Re: Webtrends Scenario analysis discrepancy

Expand Messages
• Wow! this is Great!Many many thanks for taking time to explain it so clearly. Thanks again.
Message 1 of 7 , Sep 2, 2008
Wow! this is Great!Many many thanks for taking time to explain it so
clearly. Thanks again.

--- In webanalytics@yahoogroups.com, "Nick Potter" <nickpotter@...> wrote:
>
> Ok, this will get complicated so I warn in advance (but it is I think
> one of the reasons why people have so much trouble with this).
>
> The number for each individual step (lets call this Ns) is calculated
> from:
>
> those coming from the previous step (Np),
> those coming into that step from *outside* the scenario (No),
> those coming to the step from other steps within the scenario other
> than the previous step i.e. going back (Nb)
> those leaving the scenario entirely (Ne), and
> those jumping to a step other than the next one (Nj).
>
> Therefore:
>
> Ns = Np + (No + Nb) - (Ne + Nj)
>
> Now the trouble is, the default view of the scenario shows you
> movement in the order you expect (step 1, then 2, then 3) plus people
> coming or going external to the scenario. So what you are missing is
> Nb and Nj above.
>
> The other view for the chart shows you movement in the order you
> expect (step 1, then 2, then 3) plus people jumping to steps out of
> order in the scenario. So this time you are missing No and Ne above.
>
> So what I was trying to convey in my original response is you never
> see all the figures into or out of a step at any one time and this is
> why the numbers never add up.
>
>
> --- In webanalytics@yahoogroups.com, "syedspiky" <syedspiky@> wrote:
> >
> > Hi Nick,
> > Many many thanks for your reply. appologies for my ignorance in this.
> > I have understood that there are one chart and one table explains the
> > scenario converions. But isn't it 'users coming from previous step to
> > the next step in the graphical chart' is movement within the
> > scenario? This number of users are clearly shown in the graphical
> > chart just by the 'down arrow'. Or do you mean that in the graphical
> > chart scenario Webtrends never captures all the users coming from
> > the previous step to the next step? Thanks again...
> >
> > --- In webanalytics@yahoogroups.com, "Nick Potter" <nickpotter@>
> > wrote:
> > >
> > > If I had a penny for every time my users ask me this question, I'd
> > > have a lot of pennies by now!
> > >
> > > The best explanation I got for this is that you are only looking at
> > > half of the picture at any one time as there are two charts you can
> > > view for a scenario.
> > >
> > > The first (and most common view) is movement between the scenario
> > and
> > > pages external to it. So basically it ignores movement within the
> > > scenario and only looks at visitors coming in from a page or leaving
> > > to a page that is not part of your scenario.
> > >
> > > The second view ignores all pages external to the scenario and only
> > > looks at movement between pages/steps in your scenario.
> > >
> > > So when you're looking at a single view and try to add up those
> > coming
> > > into the scenario, coming down from the previous step and take away
> > > those leaving it doesn't add up right because you're never seeing
> > > everyone coming into or leaving a step.
> > >
> > > The line I usually give my users is to uses the table at the bottom
> > to
> > > look at individual step and complete scenario conversion rates and
> > use
> > > the chart to see which pages people are entering or leaving your
> > > scenario on.
> > >
> > > Hope this makes sense!
> > >
> > > Nick
> > >
> > >
> > >
> > > --- In webanalytics@yahoogroups.com, "syedspiky" <syedspiky@> wrote:
> > > >
> > > > Hi all, I have a question with regards to the webtrends scenario
> > > > analysis. In my products conversion scenarios, the total users
> > > > (usually in the middle of the scenario conversion chart)in a
> > specific
> > > > step of the scenario never matches the total of 'the users coming
> > from
> > > > the previous step' and 'the users coming to this step from other
> > > > paths/pages'. Logically, this should be the total number of these
> > two
> > > > figures. But why this figure always less or more in most of the
> > > > scenario steps? This is creating doubts in the reliablility and
> > > > accuracy of the scenario analysis and conversion rates in
> > Webtrends.I
> > > > would really appreciate if anybody answers.
> > > >
> > >
> >
>
• Michael, Thanks. The explanation is born from frequently having to answer this question! I think its something WT needs to look at (perhaps their new UI hire)
Message 2 of 7 , Sep 2, 2008
Michael,

Thanks. The explanation is born from frequently having to answer this
question!

I think its something WT needs to look at (perhaps their new UI hire)
as it obviously causes everyone lots of grief.

