2023 Data Analytics
Requesting Jurisdiction: Saskatchewan Parks
Description: Saskatchewan Parks is currently building out its data analytics program and we are interested in learning about the practices and metrics that other jurisdictions use to make enhanced business decisions and measure performance.
Response Rate: 2/13 jurisdictions
15%
Questions asked:
What are the main components of your data analytics program in addition to standard occupancy and revenue reporting?
AB
PC
AB
Goals of the program will be:
a) Strengthening Information and Communication Outputs
a. Developing a Division-wide Data Analytics Community of Practice
i. Outputs- Data Tool kit for staff, Data Landscape for the Division on what tools are being used to what end,
b. Establishing a Day Use Statistics Visitation Program
i. To better understand volume of visitation and potential user conflicts
c. Data Administration
i. Developing an Intake form to triage requests
ii. Development and execution of Performance Metrics specific to Parks Visitation and Satisfaction
b) Enriched Data on Visitor Use and Recreation Patterns
a. Third Party Customer Profiling
b. Additional Social Science Data based research tailored to specific nodes under Master Development Planning and Capital Planning initiatives underway
(outputs include ArchGIS, Survey tools)
c) Innovation through Data
a. Data Driven-Economic Analysis
b. Data to support Maintenance Scheduling
c. Data deep dives based on government mandate and direction from Alberta’s Plan for Parks
PC
Visitation:
Monthly visitation statistics (based on a person-visit definition) by place (~130 parks and sites).
- Have archives back to 1930s.
- Visitation is a corporate indicator and base for many other initiatives in Parks Canada. Each place has a formula to measure visitation.
- Visitation can be spliced by many factors including program (e.g., parks/sites), business units (i.e., field units). prov/region, season, and distance to major urban area.
Visitor indicators:
Employ a nationally standardized onsite survey (random sampling/intercept) to collect operational indictors and other key operational information at select locations annually.
- Program has operated since 1999.
- Basis for core satisfaction (visit overall), enjoyment, and learning indicators
- Basis to identify factors such as average party size, frequency, length of stay/type of visit, party size and composition, aspects of market draw, satisfaction with programs/services (e.g., trails, washrooms, parking lots), motivation for visit, enjoyed most, info used to plan visit, purpose of visit (e.g., main stop / unplanned). Will feed attendance studies, economic impact et al.
Public Tracking indicators:
Employ quarterly surveys to track public awareness (aided, unaided) of Parks Canada, public support for PCA’s mandate, and main platforms (TV to social media) where Canadians get information about Parks Canada. Restarted tracking of corporate symbol (brand recognition) in 2022.
- Core tracking has occurred since 2010
- Awareness and support spliced by region, major urban centres (Montreal, Vancouver, Toronto), families (kids/no kids at home), age (35), and diversity (born/not born in Canada)
- Support is a corporate indicator; awareness, source, brand recognition are operational indicators
- Will use quarterly surveys to track other items as needed (e.g., travel intentions and to where)
Market analytics:
- Entry related: data is randomly collected (origin, postal/zip codes, country, groups) at welcome points/ gates through transactions and related prompts that is used to analyze tourism market draw of each park and site that participates (has sufficient data/collection rate). Data is analyzed using segmentation software.
- Analytic program has existed since ~2010
- Parks/sites receive report with info on market draw (% Canada/USA/International), break down of draw (prov %s; states, countries), distance decay (how far domestics come), composition (based on entry fees sold), some demographics (drawn from postal codes), and audience segments (Explorer Quotient (EQ) and custom PCA segments).
- Camping: the analytics team pulls the data (no Intellectual Property) to analyze the market draw for camping in each park.
- Similar to market analytics above, each park receives a report with info on camping market draw (% Canada/USA/International), break down of draw (prov %s; states, countries) by camping type, distance decay (how far domestics come), some demographics (drawn from postal codes), and audience segments (EQ and custom PCA segments) by draw and main camping types.
- Also do the same analysis/ provide analytics for reservable shuttles, parking (where in lieu of entry) and reservable guided hikes. Other offers will follow as more is made reservable on reservation system.
Camping analytics (operational):
Team who manages the system and will produce analytics such as # of reservations/cancellations, site nights available, nights occupied, occupancy rate (by season, by place, by main type of offer), source (e.g., walk up, online). Have analyzed lead time to book etc.
