Jim answers questions from fellow Drillers
Topic Overview
Hi again folks, Jim Novo here.
How do we make money on a content site? Free? Pay? Some combination of both? There’s been a lot of guessing and testing by the big media guys on how to work this, but how do the the small segments / little guys make this work, what can be measured to help? What about newer platforms like Substack, how do you measure optimization? On with the Drillin’ …
Jim’s Note: If you don’t know what Recency and Frequency are they are explained here, and RFM is covered here. “Intensity” is Views per Session, in this case a “proxy” for visitor value.
Q: Hi Jim,
Should we use:
RFI – Recency, Frequency, Intensity
RFM – Recency, Frequency, and Monetary
or
RF – Recency, Frequency
to measure visitor value, and what should these terms ideally mean? Total Sessions, Total page views, etc. Also, when you measure Frequency, do you only include the Frequency during a specific period of time (i.e. one month, or one week), or do you include total lifetime activity per user?
A: On the advertising side of the business, I think the page views/session stat is probably the best to use. The reality of the ad-based business is it doesn’t matter if they come back, you are selling impressions, not people. I don’t think you have to overcomplicate it with formulas like RF or RFM, because you are primarily dealing with audiences, not individuals. RF and RFM are about predicting if individuals will come back.
For the ad biz, you want to know which visitor sources generate the highest page views per session. But for the subscription business you want to know what parts of the site create the most loyal visitors – those who will pay for content. These are two different issues. More on this below. If you decide to track pure Frequency, yes, it can be a Lifetime activity or a more current timeframe, but I don’t think you really need Lifetime for the ad business.
Q: It’s interesting. There seems to be 2 major “customers” for your systems at my company: Marketing and Product groups. What are the most powerful deliverables, in your experience, that we can give to each?
A: I am guessing Marketing has to do with getting people to come to the site and Product is the people developing the site content.
For Marketing, the most important thing you can do is to track source, because none of these other metrics will do you any good if you can’t actually do something with them. For any given metric, ask yourself this: If I knew this metric, what would I change to improve the business? If you don’t know “source” of the customer (which ads, search engines, etc. are the visitors coming from?) then you can’t change your marketing strategy to attract the visitors with highest page views/session.
You want to know “original source,” as in when you set the first tracking element: what was the referring URL or Ad?
Visitor source is very powerful, because it is largely responsible for the long-term quality of the visitor, and it is something you can often control or influence.
You don’t have to report on source at the individual level, report on it at a meaningful level that you can use to make changes – again, if you can’t use the info to do something then it is not worth tracking. So for ads where you are paying money for visitors, you would track by ad or by network or whatever control point you have. For search, you track at the engine level – Google, MSN, Yahoo, etc. When you find the average visitor from Google does 8 pages/session, and the average visitor form MSN does 2 pages/session, now you have information that is actionable that can be used to optimize campaigns and search.
For Product, I would also track Recency in some aggregate form, say “30 day Recency” – what percent of visitors have visited at least once in the past 30 days? You will find that your different “products” have different levels of 30 day Recency, and the ones with the highest percent Recency are the ones where your customers are most loyal – this is the true definition of “stickiness,” if you think about it. This information will help guide product design – which products are the most satisfying to customers? The above approach is what I call Hurdle Rates – it’s in the book or you can read some about it online here.
And, by measuring it this way, you really have a yes/no type tracking – either they came back within 30 days or didn’t, you can use a simple “switch” that will be easy to implement on the IT side, rather than tracking the actual Recency by number of days. You have to decide how you handle visitors using multiple products – they could be 30 day Recent on weather but not on news, for example.
You can also use source for Product. Even though the average 30 day Recency for a Product may be 30%, you find that for Visitors from Google have a 60% 30 day Recency and visitors from MSN have a 10% 30 day Recency – for the same particular product. Knowing what you know about these different audiences, which one is more desirable, and what does this mean about the product?
This intersection of “visitor quality” with “product quality” creates a grid you should be able to use to tweak both ad and site effectiveness. It will show you where the very highest and very lowest quality combinations are. The highest quality visitors to the highest Recency area of the site are the most likely to pay for the content. You should be able to optimize both ad revenue and subscriptions using this matrix.
Q: Also, given the fact that we will most likely use a “phased” implementation, what do you think are the logical steps to take in phase I, II, III, etc?
A: I’d think about t like this:
1. Get your source tracking down.
2. Figure out how you will aggregate users – what will be the most meaningful way to understand the source data, how will you use source, where are the control points? Don’t bother aggregating unless you can do something by knowing the result.
3. Track page views/session, and aggregate by source for reports.
4. Track 30 day Recency as a measure of the visitor loyalty to any particular product.
With just these 3 metrics – source, views/session, and 30 day Recency across each product, you should have enough intelligence to keep you busy for several years figuring out how to optimize the business.
The next step would be to figure out who converts to paid subscribers – where do they come from, and what products do they use and stick with? Then you can refine your products and advertising to attract visitors most likely to convert to paid. I just did a study on this for another site, and found of the 38 ads they were running, a lot of them got high click-through to the site, but only 2 ads resulted in people becoming paid subscribers – and those two ads had the lowest initial click-through rates – a concept / result very common in direct marketing.
Jim
Get the book at Booklocker.com
Find Out Specifically What is in the Book
Learn Customer Marketing Concepts and Metrics (site article list)
Download the first 9 chapters of the Drilling Down book: PDF