Skip to content
PlacementGraph™
Cross border plans

Cross border advertising in Canada and the US: how should a plan disclose uncertainty?

Part of US and Canada targeting

A crowded city street filled with people, signs, and traffic.

The short answer

A plan covering Canada and the US should state four uncertainties in writing: spillover from location settings, shared sites that serve both countries, the accuracy of IP based geography, and currency. Name the evidence behind each estimate, mark what is inferred, set separate budgets and line items per country, and report delivery by matched country with an unknown row.

Key takeaways

  • Location targeting is a best effort: Google says 100% accuracy is not guaranteed, so country reach is an estimate, not a fact.
  • Country popularity data and IP based geography are proxies; MRC guidelines ask that the nature and extent of inference be disclosed.
  • A site read in both countries does not come with a published split, so the split is set by line item targeting, not by the site.
  • Display & Video 360 buys in USD, so a plan written in Canadian dollars carries a conversion that should be dated and stated.
  • Country reports need an unknown or N/A row, and the targeted and matched views answer different questions.

How should a planner disclose uncertainty in cross border reach?

Start from the standard the industry already wrote for location data. The MRC Location-Based Advertising Measurement Guidelines (opens in a new tab), final version 1.0 from March 2017 with IAB and the MMA, say that when data is inferred, "the nature and extent of inference as well as the inference techniques should be disclosed." They also treat accuracy and precision as "independent concepts, but both are necessary disclosures for location measurement." A two country plan applies that logic to each number that says where an audience is.

Definition

Cross border uncertainty statement
A short section of a media plan that names each source of doubt about which country an ad will reach, the evidence used for each estimate, whether that evidence is measured or inferred, and how delivery by country will be reported.
The four uncertainties in a Canada and US plan, and what to write for each
UncertaintyWhere it comes fromEvidence to citeWhat the plan states
SpilloverLocation settings that include interest or regular presence, not only current locationThe platform's location option and its help pageWhich location option each line uses, and that some delivery may fall outside the target country
Shared sitesOne site read on both sides of the borderCountry popularity bands for each country, read separatelyThat no traffic split per site is published, and how line items divide delivery instead
Geo accuracyCountry inferred from IP addresses and device signalsPlatform statements on accuracy and unknown locationsThat country is inferred, plus the expected unknown or N/A remainder
CurrencyBudgets in CAD, buying or billing in USDThe platform's buying currency and a dated exchange rateThe rate, its date and which currency is the commitment
Label each figure by how it is known
Label each figure by how it is knownEach country figure in the plan gets one of these labels, so the reader sees which numbers are measured and which are assumptions. 1. Measured: Delivered impressions by matched country. 2. Inferred: Country from IP or device signals. 3. Proxy: Popularity band, not a count of people. 4. Unknown: N/A or unspecified rows in reportsMORE RESTRICTIVEMeasuredDelivered impressions by matched countryInferredCountry from IP or device signalsProxyPopularity band, not a count of peopleUnknownN/A or unspecified rows in reportsMORE PERMISSIVE
Label each figure by how it is knownEach country figure in the plan gets one of these labels, so the reader sees which numbers are measured and which are assumptions. 1. Measured: Delivered impressions by matched country. 2. Inferred: Country from IP or device signals. 3. Proxy: Popularity band, not a count of people. 4. Unknown: N/A or unspecified rows in reportsMORE RESTRICTIVEMeasuredDelivered impressions by matchedcountryInferredCountry from IP or device signalsProxyPopularity band, not a count ofpeopleUnknownN/A or unspecified rows in reportsMORE PERMISSIVE
Each country figure in the plan gets one of these labels, so the reader sees which numbers are measured and which are assumptions.

What uncertainty remains when geographic traffic is only a proxy?

Two kinds of proxy sit inside a cross border plan. Before launch, a planner judges sites by audience evidence such as the Chrome UX Report (opens in a new tab), whose country dimension is "inferred from users' IP addresses." After launch, platforms place each impression in a country from signals of their own. Google's location targeting page (opens in a new tab) says location "is typically based on the Internet Protocol (IP) address" and that "100% accuracy is not guaranteed in every situation."

