Menu

Programmatic Optimization: Why Placement IDs Matter More Than You Think

Programmatic campaign optimization through placement IDs and format-level analysis

What is the single most important dimension for optimizing a programmatic advertising campaign? Audience? Format? Publisher? Placement ID? Well… it depends on your KPI. And the answer changes dramatically between lower-funnel and upper-funnel objectives.

Key takeaway: For lower-funnel campaigns focused on sales or leads, audience relevance is often decisive. For upper-funnel campaigns focused on traffic or qualified impressions, format × placement can be an especially powerful optimization dimension. The trick is to identify where your ads actually appear, not just which website you buy.

1. Lower funnel vs upper funnel: the KPI changes everything

When a campaign is designed to generate sales or leads after a click, the audience is often the most important lever. Asking someone who has never visited an advertiser's website to click an ad and immediately make a purchase? Good luck with that. 😉

For these campaigns, it can make sense to prioritize people who have already visited the site or interacted with the brand, then reach them across relevant environments and formats. Retargeting is one option (provided there is an appropriate legal basis, consent where required, and enough eligible users).

For upper-funnel campaigns designed to generate traffic or efficiently delivered, qualified impressions, the picture is different. Here, I would pay particularly close attention to the combination of advertising format and individual placement.

Campaign objectiveDimension to prioritizeWhy it matters
Lower funnel: leads, sales, post-click actionsAudienceExisting intent, prior interactions and relevance can strongly affect conversion probability.
Upper funnel: traffic, qualified impressionsFormat × placementCreative presentation and the precise ad slot can materially affect engagement and exposure.

These are strategic priorities, not universal rules. Creative quality, frequency, landing pages, measurement and attribution also influence outcomes.

2. Why the advertising format can change everything

A native ad and a standard IAB banner do not necessarily generate the same response. In my experience, the difference in observed cost per click can reach 10× to 30× between formats in some buying situations (an illustrative practitioner observation, not a general market benchmark; placement mix, objectives and click quality matter).

But comparing formats alone is not enough. Two placements using the same 300×250 creative can behave very differently depending on their position, how much of the screen they occupy and whether they remain visible while someone scrolls.

That's why I previously argued that viewability alone is not a sufficient measure of advertising effectiveness. A placement can technically satisfy a visibility threshold without delivering much attention.

3. What exactly is a placement ID in programmatic?

A placement is the specific advertising slot or integration being purchased on a publisher's property. It reflects part of the publisher's tagging and monetization setup. In some systems, it is represented by a placement ID, ad-unit identifier or another platform-specific field.

A descriptive placement label might look like this:

site.com_SSPseat_device_sitepage_format

This is an illustrative naming convention, not a universal technical standard. Actual identifiers and available dimensions vary by publisher, SSP and DSP.

Why does this matter? Because a handful of well-integrated placements may account for a disproportionate share of campaign performance. Think of a well-positioned slot, a format with meaningful screen share, or a carefully implemented sticky unit (provided it respects user experience and quality requirements).

The trader's job is to find those placements, assess their incremental value and direct more budget towards them when the data supports it.

4. The catch: placement-level reporting is not always easy to access

Many traders can readily report on domains, devices, exchanges and formats, yet struggle to obtain a consistent, actionable publisher placement identifier directly from their DSP interface.

In my original analysis, I highlighted Xandr as a platform offering useful placement-level visibility, while reporting for DV360, The Trade Desk and Amazon DSP could require API access or additional technical work (capabilities depend on the platform, permissions, integration and current reporting features).

This creates a blind spot: two ad slots on the same publisher may be grouped together even though their performance differs dramatically. The problem resembles the information loss discussed in our article on SSP throttling and bid-request filtering.

5. A practical workaround: recreate a “proxy placement” in Excel

When the true placement ID is unavailable, Media Traders can construct a proxy placement by combining dimensions that their reporting tools do expose:

Proxy placement = SSP + Seller + Site + Device + Media type + Format

Export a report at the most granular level available, normalize naming conventions and concatenate these fields in Excel or a BI tool. Then aggregate spend, impressions, clicks and the relevant quality or outcome KPIs by this composite key.

FieldIllustrative valueWhat it helps distinguish
SSPEquativSupply platform
Seller / seatPublisher seat ACommercial supply path
Siteexample.frPublisher domain
DeviceMobileScreen environment
Media typeDisplayCreative category
Format300×250Creative dimensions

For example, an Excel formula could be =TEXTJOIN(" | ",TRUE,A2:F2) if these six fields occupy columns A to F.

Important: this composite key is not a real placement ID. Several distinct slots may share exactly the same SSP, seller, domain, device, media type and format. It is an approximation that can uncover patterns, not a guarantee of ad-slot-level precision.

6. From reporting to better buying decisions

Once you have grouped performance by proxy placement, compare results against your campaign objective. Separate meaningful differences from noise: a tiny sample of impressions is not enough to declare a placement a winner.

Next, test whether the DSP actually lets you activate or exclude the relevant combination of supply-path and format dimensions. Reporting granularity and targeting granularity are not always identical. Where necessary, work with the SSP or curator to build a more precise deal.

Finally, validate that better-looking KPIs translate into genuine campaign value. Our articles on challenging DSP optimization and incremental targeting explore two complementary ways to avoid paying for apparent rather than real efficiency.

7. GreyFox Ads: making the French Open Web easier to activate

At GreyFox Ads, we believe the quality French Open Web can deliver real efficiency for advertisers. We help Media Traders outside France reach French audiences with local market knowledge, clean supply, smart targeting and transparent costs, all brought together in a single Master Curated Deal.

Because the most important optimization dimension is not always the one displayed most prominently in your DSP dashboard. Sometimes, it's the one you have to work a little harder to uncover. 🦊

Original LinkedIn post: Read the original post by Max Legrand.

About the author

Max Legrand is the founder of GreyFox Ads, a French independent programmatic curation specialist focused on efficient, simple and transparent access to quality Open Web inventory.

Want to improve your programmatic campaigns in France?

Looking for quality French Open Web inventory, smarter targeting and a simpler way to activate your media plan? Let's talk about your campaign.

Contact GreyFox AdsDiscover our Master Deal