The display line of a monthly report usually gets built by duplicating the search line. The layout is already there, the tiles are already positioned, the client already signed off on it last quarter, and you have four more reports to assemble today. So you drop the programmatic export into the same shape: impressions, clicks, CTR, spend, CPC. Then the CTR tile renders at 0.06% next to the search tab’s 4.1%, and you know exactly which number the client is going to pull out on the call. You start writing commentary to defend it before you have even published the report.
That commentary is wasted work. The problem with default display campaign reporting metrics is not that CTR is being explained badly. It is that CTR is in the headline position at all, in a layout designed for a channel where the click is the product. Display and programmatic buy attention and presence. Measuring that with a click rate is like judging an out of home poster by how many people phoned the number on it.
Below are the mistakes that keep showing up in display sections, and what to put there instead.
Why the default display campaign reporting metrics set is a search set
In search, a click is the whole transaction. Someone typed a query with intent, saw your ad, chose it over nine alternatives, and arrived. CTR is a legitimate headline because it measures the thing the campaign was built to cause. We went into this in more detail in search campaign reporting metrics clients actually need.
Display inventory does not work that way. A banner on a news site reaches someone mid-article who has no intent at all. A BVOD pre-roll reaches someone waiting for an episode to start. The click is an accident of interest or, more often than agencies like to admit, an accident of a thumb on a mobile screen. The published benchmarks that circulate for display CTR sit roughly two orders of magnitude below search, and any display CTR benchmark you quote to a client has the same weakness: it reassures them that 0.06% is normal without telling them it bought anything.
The fix is positional, not rhetorical. Decide what the display line item was actually bought to do, and put that metric in the first tile. For an awareness buy that usually means reach against the planned reach, with viewable impressions and frequency beside it. For a mid-funnel retargeting line, it may genuinely be conversions and cost per conversion, in which case click rate still is not the headline. Clicks and CTR go further down, in a table, available to anyone who wants them and no longer presented as a verdict.
If CTR is in tile one, you have told the client that CTR is the scorecard. Everything you say afterwards is you arguing with your own dashboard.
Viewability: the number clients assume is already inside the impression count
Most clients who have never bought programmatic directly assume an impression means a human saw the ad. The industry definition is narrower. Under the MRC standard, a display impression counts as viewable when at least 50% of the ad’s pixels are in view for at least one continuous second, and two seconds for video. Everything else was served, paid for in many buying models, and never had a chance to register.
The mistake is reporting served impressions as the top-line delivery number and keeping viewability in a measurement appendix, or leaving it out entirely because the export has a column name nobody on the team has reconciled yet. The client then builds their mental model of the campaign on a number that is larger than reality, and when they later hear the word viewability from someone else, your previous reports look like they were hiding something.
Put display ad viewability next to delivery, and report three things together: viewable impressions, viewability rate, and the measured rate (what share of impressions could be measured at all). That last one matters because unmeasured inventory is common, and a 78% viewability rate calculated on 60% of impressions is a different claim from the same rate calculated on 95%. Saying so in one line under the tile costs you nothing and buys you credibility for the rest of the section.
One judgment call worth making explicit to the client: viewability is a quality floor, not a performance result. A campaign can hit a strong viewability rate and still have been bought against the wrong audience. Report it as evidence the media was delivered properly, not as evidence it worked.
View-through conversions, and why the client questions them first
View-through conversions are where display sections get defensive, and the usual mistake is to add them quietly into a total. The conversions tile goes from 40 to 310, the client notices, and now you are explaining attribution under pressure in a meeting instead of in writing on your own terms.
Report view-through separately, always. Click-through conversions in one column, view-through in another, never summed into a single “conversions” figure on the display line. Then state the lookback window you are using, because the number is meaningless without it and the client’s own analytics will never match it. A 30 day post-view window on a brand that has heavy retargeting will credit display with a lot of activity that other channels also touched.
The honest framing is that view-through credit is directional. It shows that exposed users converted at some rate, and if you can get the client’s agreement to run it, a comparison between exposed and unexposed groups is worth far more than any single view-through total. Absent that, present the figure as supporting evidence next to the reach and frequency story rather than as display’s receipt.
Frequency and reach carry the story CTR can’t
A display line item has one structural question behind it: did we reach enough of the right people, enough times, without wasting money reaching the same people twenty times? None of that is visible in a click rate.
Suppose a programmatic display line delivered 2.1 million impressions, of which 1.6 million were viewable, against a unique reach of 410,000 people. That is an average frequency of about 5.1, and the commentary writes itself: the plan called for 400,000 in reach and a frequency cap around five, the buy landed on plan, and here is the frequency distribution showing how much of the spend went to users who saw the ad more than ten times. A client can act on that. They can ask for a tighter cap, more inventory, a wider audience. Nobody can act on 0.06%.
Where the export supports it, a frequency distribution table beats an average. Averages hide the tail, and the tail is usually where the argument for next quarter’s budget change lives.
Where a comparison number is fair, and where it isn’t
Clients compare channels whether or not your report invites it, so give them one comparison that holds up. CPM is fair across display, BVOD and PDOOH, as long as you are clear which CPM (served or viewable) and keep the basis consistent month to month. Share of spend against share of plan is fair everywhere. Reach against planned reach is fair.
CTR across channels is not a fair comparison, and neither is cost per click, and cost per conversion between a prospecting display line and a brand search campaign is close to meaningless. If the client’s board deck needs one number per channel, negotiate that number before the campaign starts rather than after the first report. Mismatched comparisons are the same failure we described in why one dashboard fails every channel, and display absorbs the worst of it because it is the channel most often reported in someone else’s template.
Building the display section once
The practical blocker is time. Everyone agrees display deserves its own tile set, and nobody wants to rebuild it per client per month while the search layout sits right there ready to copy.
That is a template problem rather than an analysis problem. In Screebl, a display layout is something you build once in the drag and drop builder (viewable impressions and viewability rate in the KPI row, reach and frequency beside them, a frequency distribution table, click-through and view-through conversions in separate columns, creatives in the slider so the client can see what actually ran), then reuse for every display line item by uploading that month’s export and picking the group-by per row. The search template stays as it is. The display template stops being a copy of it. If the reporting pack is the thing holding your team back from doing this properly, the plans are the place to start.
Two things are worth keeping in mind while you rebuild. First, resist the urge to show everything you now have access to: six well chosen tiles still beat fourteen, as covered in the case for six tiles. Second, write the definitions into the report itself. A client who reads what a viewable impression is in the same place they read the number is far less likely to send the email asking whether display was worth it. Our rundown of the campaign metrics every client report should show is a reasonable source for the wording.
