leaves the SkiBig3’s environment: direct identifiers such as email addresses are normalized and converted into pseudonymous, fixed-length values using the one-way SHA-256 hashing algorithm before transfer. Kedet uses these hashes to support privacy-conscious matching with platforms such as Google and Meta for audience creation, campaign measurement and reporting, while neither receiving nor storing the original plaintext identifiers. Because hashing is performed before transfer and no decryption key exists for a one-way hash, Kedet never holds raw or decryptable customer PII.
History as strategy
With CRM data now reaching every ad platform, years of SkiBig3’s customer history stopped being archived and started being usable. We built a new audience structure from that history, organized around how a resort actually grows revenue: acquisition audiences (bring in new potential buyers who look like your best past ones), conversion audiences (turn engaged non-buyers into first purchases), and customer growth audiences (increase value from customers you already have).
Inside each group sat specific audiences the platforms couldn’t have built on their own, because they never had access to the data: high-LTV lookalikes, multi-year lapsed win-back, upsell paths (Lift Ticket to Ski Stay Package, Ski Stay Package to Rentals), families and groups to ski lessons, and predictive audiences that model behaviours before they show up in-platform.
Just as important: knowing who NOT to spend on. Low-value lapsed customers stay out of premium campaigns unless there’s a specific reason to chase them.
Six of the audiences in this new structure are predictive. Not lookalike audiences, which take a seed list and find similar people the platform already knows about. Predictive audiences use SkiBig3’s actual multi-year customer behaviour to model who’s likely to convert, renew, or lapse before they show any signals the platforms can see. Users Similar to High-LTV Customers. High-Intent Non-Buyers. Customers Approaching Renewal. High Renewal Likelihood. This kind of modeling only becomes reliable at the data volume that comes from years of unified customer history. Platform-native lookalikes can’t reach here.
The list is already live
The old workflow was familiar. Request the audience, export it, share it over SFTP or email, activate it in the ad platform, and discover the data was already stale before it ran. Kedet Intelligence retired that. Audiences now refresh daily and flow automatically into every activation channel SkiBig3 runs.
Exclusion logic runs at the source, so the same guest doesn’t land in overlapping audiences and skew the results. And where platform-side tracking under-reports (Meta especially), Marketing Mix Modeling gives SkiBig3 a second read: comparing media spend directly against CRM revenue, a measurement path that doesn’t depend on any platform’s pixel.
Visibility as a superpower
A data warehouse is only valuable if someone opens it. Kedet Intelligence delivers custom reporting where the marketing team already works. Three reports do most of the heavy lifting: a CRM dashboard that shows customer composition and how guests are moving between segments; LTV-by-cohort tracking that shows how customer value has evolved across acquisition periods; and Purchase Path analysis with multi-touch attribution across the full journey. Business questions that used to take a week and three meetings now get answered in an afternoon.
One example: a single query surfaced an entire high-value guest segment (over 11,000 addressable customers) that had never been reachable through the ad platforms before. Not because SkiBig3 didn’t know these customers existed; they did. But the platforms couldn’t build the audience because the data lived somewhere the platforms couldn’t reach. Now it can.
“First-party data is really about knowing our skiers better and using that knowledge to improve the full experience. If we get that right, they’ll come back with more friends.”
— PETE WOODS, PRESIDENT, SKIBIG3
+31%
Revenue YoY
10.03
ROAS (+51% YoY)
-17%
Spend YoY
Many things contributed to this season’s results.
Kedet Intelligence is the biggest structural change, and the portfolio-level lift has moved sharply since it went live.
WHAT WE LEARNED
Key
takeaways
As paid media and data specialists, we don’t declare victory because a system is running. We look at whether the business changed. Here’s what changed for SkiBig3 since Kedet Intelligence went live:
First-party audiences outperform because they’re built from truth, not signals. Every audience War Room built for SkiBig3 was seeded with real customer behaviour across three-plus years, not proxy signals inferred by the platforms. That difference shows up in every campaign that runs against them.
Enhanced attribution changes what the algorithms learn. When platforms can finally see the conversions they were previously missing, Smart Bidding gets sharper signals to optimize against. Better signals in, better optimization out, across every channel.
Better audiences are worth more than more audiences. Revenue up 31%. Spend down 17%. The team is finding higher-value customers with less budget.
Retention becomes a strategy when the data lasts. Multi-year customer history unlocked audience types that weren’t buildable inside platform retention windows: predictive renewal, multi-season lapsed win-back, and high-LTV lookalikes seeded from history that goes back further than any platform allows.
Data operations and media operations belong to the same team. Data consultants build warehouses and leave. Media agencies work with what platforms give them. War Room does both. That’s why the data investment shows up in the revenue.