Home – New Forums Marketing mastery Is Google Analytics useless for assessing your Facebook and Mobile Traffic?

  • This topic is empty.
Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #986790
    JohnW
    Member
    • Total posts: 2,642
    Up
    0
    ::

    Google Analytics has become the default website metrics program.

    We are also seeing huge growth in mobile usage and Facebook referrals.

    So what does Analytics tell us about mobile and Facebook users?

    The answer is possibly very little that is accurate!

    Some Analytics metrics that can become completely inaccurate include:

    • Time on Page
    • Engagement

    Unless things have changed, Analytics cannot identify the time when a visitor leaves a website.

    This limitation has major implications because:

    Many mobile phone visits end in single page visits for phone numbers, addresses or prices. Mobile phone users don’t visit many site pages.

    Facebook links tend to result in a single web page visit.

    Analytics cannot identify when the last page on a website was requested.

    This means…

    Mobile phone users that check out one site web page are recorded as a visit duration of “0” seconds.

    Many Facebook users also end up as single site page visits and are therefore recorded as a visit duration of “0” seconds.

    Here’s my problem!

    I have a client who generated 500 mobile visitors to one page of their site recently. Analytics reports all of them as Facebook referrals with “0” seconds duration because they did not visit another page.

    Was my client’s Facebook promotion a success or not?
    Regs,
    JohnW

    #1159652
    Anonymous
    Guest
    • Total posts: 11,464
    Up
    0
    ::
    JohnW, post: 184144 wrote:
    Was my client’s Facebook promotion a success or not?

    Hi John,

    The simple (and possibly frustrating answer) is ‘it depends’.

    I agree, Google Analytics (GA) has its shortfalls. But its benefits far outweigh all of these for SMEs – and it has come a long way in recent years.

    But rather than a nitty gritty discussion about the technical side of GA, I think the question here is “what does campaign success look like?”.

    For the SMEs I work with, campaign success is typically based on number of leads generated or dollars generated /campaign investment. And there is no shortage of ways to measure both of these factors through GA.

    So my suggestion for your client is that they get perfectly clear on what they want to achieve and then work back from that goal to ensure they have a way to measure their ROI from Facebook.

    Everything else (i.e. page visits, CTR, impressions) can be used to deconstruct what went right/wrong. For example, if your 500 page visits came as a result of 600 people seeing your ad (i.e. a CTR of 83%), then you know you’re saying something right! But if only a few go on to perform the desired action on their website, then they need to focus their attention on their landing page.

    Setting up the right goals from the beginning is definitely the best way to go. It might take a little extra time, but it will pay for itself many times over in the long run.

    – Kieron

    #1159653
    John Romaine
    Participant
    • Total posts: 1,108
    Up
    0
    ::
    JohnW, post: 184144 wrote:
    I have a client who generated 500 mobile visitors to one page of their site recently. Analytics reports all of them as Facebook referrals with “0” seconds duration because they did not visit another page.

    Was my client’s Facebook promotion a success or not?
    Regs,
    JohnW

    You’re focusing on the wrong metrics.

    You should be focusing on conversions, not “visit duration”.

    I never once worked with a client that got excited because visitors to his website were sticking around for “a long time”.

    Analytics is as useful as you need it to be. Problem is, most people don’t know how to use it.

    If your client was running a FB promotion then they should’ve been tracking and measuring using trackable links > conversions. NOT visit duration.

    #1159654
    Simply Money Honey
    Member
    • Total posts: 78
    Up
    0
    ::

    I agree that you are looking at the wrong metrics in Google Analytics. Time spent on a site is somewhat irrelevant.

    What was the objective of the facebook promotion? Was it to sell a product? Get newsletter signs ups? If you set up goals in Google Analytics you can determine whether people reach a particular page on your website (e.g. receipt page after a sale or thank you page after a newsletter sign up). Then you will be able to see whether the traffic from facebook is actually converting.

    Cheers

    Gabrielle

    #1159655
    Aidan
    Member
    • Total posts: 1,125
    Up
    0
    ::

    To me, it all depends on what the objective is.

    I don’t imagine using time on site measurements for much, or bounce rate… (bah, don’t get me started on that one)

    …what matters is whether or not the visitors got what you wanted them to get, saw what you wanted them to see, or did what you wanted them to do. Those are easily tracked in GA.

    :)

    #1159656
    JohnW
    Member
    • Total posts: 2,642
    Up
    0
    ::

    Hi Guys,
    Thanks for your input.

    1. Google Analytics
    Let me document this as a case study to show people how complex website and social media metrics can be.

    The last few Analytics suggestions are great in theory. The problem is there are so many situations where they have serious limitations and this seems to be one of the many.

    Here we are talking about a service industry that has 8 locations in a number of different cities across 2 countries.

    The primary objective of the website is to generate new clients for all of the 8 businesses’ practice locations.

    The client wants phone calls from new clients. There are 8 Aust phone numbers and a NZ phone number.

    The client does not have the capability of centrally tracking phone calls and appointments made across the 8 location network.

    There is an online enquiry form that can be tracked but the expectation is that email enquiries should be a small part of new client responses.

    That pretty effectively kills suggestions aimed at Analytics event tracking, don’t you think?

    2. Facebook Value
    What happened on top of this situation is the introduction of a Facebook promotional program by a third party. This generated 500 referrals in a month.

    The client thinks this is a great response.

    When I view the content of the Facebook landing page, common sense indicates it is rubbish.

    The point is that Analytics provides no information to support this contention.

    My real intent of this post is to provide website owners with educational information.
    Regs,
    JohnW

    #1159657
    ITCrazy
    Member
    • Total posts: 31
    Up
    0
    ::

    I have noticed some reasonably accurate data from GA. It’s probably fair to say that Google is really only just breaking into the social media analytics field, but the features they do offer seem okay.

    Most of my traffic comes from Google search and ads, but I do get some from Facebook etc.

    #1159658
    Kieron_Brown
    Member
    • Total posts: 1
    Up
    0
    ::

    Hi John,

    In this situation, best practice would be to use call tracking numbers.

    They’re dead cheap through a company called CallTrackingMetrics (about $6/month plus call charges. Probably $30/month per number all up), easy to setup and easily tracked through AdWords and Analytics.

    I recently wrote an article on the topic which might be useful to get your client over the line if they need convincing: http://www.marketingresults.com.au/blog/2014/02/24/online-call-tracking/

    I hope that helps. God knows they’ve made my life easier!

    – Kieron

    Sent from my iPhone using Tapatalk

    #1159659
    AngusP
    Member
    • Total posts: 53
    Up
    0
    ::

    good question John, I’ve been wondering myself.

    Cheers

    Angus Pryor
    http://www.galwijihomestead.com
    http://www.mysteryshoppingguide.com.au

Viewing 9 posts - 1 through 9 (of 9 total)
  • You must be logged in to reply to this topic.