Economic impact:
Employ an expenditure input-output model to analyze the economic impact of PCA operational spending and visitor spending (e.g., GDP, jobs, taxes, labour income). Draws on many analytic programs to feed the analysis. Integrating land cover data with non-market valuation to estimate the value of natural capital assets in parks through ecosystem service analysis.
Sales:
The Finance department is responsible for tracking expenditures and reconciled revenues and their related official reporting.
- The analytics team tracks on demand sales through Point-of-Sale units nationally on a weekly basis and prepares dashboards to track progress (vs previous years).
- Data is spliced by sales type (e.g., entry, camping), program (e.g., NP, NHSs), region, and operational business units, and background data permits seeing change in fee buckets (e.g., golf, fishing fees, rentals) or initiatives.
Product/service specific:
Various operational info will be collected and/or analyzed depending on the product or service. For example:
1) E-commerce (third party)
- Sales of annual passes, when, type, $, and related market draw
- Track merchandise sales. Can pull data and splice as needed (who, amount, top items, types, etc.)
2) National information platforms and programs:
- Platforms: operational analytics are collected on a range of platforms. For example – e.g., call volumes into National Information Centre, web use analytics (public website), social media followers and popular content, e-newsletter recipients and click/read stats on newsletter content, media requests from travel reporters and resultant stories.
- Programs: examples: participants /reach of ‘Learn to Camp’ activities, volunteers (number, activities), citizen science events, contacts at outreach events.
3) Media campaigns: national advertising campaigns are evaluated using TBS tools. Promotion on third party sites (Expedia) will yield platform analytics (e.g., popular content, ROI, bookings/sales, type of accommodation).
4) Other specific: trail volumes, number of youth (school groups), commercial tours (companies, where, what), cruise ship visitors
What data analytics are analyzed which directly aid in making business decisions for the park organization?
AB
PC
AB
Not all of these are in the same stages of maturity:
- ReserveAlbertaParks overnight bookings
- ReserveAlbertaParks Customer Survey
- Point of sale bookings and sales eg passes, tours
- Point of sale customer surveys
- Kananaskis Conservation Pass sales and compliance data
- TraFX day use counters across parks regions
- Ecocounters day use counters across regions
- Potentially mobile data
- ArchGIS
- Social media engagement
- Website traffic
- Call centre data
- Visitor email and social DMs
- Provincial survey on visitation and satisfaction
PC
Some examples are detailed below, not exhaustive:
- Visitation: foundation /criteria for other analyses or teams (e.g., investment priorities, promotion /travel trade (what to showcase), product pilot opportunities, visitor safety (incidences per visitor)). Investment firm in USA uses select places to track performance of hotel stocks.
- Visitor surveys: product development, site selection (pilot studies), language profiles (areas), service standard improvements (e.g., trails, washrooms), investments (e.g., where work needed), travel trade, understanding connections and value of heritage places (what is important, enjoyed)
- Market draw analytics (entry/camping): initial market analytics and audience segmentation helped shift the orientation of the organization (from internal to public focused) for visitor products/services, outreach, information platforms – meet needs – and got buy-in on approach.
- Entry: Ongoing analytic work has highlighted the tourism pattern in PCA places, to help inform work in travel trade, promotion (what to promote to whom/where, what to showcase), nature of who is/is not coming (strong/ not as strong) in regions/places (product placement, how might to engage) etc. Reports are often shared with DMOs, and other local/regional tourism partners to share /work more closely together.
- Can also help to inform broader issues such as merchandise stock placement, selection of outreach venues, digital content strategies, product development, partner selection to reach new audiences.
- Camping: operational stats (e.g., occupancy) inform work with third party businesses for off season opportunities. Used nights occupied to produce heat maps of campgrounds to help identify under used/ or less desired campsites.
- Economics: often used for Memorandum to Cabinet / budget submissions, when engaging with external parties etc. for context / positioning. Ecosystem services has the potential to inform park expansion, restoration priorities, and asset placement.
- Sales: tracking partnership programs (e.g., the Canoo App – for newcomers to Canada – formerly known as the Cultural Access Pass), compliance (e.g., tracking annual pass use), magnitude of discounts/refunds (to see issues, where work is needed), popularity of specific merchandise items (for promotion, inventory issues, buy/sunset?), pass purchase trends (> annual vs others?), short term promotion (to boost visitation, etc.).