Nick

--- In webanalytics@yahoogroups.com, "michaelnotte" <michaelnotte@...>
wrote:
>
>
> I fully agree with you that while very colourful, the scenario graph
> in WebTrends is bringing more confusion to most users and really
> give us hard time to explain it correctly with simple word. I wish
> there would a way to hide the figures (just keep the %).
>
> I mostly rely on the table as you said because, it is not easy to
> explain and when you try, most of the time users look puzzled and
> septical (and then they don't trust the tool :-))
>
> Michael
>
> --- In webanalytics@yahoogroups.com, "Nick Potter" <nickpotter@>
> wrote:
> >
> > Ok, this will get complicated so I warn in advance (but it is I
> think
> > one of the reasons why people have so much trouble with this).
> >
> > The number for each individual step (lets call this Ns) is
> calculated
> > from:
> >
> > those coming from the previous step (Np),
> > those coming into that step from *outside* the scenario (No),
> > those coming to the step from other steps within the scenario other
> > than the previous step i.e. going back (Nb)
> > those leaving the scenario entirely (Ne), and
> > those jumping to a step other than the next one (Nj).
> >
> > Therefore:
> >
> > Ns = Np + (No + Nb) - (Ne + Nj)
> >
> > Now the trouble is, the default view of the scenario shows you
> > movement in the order you expect (step 1, then 2, then 3) plus
> people
> > coming or going external to the scenario. So what you are missing
> is
> > Nb and Nj above.
> >
> > The other view for the chart shows you movement in the order you
> > expect (step 1, then 2, then 3) plus people jumping to steps out of
> > order in the scenario. So this time you are missing No and Ne
> above.
> >
> > So what I was trying to convey in my original response is you never
> > see all the figures into or out of a step at any one time and this
> is
> > why the numbers never add up.
> >
> >
> > --- In webanalytics@yahoogroups.com, "syedspiky" <syedspiky@>
> wrote:
> > >
> > > Hi Nick,
> > > Many many thanks for your reply. appologies for my ignorance in
> this.
> > > I have understood that there are one chart and one table
> explains the
> > > scenario converions. But isn't it 'users coming from previous
> step to
> > > the next step in the graphical chart' is movement within the
> > > scenario? This number of users are clearly shown in the
> graphical
> > > chart just by the 'down arrow'. Or do you mean that in the
> graphical
> > > chart scenario Webtrends never captures all the users coming
> from
> > > the previous step to the next step? Thanks again...
> > >
> > > --- In webanalytics@yahoogroups.com, "Nick Potter" <nickpotter@>
> > > wrote:
> > > >
> > > > If I had a penny for every time my users ask me this question,
> I'd
> > > > have a lot of pennies by now!
> > > >
> > > > The best explanation I got for this is that you are only
> looking at
> > > > half of the picture at any one time as there are two charts
> you can
> > > > view for a scenario.
> > > >
> > > > The first (and most common view) is movement between the
> scenario
> > > and
> > > > pages external to it. So basically it ignores movement within
> the
> > > > scenario and only looks at visitors coming in from a page or
> leaving
> > > > to a page that is not part of your scenario.
> > > >
> > > > The second view ignores all pages external to the scenario and
> only
> > > > looks at movement between pages/steps in your scenario.
> > > >
> > > > So when you're looking at a single view and try to add up
> those
> > > coming
> > > > into the scenario, coming down from the previous step and take
> away
> > > > those leaving it doesn't add up right because you're never
> seeing
> > > > everyone coming into or leaving a step.
> > > >
> > > > The line I usually give my users is to uses the table at the
> bottom
> > > to
> > > > look at individual step and complete scenario conversion rates
> and
> > > use
> > > > the chart to see which pages people are entering or leaving
> your
> > > > scenario on.
> > > >
> > > > Hope this makes sense!
> > > >
> > > > Nick
> > > >
> > > >
> > > >
> > > > --- In webanalytics@yahoogroups.com, "syedspiky" <syedspiky@>
> wrote:
> > > > >
> > > > > Hi all, I have a question with regards to the webtrends
> scenario
> > > > > analysis. In my products conversion scenarios, the total
> users
> > > > > (usually in the middle of the scenario conversion chart)in a
> > > specific
> > > > > step of the scenario never matches the total of 'the users
> coming
> > > from
> > > > > the previous step' and 'the users coming to this step from
> other
> > > > > paths/pages'. Logically, this should be the total number of
> these
> > > two
> > > > > figures. But why this figure always less or more in most of
> the
> > > > > scenario steps? This is creating doubts in the reliablility
> and
> > > > > accuracy of the scenario analysis and conversion rates in
> > > Webtrends.I
> > > > > would really appreciate if anybody answers.
> > > > >
> > > >
> > >
> >
>
Your message has been successfully submitted and would be delivered to recipients shortly.