Digital marketing in Paris is measured through a consent wall
Enforcement of consent rules here has been unusually active, which means the banner on your site is not decorative and refusal is a real, common choice rather than an edge case. A large share of the people who visit will decline tracking, and everything downstream is affected: audience sizes, retargeting pools, conversion counts, the automated optimisation that platforms perform on your behalf.
Most proposals do not mention this, and a plan that assumes complete data will keep reporting confidently while the numbers drift away from your accounts. Raise it in the first meeting. Ask each candidate what proportion of visitors they typically see consenting in this market, what they do about the rest, and how that changes the first quarter. A digital marketing team that has worked here seriously will have a concrete answer and will not promise you full visibility.
What refusal actually does to your digital marketing numbers
Three things happen at once. Measured conversions fall below real conversions, so every campaign looks worse than it is. Audience lists built from site behaviour shrink, so retargeting and lookalike targeting weaken. And the automated bidding systems receive fewer signals, so they learn more slowly and need more budget to reach stability.
The temptation is to compensate by loosening the consent mechanism. Do not. The regulator has shown it will act on banners designed to make refusal difficult, and the reputational and financial cost of that is larger than the measurement gain. The correct response is to accept partial data and build a reporting approach that survives it, which is a skill rather than a setting.
Platform reported results and business results stop agreeing
Once data is partial, the totals in your advertising accounts and the totals in your own systems diverge, sometimes dramatically. Each side is right about something different, and every review meeting can be spent arguing about which is real. Settle the question before it arises by naming one source of truth, and make it yours: the orders, the signed contracts, the bookings, the revenue in the finance system.
Platform numbers then become directional, useful for deciding where to move budget within a channel and not for judging the channel against the business. Write that hierarchy into the reporting arrangement. It is a small piece of governance that removes most of the friction from the second year of a digital marketing relationship.
How a candidate proves a channel works without perfect data
This is the question that separates the field. The honest methods are old and slightly crude: controlled periods where a channel is switched off, regions or segments held back deliberately, comparisons across stretches where spend changed sharply, and a simple question asked at the point of enquiry about how the customer found you. Modelled conversions have their place, provided everyone understands they are estimates rather than counts.
Ask each candidate which of these they have actually run, on what kind of client, and what the result was. Then ask what they would do in your first quarter to establish a baseline that later work can be judged against. A team that has never established a baseline will always be able to claim credit for whatever happens next.
Server side collection is a digital marketing project rather than a setting
Many agencies will propose moving measurement to a server side arrangement. It can genuinely help with data quality, but it is engineering work with ongoing cost, it does not remove the need for consent, and it needs somebody who will maintain it after the person who built it moves on. Treat it as a project with a scope, an owner and a price, not as a line item inside a monthly fee.
Ask who builds it, who owns the infrastructure, what happens to it if you change agency, and what the recurring cost looks like. If a proposal includes it without a clear answer to the last question, the cost has been hidden somewhere you will find later.
Reporting honesty becomes a contractual habit
Where the data is incomplete, the temptation to present the flattering version is constant, and it compounds. Build the opposite habit deliberately. Agree that every report names what is uncertain, that a month where the team does not know why something moved is reported as such, and that the same figures your finance function sees are the ones discussed.
Fix a rhythm as well: a short written note monthly, a working session each quarter where a decision is actually taken, and one annual review of whether the whole arrangement is still the right one. Ask to see an anonymised report from a current client before you sign. If it contains only good news, you are looking at the output of a process designed to reassure rather than to inform.
Choosing between digital marketing agencies in Paris
Brief three or four firms with the same written information and require the same outputs: named people with contracted hours, a first quarter plan that includes establishing a baseline, a description of how they would measure without complete data, and a reference at a company with a comparable consent profile. Ask what they would refuse to promise.
If the requirement also reaches the site, the product or earned coverage, the neighbouring listings for web design studios in Paris, local web development firms and public relations teams working here are worth reading before you concentrate everything with one supplier. The full directory of digital marketing agencies helps when you are buying across several markets, and you can request proposals from several teams in one step so the replies arrive together and can be compared properly.