We've been reporting the monthly fraction of clients' ad spend on each major PPC engine for over a year now. See, for example, Feb 2007; July 2007; Nov 2007, etc. We plan on continuing these reports in 2008.
So, here are our January 2008 PPC search engine share numbers:
|Engine||Jan 2007 share||Jan 2008 share||YOY Change|
That "78.7%" figure means that for each dollar our agency spent on paid clicks for our clients last month, 78.7 cents went to Google. Reporting share in this way removes month-month seasonal factors as well as our own firm's growth serving new clients.
Here is PPC ad spend share graphed monthly.
With Microsoft's recent offer to buy Yahoo, we thought we'd dig into these data again in light of that potential acquisition.
Observation #1: Google is beyond dominant.
Most folks outside the industry do not realize just how incredibly dominant Google is in online advertising today.
Yes, Yahoo and Microsoft are in second and third place. But at nearly 80% share, Google is so far ahead, that 2nd and 3rd, alone or joined up, doesn't matter all that much in the grand scheme of things.
That's not to say there aren't great clicks to be bought on Yahoo and Microsoft. There are. That's not to say ignore Yahoo and Microsoft. We don't, and other smart advertisers don't.
But at 80%, Google is four times larger than the other engines combined. Four times! That's way beyond dominant.
Observation #2: Google steadily seized share from Yahoo during 2007.
For our clients at least, Google clicks improved in quality relative to Yahoo clicks during 2007. As a result, our technology and our analysts moved spend from Yahoo to Google.
We never sat down and had meetings about moving budgets. Rather, our systems noticed that ever-so-slightly better clicks could be had on Google and so shifted spend there. "Better" in this context means "more likely to generate sales dollars or margin dollars for our clients".
This shift from Yahoo to Google was imperceptible on the day-to-day scale. The search wars are fought one click at a time. One bid at a time. A penny here and a penny there. But looking back over '07, the trend becomes clear: Google won 5 points of share at Yahoo's expense.
Observation #3: Microsoft has good clicks, but no inventory.
Here's a plot of conversion rates by engine over the last two quarters.
"Conversion rate" is the likelihood a paid click generates an order on the client's site. The formula is simple: tracked orders divided by corresponding tracked clicks.
The absolute scale on that chart doesn't matter so much. Our agency-wide conversion rate would go higher if we had more clients selling low-ticket goods, and it would go lower if we had more clients selling high-ticket goods. What does matter is the relative position and trend of the lines.
The conversion rate chart shows that Microsoft clicks convert far better -- in fact 10% better -- than Google clicks.
On first glance, that observation seems to cast a different light on Google's dominance in the PPC search ecosystem. In truth, it doesn't.
Observation #4: Google's winning $150B+ recipe is click volume plus click quality.
You can't evaluate click quality without considering click volume. Scale matters.
In January, our firm's clients bought 13.6 clicks from Google for every 1 click they bought from Microsoft.
For an apples-to-apples comparison, had we only purchased the most effective top 7% of the click inventory on the Google side (eg the highest converting 1/13.6th of the traffic), I'd wager Google's conversion from those top super-clicks would soar north of 15%. Just a guess, and I'll try to pull some actual data on that later.
Because of the volume differences, comparing MSFT click quality to GOOG click quality is like comparing a single-location local coffee bar to Starbucks. Sure, the small store has better atmosphere and coffee. But you have to take into account Starbucks' behemoth scale. Starbucks does a passable job everywhere, in 15,000 stores. I doubt the local store could scale and maintain any level of quality.
Observation 5: Microsoft has good monetization on small volume.
On low volume, higher conversion means Microsoft has good monetization of their SERPs, garnering strong CPCs.
The increasing trend across the fall reflects the holiday advertising frenzy.
Higher CPCs means Microsoft commands strong effective CPMs.
Someone might argue that Microsoft's higher click quality suggests Microsoft has built a superior algorithm, a better mousetrap as it were.
If that were true, and if Microsoft could move their conversion rates onto Yahoo's traffic, the post-merger "Mahoo" would be formidable Google competitor.
I don't think that is the case.
Microsoft's small share in the industry is strong evidence that their quality doesn't scale.
Every advertiser of any scale has tested paid search on Microsoft. If a deep well of high-quality clicks existed there, our robots and our competitors' robots and the tens of thousands of SMBs running search by hand would have stampeded to that watering hole, ramping up their their spend on Microsoft during 2007.
That didn't happen.
Significant advertisers have tested Microsoft, bought all the decent relevant click inventory they could there, and quickly turned their attention back to Google.
I really wish Microsoft and Yahoo all the best. Together or apart, I'm rooting for them to succeed.
Our industry would be strengthened if Google has stronger competition, and it would be great if the merger yielded increased transparency.
But I don't think the Microsoft merger poses significant risk to Google. I think Google's biggest threats are internal.
And as usual, we'll report February 2008 PPC ad spend share numbers in early March. Stay tuned: 2008 promises to be an interesting year!
Our firm manages search for 100+ online retail sites, most from the IR 500. Some sites are small PPC advertisers. Several are among the web's largest advertisers. Our client mix is about 85% B2C, 15% B2B. We don't buy media with any a priori budget for each engine; rather, our portfolio optimization technology as many good clicks as it can. We manage search from a direct marketing perspective. That is, we buy as many clicks as we can that yield sufficient sales to meet or beat our client's efficiency targets.
Accordingly, changes in how our agency spends money across the engines reflects changes in click quality, typically measured by conversions or sales per visit.
If there is a bias to our data, that bias is that RKG primarily serves direct-response retailers, and so our client roster doesn't reflect the advertisers who buy search for branding.