- National platforms: website (pinch points, website navigation improvements, content), Info centre (staffing / demand, anticipated issues to be prepared for), social media (what content works/does not, uses of each / patterns).
Do you complete any predictive data analytics and if so, what are you analyzing predictively (visitation, revenue, site type sales, etc.?)
AB
PC
AB
No, we are not currently using any machine learning tools at present.
PC
For the last 5 years or so, we have undertaken some forecasting activities related to visitation and revenues.
- Visitation: asked regularly for annual visitation before the season has even begun and where we are headed during the season.
- Created a multivariate regression model for national visitation (needs updating post pandemic) to forecast ‘magnitude’ of visitation ahead of the season.
- During the season, we employ rolling projections of visitation using proportional patterns. Visitation by month (% of year) does not change much year over year. Using this assumption, each month we can project year end total based on current supplied figures. Usually, the projection made after June visitation figures are supplied is within 1 to 3% of actual year end total.
- Revenue: as revenue is linked to visitation, we are often asked to forecast revenues throughout the year (as analytics team sees sales through the system months before Finance does). Using a similar approach to visitation (proportions, but spliced out for big buckets – e.g., entry, camping, other visitor related), we do rolling $ projections. Timing of camping reservation launch always creates challenges as we are constantly having to adjust the proportions to accommodate it. Revenue and visitation projections have been used to project revenue replacement needs in 2017 (free admission) and 2020 (pandemic) when Memorandum to Cabinet had to be submitted ahead of the season for funds.
- Less formal “predictive”, and more looking forward: the analytics team will often be asked to provide advice on various issues and what it means for operations, programs and services. It is about understanding the social trends (population, travel, sentiment etc.) and working it through PCA operations
What has made the most difference in enhancing your park organization from a data analytics program perspective?
AB
PC
AB
- Committing to being a data driven organization. This is a work in progress – but intentionally curating and managing information in an accessible way to decision makers is a goal we are pursing with more vigour as we stand up our own program
- Moving (almost) all e-commerce and bookings into a single platform when new reservation system is launched in 2024
- Introducing a day use fee for Kananaskis in 2021
PC
- Standardization: with ~130 places having their own unique needs and wants, common templates and standardization of tools/analytics is necessary to maximize efficiency (delivery time, strategic use of limited capacity resources) and focus more on key analytics / info. Custom or additional ad/hoc can follow, where capacity /ability exists.
- Automation: where some repetitive tasks are done (data cleaning, annual dashboards, reports), automation saves time and resources (and is linked to standardization). We update and apply scripts to automate cleaning of data, automate analytic templates, and maintain dashboard structures over time.
- Market analytics: Parks Canada employs segmentation software to standardize insights and have a common approach. A decade ago, you would have had 130+ separate studies trying to identify who was coming using basic demographics through visitor surveys, with no ability to locate more, understand choice, needs etc. The standardization and common approach have allowed Parks Canada to understand similarities/differences in who we are attracting, where, to what and the gaps to inform other strategies/ approaches; it has also allowed us to see the entrenched tourism patterns in Canada through our data.
- Infographics: no one wants a 20-page, let alone a 50-page report of numbers or lots of text. The analysts may clean and interpret 1000s of lines of data or spend weeks coming up with a model, but you must deliver the analytics in an easy to consume and use way. Infographics provide the nuts and bolts, and they are used more often; the 50 pages of numbers or the 1000 lines of data are available if needed.
What tools and data and from what sources do you collect data? Please share as much detail as possible.
AB
PC
AB
See question 3
PC
There is overlap on this question with what is provided above. So, will detail the tools/sources in detail here.
- Visitation: undertake studies to develop formulas. Formulas will involve some mix of data sources, which may be include traffic counter data, Point Of Sale transactional data, third party data (e.g., tour operators), manual counts, and adjustment factors (e.g., party size, removal of non-visitor related) derived from visitor surveys.
- Visitor surveys: self administered paper or QR code survey involving random intercepts on site between May and October. Administered by students and managed internally. Minimum standards must be met to move to analysis.
- Tracking surveys: public, general population surveys. National sample of between 1000 to 2000 people, weighted by demographics and geography. This work is contracted to a third party.
