Directing Traffic
February 4, 2008 by Geoff Jennings
Statistics are king in businesses such as online job boards, where success or failure is measured by the amount of traffic to a site.However, there are two methods of collecting data used by the online recruitment industry in Australia.
The industry standard is the panel-based methodology used by Nielson/NetRatings. This tracks data from a panel of about 400 consumers. They report the stats to the market on a monthly basis. As data relies on subscription at a fee, several of the smaller job boards, such as JobX and Jobs, Jobs, Jobs do not participate in this form of data collection.
Alternatively, Hitwise Australia collects data directly from ISP networks. Stats from Hitwise are updated on an hourly basis. The data from this method is therefore more raw and “real-time”. It gives the stats-geeks a chance to analyse the comings and goings of consumers, peak times for views etc, providing an excellent opportunity to better understand consumer practices. Hitwise data is not reliant on individual job board subscription, as is the case with the NetRatings data. However, this data excludes any information from the Telstra ISP networks.
First up, let’s all agree on this: before any one method can be considered the industry standard, it should include data from all possible sources. Therefore, both methods currently in use are deficient in this regard. Hitwise, for its exclusion of Telstra data, and NetRatings for its exclusion of minor job board data.
Arguably, the effect on the data purity of these two deficiencies differs. The NetRatings data gives us no indication whatever of the popularity of the minor sites. It is merely indicative of the popularity of the three main players, Seek, CareerOne and MyCareer. However, the Hitwise data, while excluding a whack of info from the Telstra ISP, allows us a snapshot of where all the industry competitors sit in popularity. It would be difficult to contend that a greater portion of consumers of any one site would be more likely to be Telstra ISP customers. Therefore, as the exclusion affects both the big and smaller players, while it alters quantity of data, it probably does not alter the quality of it.Whatever method of data collection one considers, if we have a look at the figures for January, Seek is the undisputed market leader.CareerOne have seen a better- than- average traffic rise than their close competitors MyCareer and have assumed the number two position. If this was due to spending more dollars on advertising then MyCareer can quickly buy back the number two spot in February.
However, when we consider the popularity of Linkme for the month, there is a conflict between the two data. Linkme have dropped off the NetRatings list with a 97% decline. However HitWise shows a rise in their traffic and rankings. This discrepency may be due to the launch of the new site in January. NetRatings is possibly still tracking the old one and Linkme have moved on.Admittedly, there are several other factors to be considered in the question of success and failure of the job boards. These include: number of c.v.’s on their databases, how many job alert emails go out each day and how good are the quality of applications. Volume isn’t always better.
However when looking at popularity alone, based on the above, we need to challenge the idea that NetRatings is the industry standard of popularity rating for the recruitment industry. But, even if we all concur that a new metric is required for the measure of popularity, will we ever be able to agree on how that measure should be derived?
One thing we can agree on is that NetRatings is not the “industry standard” that some people make it out to be. It is a very small cross section as a result of being totally over priced for the service they deliver to small to medium businesses.
If NNR is to become the industry standard then a greater representation from the industry is required.
We have made it clear that we would love to be involved and are happy to make our numbers transparent, however $15,000 a year for a few half arsed reports and a comparison to SEEK is of little use.
If NNR come to the party and offer a pricing model that reflects the amount of traffic or size of the business (or any other fair metric) then I know a number of other job boards including ours who would jump on.
Until then NNR’s only use is to support the ongoing pissing contest over positions 2 and 3 between CareerOne and MyCareer.
Geoff
Whilst Nielsen do have a panel based methodology in their Netview Panel Product the standard methodology for the online recruitment industry in Australia is from the Nielsen Market Intelligence Product. MI is site centric ie cookie based and is relatively accurate.
The metric for ranking in MI is now 28 day daily average UBs rather than monthly UBs.
At SEEK we have a belief that the best measure is one which represents true audience engagement. The best competitive measure at this point that we have identified is the total time metric that Nielsen now publish in MI. This measure represents the total audience time on the site for the month. In other words how many jobseekers ,by how often, by how long. From this measure we can see that approx 7 out of every 10 minutes spent on the top 3 jobs board is spent on SEEK. We find that this message is more meaningful to our customers than 2.8 Million UBs per month.
Re the small guys - Brett you make a good point and NNR should look at their pricing - having said that we have seen a number of new entrants with large marketing spends who have clearly made a call not based on $, to not enter MI. Wouldn’t entering MI provide them and more importantly their customers a clear view on the value they are bringing to the market in a clear and transparent manner using a consistent methodology?
“Statistics are king in businesses such as online job boards, where success or failure is measured by the amount of traffic to a site.”
Surely success should be measured by the ability to connect job advertisers (not recruiters, actual employeers) with good quality applicants. High traffic will help in a pure numbers game, but better processes will help more. The seek.com.au process is not good.
Justin - yes the bigger guys with lots of money who opt not to enter MI do so for different reasons. If you are telling people you compete with Seek, MyCareer and CareerOne then the last thing you want is for the world to see you with your pants down.
Hey Craigvn
I agree that it would be great to get a ‘quality’ metric in the industry.
Problem is that the recruiters and employers don’t provide that data to anyone, least of all a job board, and never will.
Since recruiters are making say $7,000 per placement, and the most online expensive job ad in Australia is currently around the $200 mark, you can see why they’ll probably never be telling us job boards that information!