My last Plusbox performance post covered about 6 month’s of data, and performance details about a fairly young Google product. Now, with nearly a year of data under our belt, it’s time for an update!

First a note about product features that I didn’t cover before, and Google kindly alerted me to. (Thanks!) Google Base Feeds (Now, referred to as Google Merchant Center Feeds) were developed to support Google Product Search. As such, most Alpha level features of the Plusbox product looked to utilize the existing data, and not much else. As the product evolved, Google allowed for additional, AdWords Plusbox specific columns to help the advertiser use one Google feed, to support two distinct Google products. The most notable ones are:

1. adwords_queryparam: Google will append “kw={keyword}” to the referring URL, for on-site analytics to handle. (Google example: If the destination URL in the offer is http://example.com/offer26?q=5 and the ad’s keyword is ‘digital cameras’, a user who clicks on the offer from a Product Plusbox will visit the URL
http://example.com/offer26?q=5&kw=digital+cameras).
2. adwords_prefer_for_query: Gives the advertiser the ability to specify key, user queries, and designate particular items as preferred for that query. Note that you can submit more than one keyword for each item. Also, you can use the same preferred query for multiple items in the feed. (Google example: If ‘red luggage’ is set as a preferred query, The item with this attribute will be preferred when the user’s query is ‘red luggage’ but not if the user’s query is ‘big red luggage’ or ‘luggage’).
3. adwords_redirect: Gives the advertiser the ability to specify a unique URL for AdWords Plusbox use only. The URL listed here will not be used in Google Product Search and will only be used in Plusbox ads. This allows advertisers to independently track each Google service utilizing this feed.

In my first and second posts on the topic, Plusbox ads had a fairly defined feel, with 3 product offering, text, and so forth. We’ve seen a number of new formats that they are testing.

Up to 6 products listed on top rail. Horizontal layout. Pricing. No item descriptions.

Up to 4 products listed on right rail. Horizontal & Vertical layout. Pricing. Titles present, but harder to read.

Finally, I found a few points in the data that are worth noting. First, it’s interesting to see the interaction between Google’s Plusbox serving frequency and overall user Plusbox engagement (expansions - that is to say, clicked on the Plus symbol of the ad, to see Plusbox product offering). Here I’ve outlined Plusbox ad impression volumes by week, compared to the same ad’s Plusbox expansion rate.

The product was fairly new in late 2008 / early 2009, but user expansion rates were consistently higher through much of 2009, until roughly July. Since then, we’ve seen much lower Plusbox expansion rates. Granted, expansion rates dropped from 0.7% to 0.2%, so the overall effect to the program is small. My hunch is that users were eager to test the new ad formats through the early quarter of 2009. Over time, they have grown weary of the product results displayed within, and have been less likely to reengage.

Second, it’s important to see if we are still seeing overall ad click-through-rates (CTRs) to be higher with the presence of the Plusbox ad, than without. (Regardless of any Plusbox icon expansion) Yes, this finding still holds true!

This still seems to be the most beneficial component of the current Plusbox product. There is some recent evidence that leads me to believe that the product is getting better. This could be a result of Google showing more appropriate products, testing the different layouts, or advertisers using the increased functionality to improve the results. User engagement with the Plusbox icon is still very limited, and I expect it to remain so for some time.