Question

Topic: Advertising/PR

Attributing New Members To A Tv-advertisement

Posted by johnjanssen1984 on 5/14/2015 at 4:21 PM ET 250 Points
Dear Marketing Fanatics,

I am working for an e-commerce company that is broadcasting a Television commercial. Our culture is very data driven and fact based, therefore it is essential for the marketing department to be able to show how many new members and ROI a certain marketing (e.g. online marketing ads and voucher distribution) action delivers.

For my research into the performance of the TV commercial I’d like to indicate the revenues gained from the TV members. Therefore I first need to attribute all the new members to either TV or Non-TV. One attribution system used by our company attributes all the new members to TV when a new member subscribes within 8 minutes of the broadcasting time.

To have a more representable view on the revenues one receives from the TV members the model should be made more strict to my opinion, leading to a smaller number of TV members but with a higher likelihood that is due to TV. Therefore I made my own model in which I took the 5 minute rule to attribute the members to TV.

Do you guys think this is a good way to approach the attribution model or do you have any further suggestions?

Thanks so much for responding!
To continue reading this question and the solution, sign up ... it's free!
SUBSCRIBE
Or, SIGN IN

RESPONSES

  • Posted by Gary Bloomer on 5/14/2015 at 4:40 PM Accepted
    What do you mean by "the TV members"? Are you referring to viewers, to stations, or to something else?

    How will you track the actions you're inviting viewers to take? How will you know that a specific action taken at 9:00 p.m. is directly attributable to a slot you aired 5 minutes ago? How will you know if the action is related to the broadcast of the slot? My point is that you won't, not unless you have live operators asking questions along the lines of " … and how did you learn about us?"

    Over how many weeks will your spots run and to what kind of demographic will they be shown?

    How will you track actions from PPC and press ads that appear on specific sites or in particular publications? Unless or until you know where the actions are driven from, how can you assess the return on your ad spend per message, per medium, per time slot, per viewer or reader catchment area?
  • Posted by mgoodman on 5/15/2015 at 6:32 AM Accepted
    The objective of television advertising is usually to create broad awareness, in hopes that it will eventually lead to a sale. TV isn't known for driving immediate customer response.

    When companies want to measure the impact of television advertising, they sometimes compare sales trends in two different areas -- one with TV advertising and one without -- over a period of several months, or even a year.

    If you find a better way, you can probably sell or license it to lots of businesses who have given up trying to measure the direct effect of television advertising.

    I have a client who has been advertising on television for 20 years. He has no clue whether it's a good investment or not, and business doesn't drop off if he stops for a few months. Yet he's convinced it's working because customers tell him they saw his commercial. (They tell him this even when he's been off the air for several weeks!)
  • Posted by johnjanssen1984 on 5/16/2015 at 10:53 AM Author
    Hi Guys,

    Thanks for being critical on the subject.
    I am aware that the effect of TV is for the largest part immeasurable: brand awareness and customer reactivation. However the one measurable part is direct response, which can work as a sample for the whole cohort of members acquired due to TV. One can argue that TV is a mass media that is not measurable however you can also attempt to measure it with all the side notes in mind.
    To give an answer to the question about how do you know what a TV members is; it is a subscription only website so when a member subscribes we know the exact time which we can then compare to the tv spot time.
    So indeed we are also looking at the overall sales increase on a macro level but don't you think this estimate is just way too rough? Sales is so dependent on seasonality that you can't just attribute the whole increase to TV..
  • Posted by Jay Hamilton-Roth on 5/16/2015 at 1:10 PM Accepted
    Why not have a separate landing page for TV spot, to better track signups (rather than assume if a signup occurred soon after a spot aired)? And if you have the ability, separate spots with different landing pages for each, to track which spot (useful for time-shifted viewing) produced which signups.
  • Posted by johnjanssen1984 on 5/17/2015 at 5:42 AM Author
    Good point!
    This would be the solution to the problem however it is not very sexy if you're advertising for a brand and it then says for example go to: www.apple.com/tvcommercial
  • Posted by mgoodman on 5/17/2015 at 7:28 AM Moderator
    With bit.ly you can create custom URLs so each landing page can have either a relevant word/name and/or a short nonsense name (like bit.ly/Soap23, etc).
  • Posted by Jay Hamilton-Roth on 5/17/2015 at 8:50 AM Member
    Or, you could craft subdomain names: tv.apple.com, for example (that would forward to the relevant landing page).
  • Posted by Shelley Ryan on 6/3/2015 at 6:06 PM Moderator
    Hi Everyone,

    I am closing this question since there hasn't been much recent activity.

    Thanks for participating!

    Shelley
    MarketingProfs
BACK TO TOP

Post a Comment

Related Discussions

Community Info

Top 25 Experts
(Advertising/PR)