- Market analytics (entry /camping): transactional data from Point-of-Sale machines and Parks Canada reservation system platforms. Cleaned and managed data is then used with ENVISION / PRIZM system (have inhouse) to undertake market analytics / audience work. Parks Canada had a templated report developed in ENVISION that automates the analysis (shifted analysis from 2 days per report to 1 hour).
- Camping (operational): Parks Canada reservation system platform
- Sales: Parks Canada reservation system platform and Point of Sale system (~300 machines across Canada)
- Product/Service specific: examples
- E-commerce: e.g., third party booking or sales platforms Parks Canada contracts/has space on and on-line Discovery Pass purchases
- E-newsletter: contract with service provider provides stats
- Media reach: public surveys (national, general population post campaign as per Treasury Board Secretariat requirements for $1M+ campaigns), service provider (e.g., Expedia) where agreements are in place.
- Social media/web: Parks Canada accounts
- Call centre: call centre logs /emails
- Trails: local trail counters for use volumes / calibrated
What data specifically is analyzed which later is turned over to marketing to help promote an underperforming product or service?
AB
PC
AB
We are not currently intentionally marketing an underperforming product or service – but that is a future goal as we grow in sophistication
PC
See above
Marketing uses many of the data/analytics above to inform different aspects of their work (what to promote – places, offer; to whom – groups, regions, country, channels) as what performs well or not can vary across the system.
In other organizations, marketing and research/analytics are more closely linked because of revenue dependency (e.g., promote X, see pick up/response in sales, respond; charge X, see response, change price or adjust package deal). The extensive network of places, many products, policies, and distributed earning means it is difficult to draw a straight line ‘dose-response’ relationships between actions and outcomes in our organization or pivot/rapidly mobilize to market.
Aside from marketing, how else does the organization use analytics as it relates to underperforming products or services?
AB
PC
AB
(again, not really focused so much on underperforming but on feedback in general)
- RAP Survey- where users report on experience and areas of improvements – is shared weekly with field managers to get immediate feedback on products and services.
- Use online engagement to modify social and web pages
- Use occupancy info in planning for capital investments, operations, FTEs, third party contracts
- Budget development and revenue forecasting
PC
Depends on what the product/service is. For example:
- Hot spot analysis of campsites – highlights under-used campsites and inquires into why (flooded, not near the showers, not near water, bumpy site) and how to address.
- Market analytics (based on postal codes from gates/camping) provided insights into who was/was not coming, which informed outreach strategies (where we went, what we did), platforms and content, engagement programs (e.g., youth programs), and even the types of partners/partnerships we sought/looked at to extend objectives.
- Social media – views of Parks Canada videos, or reads of content in e-newsletter, website content, etc. provides insight into what is engaging people, and what is not. Can help shape content or how content is presented.
How does the organization leverage analytics for top performing products or services? How does that translate to make other products or services successful?
AB
PC
AB
TBD
PC
Similar to above.
Becomes a lesson learned of what works and having the information available to demonstrate. Then requires understanding if it can be applied in different contexts / elsewhere, or to other products.
What is the frequency of reporting on the various components of your data analytics program?
AB
PC
AB
- RAP survey- weekly in season
- KCP sales- weekly
- Performance Metrics on visitation and satisfaction in business plan- annual
PC
Parks Canada operates on an April 1 to March 31 fiscal year (FY), so many corporate and operational indicators must be ready shortly after FY end for reporting in various corporate reports for accountability. Other analytics are when/where required.
- Visitation: received monthly, where possible. Officially released after end of FY (usually May/June), but often get media/other requests through year. A mid-year status update is done for internal-only audience.
- Visitor surveys: conducted during the peak season (May through Oct) and analyzed in the Fall. Business units receive their results in Fall/Winter. Key operational indicators are reported in Parks Canada Agency’s annual report (drafted ~June, after FY end) and in State of Protected Area reports (every five years) and used in state of reports ahead of management plans (depending on schedule).
- Tracking surveys: occur quarterly. Internal-only circulation each quarter (infographic). Results released after FY end internally and on Library and Archives website (required by Treasury Board Secretariat). Results will be reported in PCA annual report (drafted ~June).
- Market analytic reports (entry / camping): depends on seasonal operational data (e.g., visitation and reservations). Entry related data from Point of Sale pulled between Sept 1 and Oct 30 and analyzed; business units receive reports Oct/Nov. Reservation data for season received ~Nov; data is cleaned Nov/Dec and reports/analytics prepared for business units Nov/Dec/Jan. Subsequent analytics (from operational and analytics teams) will occur in Winter usually (summary infographics, occupancy etc.).