The MRC guidelines are blunter about mobile: "The use of mobile IP addressed is generally not accurate or precise for mobile location," while the data "may be more accurate and precise for location of static devices such as desktop PCs, connected TVs or OTT devices." Research points the same way. A 2021 study of two commercial IP geolocation databases (opens in a new tab), benchmarked against more than 2 billion location samples, found that once cost is factored in, IP geolocation may be a better choice than GPS under certain campaign characteristics, "despite its inferior performance." It gives no Canada or US error rate, so a plan should not invent one.

How do shared sites split delivery between countries?

They do not split it by themselves. A national news, recipe or review site read in both countries sells impressions to buyers in both, and each impression carries the country the exchange inferred for it. In the OpenRTB 2.6 specification (opens in a new tab), the device's Geo object carries a country "using ISO-3166-1-alpha-3" and, for IP based locations, an ipservice field naming the provider "used to determine geolocation from IP address." The split happens impression by impression, at bid time.

That has two consequences for the plan. First, a shared site can appear on both country lists without double counting, because a line item targeted to one country bids only on impressions whose inferred location matches it. Second, the share of a shared site's traffic that is Canadian is not something a planner can read from CrUX: ranks are relative bands within each country, not percentages of one site's visits. Write the shared site once per country list, with each country's band beside it.

Handling a site that sits on both country lists

  1. Record both bands. Note the site's popularity band in Canada and in the US, or that data is missing for one of them.
  2. Assign it to both lists. Keep the site on the Canada list and the US list if the brief fits both audiences.
  3. Separate the line items. Run each list on its own line item with one country targeted, so delivery divides by inferred location.
  4. Watch the overlap. Compare each line's delivery on the shared site in the first reporting period and note it in the plan.
  5. Never assign a site's whole budget to its home country. A US based site with Canadian readers still needs a Canadian line to reach them.

How should budgets be split across the border?

No published ratio settles it; the split follows the client's goals in each country and should be written as a decision, not derived from a site's traffic. What the plan can make exact is the currency. Display & Video 360's billing page (opens in a new tab) says "All buying in Display & Video 360 happens in USD," and that conversion rates for reporting or invoicing come from "a commonly used bank feed of daily exchange rates." In OpenRTB 2.6, bidfloorcur defaults to "USD" when a request does not state another currency.

The Bank of Canada's annual average exchange rates (opens in a new tab), "expressed as 1 unit of the foreign currency converted into Canadian dollars," show the US dollar at 1.3698 for 2024 and 1.3978 for 2025. A Canadian budget fixed in CAD buys fewer US dollars of media when the rate rises. State which currency the client committed in, the rate used and its date.

Budget lines for a two country plan
Budget lineCurrency to stateWhat to write beside it
Canada line itemsCAD when the budget holder commits in Canadian dollarsRate and date used to convert into the buying currency
US line itemsUSDNo conversion, or the CAD equivalent with rate and date if the client reports in CAD
Shared site dealsThe deal's stated floor currencyWhether the floor was quoted in CAD or left to default to USD
Reserve for rebalancingSame as the commitmentThe rule for moving money between countries after the first report

How do you report delivery by country?

Report what was delivered where, and keep the unknown remainder visible. Google Ads offers two views in its geographic performance reports (opens in a new tab): "Targeted locations" shows performance by the locations you targeted, and "Matched locations" shows it "based on the locations where your ads appeared," which "could be users' physical locations or locations of interest." The same page lists "Unspecified" areas when location cannot be determined and warns that metrics below country level "may not add up to campaign totals."

Display & Video 360's geography targeting page (opens in a new tab) says a report showing "N/A" means it "could not determine a particular geographic dimension based on the user's IP address," and that its reports use data "received in the bid request at bid time," so other platforms measuring at serving time can differ. A country report built from two systems therefore needs a note on which system each number came from.

Country delivery report for a Canada and US plan

  • Delivery is shown by matched country, with Canada and the US on separate rows.
  • An unknown, unspecified or N/A row is shown, never folded into either country.
  • Each number names its source system and whether it was measured at bid time or serving time.
  • Spend appears in the buying currency and in the client's commitment currency, with the conversion date.
  • Shared sites show delivery per country line item.
  • The location option used by each line item is restated, so readers can judge spillover.
  • Any change to the budget split since launch is recorded with its reason.

