What is Viewability?
A lot has been written about viewability but set the context for this post we will describe viewability in simplest terms .
Viewability represents how much an Ad unit is seen by the users. This is to esnure that Advertiser pays for the impression that was actually seen by users. To make this statement quantitative , IAB and MRC endorsed the following standard :
Viewable Display Ad Impressions are
Google has a well formatted info-graphic explaining the concept and factors that impact view-ability
How does Viewability effect the CPM?
Well, It depends. No one can you an answer in terms if revenue lift, that would be worth taking action on. There are multiple factors define and result in any impact that might come. Let us look at a few of those factors here:
- Demand Partners :
- AdEx /AdSense will most probably result in an (e)CPM lift but how much isn’t really quantifiable
- DSP/SSP or networks that you use and Supply Structure.
- Your site :
- Page Structure
- Reading Depth
- Demographic Split on Internet connections of your users
- Co-relation between Site Speed and Ad speed.
- Ad Calls Setup
- SRA and MRA
- Use of Lazy Load
- Multiple Client Site Scripts that build latency
- Refresh , if used and triggers for the same.
Ideal Approach would be to run multiple A/B experiments to check the revenue impact after you have improved the Viewability of few units against similar units or on a different section. For example
- Same Unit can be executed under under lazy load and eager load for equal opportunities in a round robin format.
- On the same DSP or SSP, you can check for the change in Coverages and Win Ratios in a can 10% increase in reserve price .
The tests should normalise against reduced ad calls ,(to nullify the implicit gains on ecpm lift).
This may not give you an exact value for the change in ecpms but will give you good measure to decide further resource investment in this direction.
You should also pay attention to the device format, geo and Audience size that is exposed to improved Units. It is very much possible that for a small audience size, or for new users acquired, you may not see any impact at all. This does not mean the experiments were inconclusive , this may just mean that experiments environment should be altered to cover a larger audience across multiple geos .
A continuous effort in this direction will give positive results in medium to longer term . The results should improve the performance for Direct sold campaigns as well., especially for in-view ad Units
Improving sites or Apps User experience , Page speed Score , navigational efficiency , while keeping ad placements inherent to design will also help improve Viewability and CTRs. One of the main reasons for poor view-abilities is that Product design Prototypes rarely consider ad placements during ideation and UX testing phases. Ad slots are introduced as afterthought , there by disrupting both the UX and performance of ad units.
Improved Viewability should, in general , not create negative effect on your KPIs like PLT or campaign deliverable. However, This is not an exact science to help you quantify programmatic ecpms lifts .
As a publisher , you would need to decide where and how to invest your tech resources based on which factors need most correction in your site.