1. Measure The Average Time Lag From Impression To Conversion
This lag indicates how long you need to wait before being able to truly measure the returns of your current program. For example, there could be an average conversion delay of 8 days, meaning that roughly 50% of all sales occur between the day of click and day 8 — so you definitely want to further look into this and find out how much of an impact this post-click lag has on your daily numbers.
2. Move Away From Average Order Value (AOV), Use Customer Lifetime Value (CLV) Instead
You should ask yourself how many times unique users buy a product from your site and how much revenue it is likely to bring over time. More specifically, you want to move away from regular AOV calculation, which is essentially the average value of a transaction (where AOV=Revenue/Number of Sales) and consider the CLV (where CLV=Revenue over time/Count of unique users).
And you want to run this analysis for each individual channel and type of campaign, down to the most granular levels available such as the keyword level in paid search.
Once you have those AOV and CLV numbers put together, you can easily measure how much more revenue can be expected from a user after the first conversion. Let’s say the AOV is $80 and the CLV $110. That means that, on average, each transaction is worth $80 — but, users tend to return and are likely to buy an additional $30 of your products over time.
Multi Channel Attribution
3. Factor In Those Proxy Conversions (Prospects)
Most conversions (such as sales) are preceded by micro-conversions such as users seeing a banner, visiting a product page, subscribing to a newsletter, watching a video about your product, adding an item to the shopping cart, downloading an app, etc.
Analyzing patterns from those proxy metrics to the end conversion is key to understanding how users behave in general. It’s important to determine how to best utilize those proxy metrics to gauge how likely users are to effectively convert in the end.
For instance, if you have noticed that 10% of those users subscribing to your newsletter will eventually buy something on your site in the next 12 months, then you can do the math and estimate how much revenue can be expected from today’s newsletter subscriptions.
4. Analyze The Relationship Between Ad Spend & Revenue For Each Channel
Whether you have a limited or unlimited budget, given a certain efficiency target, you should definitely analyze the relationship between ad spend and revenue for each individual acquisition channel. More specifically, one can use regression analysis in order to establish a correlation such as Revenue=f(Ad Spend) where “f” is a function that can be determined by analyzing historical data.
Similar analysis can be run across all channels, and one thing will always be constant: whatever the online marketing channel, the marginal cost of a conversion is greater than the average cost. What varies though, is the marginal and average cost for each channel, and you want to make sure you go after those cheap marginal conversions first.
For example, looking at the below table, you’d notice that even though your current ROAS in paid search (7.4) is greater than your emailing ROAS (6.6), the incremental ROAS is greater for the emailing campaign — so if you have more budget, you should use it for your emailing campaigns. Those diminishing returns are a reality and need to be taken into consideration, then compared between channels for a more efficient marketing mix.
5. Analyze Cross-Channel Revenue Transfers (Across digital channels)
Most sophisticated online tracking suites offer cross-channel attribution, or at least some basic insights, which has become crucial when optimizing your online marketing mix.
For instance, say paid search is bringing $1M of revenue a month and the numbers are showing that 15% of this revenue originated from a banner campaign. You should adjust both your paid search and banner ROAS goals in order to take into account those assisting impressions/clicks, and potentially decrease your short-term ROAS goal for your short-term banner campaign as you know it is indirectly helping your marketing mix.
6. Multi Channel Attribution across Multi Screen – How User engage with different devices (desktop, laptop, Tablet, mobile) & convert.
7. Multi Channel Attribution Online to Offline Store – User engage in online medium & convert through offline store.
8. Predict Seasonality
A fairly straight-forward way to go about predicting current performance is to leverage your own account history or, maybe even better, Google Trends. Based on recent performance and weekly search volume trends, you can easily translate the revenue you currently track into predicted revenue.
Article Source: http://marketingland.com/6-analytics-tips-to-predict-your-online-marketing-revenue-in-2014-71083?utm_content=buffer6f6e2&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer