In my last installment, I looked at the concept of Web site profitability, and laid out the minimum financial requirements for Dawn's pay site, along with the monthly membership volume needed to sustain it. I also took a glance at whether trial memberships were profitable or not. Let's continue now with traffic requirements:
The More the Merrier
Ok, so we need to maintain an average of 105 members a month. For the moment, lets (perhaps unwisely) forget about retention rates. Although Dawn Elizabeth revealed that with her previous pay site, 3-6 months was the norm, I'm going to look at a "worse case" scenario that involves a one month retention, and base these figures on that. Of course I realize that doing so may lead me with results 3 to 6 times more conservative than necessary, I'll be very happy to see a huge "bonus" in my income, rather than a shortfall due to unrealistic expectations.
Given this, let's look at how many "1 Month" memberships I need to sell on a daily basis. With a rough base of 105 members /mo. to shoot for, this gives me an average of 3.5 members a day. Since it would be messy to saw our members in half, and I'd rather make more money than less, let's call it 4 members each day, for a new monthly average of 120 members. Still quite a very reasonable amount.
But wait, let's shake things up a bit. Dawn Elizabeth currently offers three sign up options: trial to recurring credit card, recurring check, and one time phone (Web900) charge. I have no relevant information to assign percentages to the frequency of use of these methods of payment, which will play an important role in these equations, and they are an area that I will chart over time. This leads to credit card use as the only "trial" option, which further complicates things due to the drop-off of members who cancel their trials. I have also heard figures that range from 20% to 80% of trial memberships recurring into full monthly ones — a range that seems uncomfortably vague to me.
So let's add an additional membership (for a total of "5" per day) and change the requirement to being one of "5 new signups a day." This allows two of these signups to cancel before the end of their trial without affecting our calculations too badly, and the loss should be mitigated by those members who stay longer than one month. As I said before, I'm simply trying to develop models that will be fine tuned as historical data is gathered, allowing for more accurate future predictions.
Taking Time for Traffic
Now that we know the approximate number of monthly members we need to maintain, and how many new signups are required daily, we need to ascertain how much traffic will be required to attain this volume, and how much that traffic will cost, based on our projected conversion ratios. Here's a simple formula that will begin to do just that:
Required Traffic Volume = Target Membership Level X Inverse of Conversion Ratio
Using our reasonable volume of 5 signups a day to maintain an average monthly membership base of at least 100 members, we can easily project our daily traffic requirements in "uniques" (UNQ) based on our Average Conversion Ratio (CR) as charted below:
CR UNQ CR UNQ
1:100 500 1:600 3,000
1:200 1,000 1:700 3,500
1:300 1,500 1:800 4,000
1:400 2,000 1:900 4,500
1:500 2,500 1:1000 5,000
Additionally, all of the more traditional adult traffic sources would be tested and made use of.
If we pick a middle of the road ratio, say 1:500, then by sending an average of 2,500 targeted 'uniques' into the site we should be able to make our daily signup quota. Until we assemble real world data on our average conversion ratios, we should not be overly optimistic about the results, although I believe that by using the right mix of traffic and fine-tuning the tour, we should get conversions closer to 1:300, or even better.
So where will this traffic come from, and what will it cost? One promising option is to purchase it outright through Pay Per Click (PPC) Search Engines. My preliminary investigation has shown that we would be able to secure highly targeted, first-page listings at a variety of these sites for a penny a click. Working that down shows that at a 1:500 conversion ratio, one sale would pay for the other four, which still leaves me in a very profitable position. Also, I would expect to convert much higher than 1:500 using carefully constructed Web pages to receive the traffic from our carefully targeted key phrases.
Additionally, all of the more traditional adult traffic sources would be tested and made use of, including free and AVS feeder sites, link lists, top lists, pic posts, and TGP / TGP2 — although most of these traffic sources would need to be carefully washed and filtered, so as to adversely affect our conversion ratios no more than is absolutely necessary.
As you can see, many of my considerations are still "ballpark" estimates, and I'm sure that I'll have a few more as time goes on. I will be really interested to see how this all plays out as we start to assemble a base of historical data, and implement the ideas that I have remaining in the background. As always, you'll be the first to know! ~ Stephen