- Economics: traditional $ impact prepared every 2 years or so. 2018-19 version results are pre-pandemic and 2020-21 version shows impact of pandemic. Will do 2022-23 this year.
- Sales: prepared weekly during the core operating season and circulated internally. Switches to monthly during off season.
- Product/service specific: tracked through year; final figures for FY end where needed. Otherwise used internally as needed
What’s next in your data analytics program and why are you moving in that direction? Or what data do you wish you had better access to?
AB
PC
AB
Developing the actual program based on the goals set out in 2.
PC
Parks Canada is moving towards integrating its digital and information channels to provide a seamless visitor experience and to integrate data in the background to have better picture of who is coming/using products/services, frequency, and engagement to meet needs (e.g., alerts, promotions, information). This is also being influenced by Government of Canada’s Policy on Digital and Service that is requiring digital services from end to end with measurement and reporting. The process will allow us to create new data collection programs (e.g., camper survey through reservation system), transition other programs (e.g., move from on-site visitor survey to digital one via digital pass/reservable entry database for entry, and use origin info from this integration for market analytics without collecting at gates).
Pressure on analytics right now is to measure volume of use of every attraction, service offer (trail, road, parking lots) to better understand how to manage volumes / move / control traffic/use.
Wish list – resources and costs are going to determine this in a national system.
- Visitor spend data – for a national network of ~130 places in every province and territory and visitation from some ~130+ countries, as we are losing a key data source from Statistics Canada
- Itinerary data – how visitors move across the network in a year or over time, via pass use (tracking on where used, how often, and what types of visitors).
- GIS capacity to map spatial market data to visualize patterns and do more strategic equity analysis
- Visitor data on a more frequent basis from more of the system (e.g., first, repeat, party size, composition, where stayed)
- Recreation data – proportion of people that undertake different recreational activities (hike, fish) to splice out visitation for other uses.
- Direct use data to determine value of use of places that are less tourist-oriented (e.g., urban parks)
- Public sentiment in regions about role of tourism / PCA advertising
- Commercial tour operators: earnings, where do they go, special rates?
- Means to analyze travel reviews in public domain
- Means to get reach and value of earned content marketing (travel content – stories, blogs et al)
What do you find to be the most useful aspect having a data analytics program and what does your senior leadership value as part of your analytics program?
AB
PC
AB
- support effective operational and capital budget requests
- support operational decision making on visitor use and management, programming, user conflict
- third party opportunities
PC
- Having real insights / line of sight on issues
- Having information to inform decisions and position needs
- Having people who understand data and what quality means, how to use data, interpret / analyze data, and can convey the story in data and link it to program needs for operations.
Please share any questions or issues that you feel may relate to this jurisdictional scan:
AB
PC
AB
Alberta Parks is very interested in the outcomes of your research and is in a similar position of developing a data analytics program. Please be in touch with your findings- Nancy.macdonald@gov.ab.ca Tazim.Nazarali@gov.ab.ca
(Alberta Parks is currently focused on hiring 2 data analysts and working with Service Alberta on data visualizations and dashboards, as well as working on structuring data in a more accessible way eg information systems and business processes)
Note to final question below – Alberta Parks does not yet have specialists dedicated to this topic – data management is a subcomponent of the work by a variety of FTEs.
PC
n/a
Enter any link or website that can support your response (optional)
AB
PC
AB
n/a
PC
A lot of analytics are on Parks Canada’s internal intranet site for access by staff. Some external sources:
- Economic impact: available on PCA public website https://parks.canada.ca/agence-agency/bib-lib/rapports-reports/impact-economique-economic-impact/impact-economique-2018-2019-economic-impact
- Visitation: available on Open Gov https://search.open.canada.ca/opendata/?dataset_type=dataset&search_text=parks+canada+attendance
- Tracking and advertising campaign evaluations: available on Library and Archives Canada website https://epe.lac-bac.gc.ca/100/200/301/pwgsc-tpsgc/por-ef/parks_canada/
Upload any other relevant materials that can support your response (optional – e.g. policy paper, samples, etc.)
AB
PC
AB
n/a
PC