Time to purchase vs Time Lag for business cycle data

[datarat]

Hi All,

Using just GA is it possible to get accurate information about buying cycle duration - i.e. length of time from first interaction to purchase?

Time to purchase isn't what I'm looking for as it does not include direct visits.

Time lag seems off as it suggests that the majority of conversions happen on day 0 (as does time to purchase).  I know that any conversions outside of the time window default to 0 but I find it hard to believe that a clothing retailer has a similar buying cycle to a furniture store.  Of course I could be wrong here, I'm making a big assumption.

Neither report feels very robust to use for estimating split testing duration requirements.

I know that there are other methods available, by comparing raw clickstream / cookie data but I do not have this data available, nor do I have the time to get it.

Any info / advice would be great.

Thanks

[TomRayner]

Hey

In a word, no.

GA is pretty horrendous for this.  There are several very convoluted ways of trying to get this to work - but even if you start to get there, you're hampered by the fact that all of the information is anonymised - ie, you won't see what user or user journey was responsible for which purchase (unless you have ecommerce set up and each product is a different price).

There are a lot of other solutions out there that can do this for you, each with varying price packets.  One of the better ones out there is KissMetrics - starting at $200/mo - and they have a lot of support people on hand both in pre and post sales.  I'd approach them with your query and take it from there and see if it is right for you.

If that is too high a price for now, I've heard good things about Keen IO and Calq IO - again, reach out to them with what you're trying to achieve.  Be specific as well - they should be able to help you with your queries and the one that offers the best, most comprehensive support could be your winner.

[datarat]

Hi,

Thanks for your reply.  To clarify, in your opinion Time to Purchase and Time Lag are useless reports?

I've not had a chance to look at KM, KI or CI.  How do they work?  I assume they cannot work retrioactively and are cookie based (cookies with a long TTL).  There is an obvious point of failure here - I read somewhere (shaky citation I know) that cookies are cleared much more frequently now (monthly) so this data has to be heavily caveated at best.

Thanks again for your help.