All products have a demand cycle of their own. Understanding this becomes important when making marketing decisions. Sales reports may give marketing teams plenty of data on seasonality and demand trends, but they may miss out on an important KPI in the online context – the amount of interaction a visitor has with a site before converting.
The way customers interact with a site is very complex. A visitor might search for a product, click on a PPC ad, enter your site, visit your competitor’s site, then come back to compare prices, postpone decision making, and come back after two weeks to complete the purchase.
At Position², we call this the Holy Grail of Web Analytics.
Google Analytics provides two metrics as part of their Ecommerce Reports – the ‘Days to Purchase’ and ‘Visits to Purchase’ reports. These pan-session metrics tell us how many visits have occurred (and days elapsed) since someone’s first visit, until a purchase action took place. You can get a good understanding of your customers’ behavior by analyzing these reports.
A Closer Look
To illustrate, let’s look at one of our clients in the manufacturing sector.
Less than half of the purchases occurred on the same day as the first visit, and almost 38% of them took place after 5 visits or more. This suggests a long sales cycle with multiple interactions with the site before conversion.
While this in itself is good to know, it is not actionable. What a marketer really needs to know is which customers convert on Day 1 and which ones need more handholding and cajoling before they convert.
The Next Step
We segmented the above data and looked at time to convert across different acquisition sources.
A look at ‘Days to Purchase’ data across sources shows that the site average has been influenced by ‘Direct’ visitors. More than 70% of PPC visitors converted with 48 hours, indicating a strong PPC strategy that targeted the right audience with the right communication. Direct visitors usually take the longest time to convert, whether they are existing customers or visitors from offline campaigns.
In contrast, the ‘Visits to Purchase’ data for PPC does not look as advantageous. Almost 44% of purchases ocurred after 3 or more visits. This could imply a not-so-attractive offer, or a Landing Page (LP) that does not work very positively with visitors. The numbers stack up very well, though, for Organic visitors, with 65% of sales occurring over 1 or 2 visits.
We went on and broke down the search visitors by keyword type to see if there was a difference in behavior for brand keyword visitors.
As expected, the brand keyword visitors converted much faster. The next logical step was to segment data by visitor source.
Interestingly, the Non-Brand visitors for both PPC and Organic took a longer time to convert, with one-third of Non-Brand PPC visitors requiring more than 9 visits to convert. One promising solution was to analyze the ‘Top Content’ of this set and design a separate landing page that addressed them better. Ideally, we would test this new LP against the existing one to compare their performance. Another possible recommendation would be to capture the visitor’s email ID in a form and send reminder emailers or enticing offers.
The above analysis reveals a cycle of first time visitors entering the site through PPC non-brand keywords and returning several times before converting. Since these visitors know the brand name from multiple visits, they use organic search (Segment-organic return visitor using brand keywords), or type the URL directly into the browser (Segment- PPC return visitor using brand keywords, or Direct visitor if the campaign cookie is deleted), to make the purchase.
This knowledge gets us closer to the Holy Grail in terms of better understanding our customers, and significantly provides us with actionable information to help win them over.