What does an uncertainty statement look like on a real brief?

Fictional example

A Burlington chocolate maker's two country holiday plan

An unnamed small batch chocolate maker in Burlington, Vermont sells online to the US Northeast and ships across Canada. Its agency builds a holiday display plan with a US list and an English Canada list. Several national food and recipe sites appear on both.

The plan's uncertainty section reads: country reach is inferred from IP based signals and is not guaranteed; the Canada lines use a presence only location option; popularity bands are relative ranks among Chrome visits, not audience counts; shared food sites sit on both lists and are split by line item targeting; the Canadian budget is committed in CAD and converted to USD for buying at a stated rate and date. The first report shows matched country, an unknown row and spend in both currencies.

How does PlacementGraph record audience evidence for a two country plan?

PlacementGraph shows audience evidence for each site.

Frequently asked questions

Does Google's Presence option remove cross border spillover?

It narrows it. Google's location help says its default option uses physical location and location of interest, and that location of interest "isn't restricted to a person's country." Choosing presence only removes the interest route, but Google still states that location targeting is a best effort and that 100% accuracy is not guaranteed.

Can a Canadian visitor on a VPN be counted as American?

Yes, it can happen. The MRC guidelines say IP based location is "subject to VPN challenges such as VPN spoofing or corporate IPs," and that IP addresses "may also be based on an application server, which can vary significantly from the actual location of a device or user." No platform publishes a rate for this in Canada or the US.

Why do country totals differ between the buying platform and the ad server?

They measure at different moments and with different methods. Display & Video 360 reports location from the bid request at bid time, while Campaign Manager 360 reports on data received at serving time, and Google says comparisons with third party platforms "will likely lead to discrepancies." Pick one system as the record for country delivery and say which.

Can related areas pull a US campaign into Canada?

Google says that when you target an area it may also show ads in "nearby, closely related areas" that could not be targeted on their own because of low population or limited data. For a campaign near the border, check the matched locations report early and record any delivery outside the targeted country.

Sources

  1. MRC Location-Based Advertising Measurement Guidelines, Final Version 1.0 (opens in a new tab)Media Rating Council with IAB and MMA, March 2017
  2. About targeting geographic locations (opens in a new tab)Google Ads Help
  3. About measuring geographic performance (opens in a new tab)Google Ads Help
  4. Geography targeting (opens in a new tab)Display & Video 360 Help
  5. Billing currencies in invoices (opens in a new tab)Display & Video 360 Help
  6. Annual exchange rates (opens in a new tab)Bank of Canada
  7. OpenRTB 2.6 specification (opens in a new tab)IAB Tech Lab
  8. CrUX dimensions (opens in a new tab)Chrome for Developers
  9. A deep dive into the accuracy of IP Geolocation Databases and its impact on online advertising (opens in a new tab)Callejo, Gramaglia, Cuevas and Cuevas, arXiv, 2021, revised 2022

Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.

HOW AGENCIES WORK WITH US

From your first brief.
To your next campaign.

Share a brief in plain language. Get contextual websites and YouTube channels that fit it, with the evidence behind every pick and a deal ready for your usual buying platform.

  1. 01

    Share your brief

    The campaign in plain language.

  2. 02

    Get your shortlist

    The strongest contextual sites and channels, with the evidence behind each.

  3. 03

    We build the deal

    Your shortlist becomes a curated deal through our exchange partners.

  4. 04

    Activate and grow

    Buy in your usual platform. Agencies earn benefits as their investment grows.

LET’S WORK TOGETHER

Your next campaign starts here.

Your details will only be used to respond to your enquiry.

Prefer email? diana@placementgraph.com

YOUR PRIVACY

You choose whether to allow optional analytics. You can use the whole website either way.

Change or withdraw consent at any time using “Cookie preferences” in the footer. Turning analytics off stops future collection and removes this site’s analytics cookies; it does not erase information already sent. The page may refresh to apply withdrawal.