|
A lot has been written about domains and how you can buy, sell, develop them or even treat them as stock items. I plan on taking a different approach and dealing with domains as an unique asset class that is investable by the professional investment community just like any other class of asset such as stocks, land etc.
If domains can be legitimised as an asset class then it's much more likely that external funds will begin to flow into the industry. This potentially means that rather than receiving 2 year multiples we start getting 10-15 years for our assets. Presently, other than opportunistic sales I don't see any individuals or organisations that are beginning to work on this problem and move more out of the revenue to the capital value game.
For domains to become legitimate assets for professional investors a number of things need to happen. The first has to do with transparency, the second risk and the third liquidity. Over the next few articles I plan on disassembling these concepts and comment on my own personal journey in defining domains as professionally investible assets.
Transparency
Over the years I've written a lot about transparency and I've come to the conclusion that transactional transparency that relates to why domainers earn what they earn from PPC is not really valued by any of the stakeholders other than the domainer themselves. This means that unless the incumbent stakeholders in the PPC value chain are forced by market forces to become transparent they won't. So rather than complaining, ranting and raving I decided to do something about it with our domain management and optimisation company ParkLogic and work towards a more transparent future.
 The first step towards transparency is data normalisation so that each domain monetisation solution can be compared against each other. At the moment we have a situation where the classic measurement of RPM is useless for making a comparison because each parking company counts their traffic differently.
For example, if a parking company aggressively filters their reported traffic then they will be artificially increasing the RPM by lowering the number of users reported for potentially earning the same revenue.
Likewise the measurement of revenue per day fails to deliver a consistent comparison as traffic often varies in monetisability over time. For example, if I point traffic at parking company A for two weeks and then parking company B for two weeks you can't compare the daily rates because they are at different points in time.
The solution is to send all of the traffic to all of the parking companies in as real time as possible. This is often known as a round robin system, traffic splitting or A/B testing. In A/B testing all of the traffic (completely unfiltered) needs to be sent to either solution to get a statistically significant result to determine a normalised RPM. There are a number of challenges with A/B testing the largely revolve around revenue leakage.
For example, if I test Parking Co A and it generates an RPM of 50 and then test Parking Co B and it has an RPM of 100 then for every piece of traffic I'm sending A I'm losing 50 RPM.
Let's take a look at the maths. for 2000 pieces of traffic (assuming they are all monetisable) Park Co A generates $50 and Park Co B $100. It also means that the domain owner could have potentially earned $200 rather than the $150 due to sampling Park Co A. To be successful this revenue leakage must be less than the additional revenue the domain has earned due to the revenue increase of Park Co B.
The name of the game then is to minimise the traffic that is sent to non-performing solutions while still maintaining a statistically sound sample of who is paying the most. For higher traffic domains sampling can be conducted more frequently compared to low traffic domains. The sweet spot is to accurately determine what is meant by high and low traffic. We have spent the last several years refining this logic at ParkLogic and let me assure you that this is only stage one of the data normalisation and traffic optimisation process.
Once you have these metrics right (and we think that we're pretty close) then the direct comparisons that can be made across the wide variety of monetisation solutions become increasingly transparent.
There is a whole level of sophistication around normalisation that I haven't alluded to but I believe that the fundamentals will encourage you to either contact me or investigate the matter further. A whole other side of traffic normalisation and revenue leakage minimise is true traffic optimisation but that is separate from normalising the data to gain a foothold in the transparent world.
It's important to note that there are also a number of different definitions of transparency that have been bandied around that actually just confuse the issue. Transparency is not trying to tie a parking company up like a pretzle and demand that they provide audit rights to their data. Ultimately transparency needs to relate back to the monetisation source (eg. Google/Yahoo/Direct) not the parking company which is just an intermediary.
Remember that transparency is one of the necessary components to playing the domain capital value game and providing a true exit for domain owners that have been patiently holding onto their portfolios of domains year in year out. The reason why it needs to go back to the source is that is what a professional investor will be assessing their risk against. This will be the subject of a future article.
Providing visibility into ultimate revenue source, consistency of earnings all help investors make decisions on what they will pay for an income producing asset. This is the ultimate game for a domainer where investors pay 15 times earnings because of the transparency in the metrics versus the current much lower multiples.
Trackback(0)
|
Article Archive 
G...
TrackBack URI for this entry
Subscribe to this comment's feed
