Tracking PPC effectiveness and ROI?

[cgchris99]

We have starting programming our online store to track referrals from PPC.
At what point do you decide if your PPC was effective?

What I mean is a customer clicks the PPC ad and doesn't buy during this visit. Does this mean that there is no return on this click?

But if they come back in a couple days and purchase should this be recorded for this PPC campaign?

How long after the initial PPC click should a purchase show as a positive for the PPC campaign?

Any thoughts?

Thanks

[wsm2006]

I would give it as much as 30 days.

Shoppers are information seekers (an important acknowledgement for e-tailers) and rarely buy on the first visit.

My wife for example returns at least three times to the same site before making a purchase and she buys stuff ALL the time online.

[MikaelTC]

Do you have cookie tracking on your site? I know there's a debate on the effectiveness of cookies and the rate of cookie deletion (the estimates I heard some months ago range from 5%-55%, which means no one has a clue).

Right now it's the best tool we have to track our sites. You can get pretty good tracking for cheap (or free if you can get a Google Analytics invite).

[nadmedia]

You want to treat each click as its own. For an example:

(Does not take into the account of fraud)

Your bid for the keyword "water bottle" cost you $10.00 per click.

If you receive an average sale of $40.00 for every four clicks, your ROI at CLICK THROUGH level is $0.00. If your average sale is $80.00 for every four clicks your ROI at click through level is $10.00. That is why it is important to TRACK, TRACK, TRACK.

The returning customer that makes a purchase will be tracked back to the original click and previous sources. To be successful with internet marketing you need to track each transaction as far back as you can and understand what is going on at each level such as: referrer url, keyword, landing page, time of day, time on page, etc. The more information you collect the more successful you can be, depending on how you use and interpret the stats.

[MikaelTC]

Yes, you want to make sure that you are including all click costs in your calculations. If you have more than one advertising channel (organic search, paid search, banner &c.) you need to make sure that you are correctly associating sales to the source. Wouldn't want to cancel a campaign thinking that it doesn't work only to find that your sales have stopped.

>But if they come back in a couple days and purchase should this be recorded for this PPC campaign?

Depends, did that user directly enter your url? If so, then yes, that should be counted for PPC. If that user came in from another ad you enter some interesting territory. Depending on whom you ask you'll find many ways of calculating the "source" of the sale. I think the most common, and the simplest, is called 'Credit Last' and attributes 100% of the sale value to the most recent channel. If I click on your ad for "water bottles" and then I see your banner on a water bottle site, click through and purchase, the credit would go to the banner.

>How long after the initial PPC click should a purchase show as a positive for the PPC campaign?

As long as you like. It's good to keep track of all the information you can collect so you can look at it later. If you know that 80% of your customers purchase within two weeks, you can know that your new advertising program might not look very profitable until at least two weeks out. Or, if you have to cut a program to save money, you can still expect to make some sales from that channel for two weeks.