6 ms·

What do we really know about the effectiveness of digital advertising? The effectiveness of online ads is easily measurable: Return on ad spend (ROAS) = (reven

What do we really know about the effectiveness of digital advertising? The effectiveness of online ads is easily measurable: Return on ad spend (ROAS) = (revenue from ad referrals) - (money spent on ads)

cratermoon 5y ago
> revenue from ad referrals And exactly how do you get accurate figures for that?
- tofukid 5y ago
  It’s simple. You use a URL with the campaign id, and store that when the customer clicks through, then when a sale is made you count that towards the campaign. In offline advertising people would achieve this using campaign-specific phone numbers or discount code given in the ad. Online advertising math is not rocket science. If ROAS is positive you continue the campaign, if not you abandon it or try other methods.
  - klvino 5y ago
    Guess we'll skip branding and awareness campaigns, ignore all post-view purchases, not use cross-device or cross-channel attribution, and not look at attribution of offline conversion/visitation activity. If the only thing you do is Search and Email Blasts, the query string on your URL might work okay.
    - namanyayg 5y ago
      The query string method will not provide stats for the above point; but the newer methods e.g. using the FB pixel that tracks users based on the user's email/userid you create can accomplish much of that (only offline conversions are difficult, but there are still modern ways to match an offline customer with their past ad exposure e.g. if you get their email/phone on purchase...)
  - fieldcny 5y ago
    Maybe if your business is simple and purchase cycle fast. For the majority of companies that’s not the case. There are other confounding factors, like the margin attached to that revenue. Sometimes understanding the actual profitability of a sale can take years. As others have said for big brands it can be incredibly hard to do real attribution.
  - cratermoon 5y ago
    Direct conversion to sale via clickthroughs are an almost invisibly small part of online ads . No advertiser uses only that metric for determining ROAS.
    - tofukid 5y ago
      Which metrics are you referring to? I can’t speak for ad agencies, but every company I’ve worked with to do advertising measured ROAS using customer attribution (which ad campaign brought which customer, and how much did that customer spend).
      - cratermoon 5y ago
        > which ad campaign brought which customer And how do they determine that?
  - erostrate 5y ago
    TFA spends a long time belaboring the point that this is not correct. Because many people who click on your URL would have bought the item anyway.
- Qi_ 5y ago
  Facebook for example has a “tracking pixel” script that can be included in your website to get detailed info about conversions from Facebook ads.
  - fieldcny 5y ago
    If it were really this easy people would be doing it, the world really is a complex place.
newshorts 5y ago
Exactly what the marketers thought. The point I took from the article is that you have to factor in what you would have gotten for free.
- sangnoir 5y ago
  You can stop advertising for a while and compare the outcome. Or take 2 similar markets and only advertise in one of them. Online tracking enables fine-grained analytics, there's little left to guessing - especially for small or medium businesses who do not (yet) have high brand familiarity.

cratermoon 5y ago

> revenue from ad referrals And exactly how do you get accurate figures for that?

tofukid 5y ago
It’s simple. You use a URL with the campaign id, and store that when the customer clicks through, then when a sale is made you count that towards the campaign. In offline advertising people would achieve this using campaign-specific phone numbers or discount code given in the ad. Online advertising math is not rocket science. If ROAS is positive you continue the campaign, if not you abandon it or try other methods.
- klvino 5y ago
  Guess we'll skip branding and awareness campaigns, ignore all post-view purchases, not use cross-device or cross-channel attribution, and not look at attribution of offline conversion/visitation activity. If the only thing you do is Search and Email Blasts, the query string on your URL might work okay.
  - namanyayg 5y ago
    The query string method will not provide stats for the above point; but the newer methods e.g. using the FB pixel that tracks users based on the user's email/userid you create can accomplish much of that (only offline conversions are difficult, but there are still modern ways to match an offline customer with their past ad exposure e.g. if you get their email/phone on purchase...)
- fieldcny 5y ago
  Maybe if your business is simple and purchase cycle fast. For the majority of companies that’s not the case. There are other confounding factors, like the margin attached to that revenue. Sometimes understanding the actual profitability of a sale can take years. As others have said for big brands it can be incredibly hard to do real attribution.
- cratermoon 5y ago
  Direct conversion to sale via clickthroughs are an almost invisibly small part of online ads . No advertiser uses only that metric for determining ROAS.
  - tofukid 5y ago
    Which metrics are you referring to? I can’t speak for ad agencies, but every company I’ve worked with to do advertising measured ROAS using customer attribution (which ad campaign brought which customer, and how much did that customer spend).
    - cratermoon 5y ago
      > which ad campaign brought which customer And how do they determine that?
- erostrate 5y ago
  TFA spends a long time belaboring the point that this is not correct. Because many people who click on your URL would have bought the item anyway.
Qi_ 5y ago
Facebook for example has a “tracking pixel” script that can be included in your website to get detailed info about conversions from Facebook ads.
- fieldcny 5y ago
  If it were really this easy people would be doing it, the world really is a complex place.

newshorts 5y ago

Exactly what the marketers thought. The point I took from the article is that you have to factor in what you would have gotten for free.

sangnoir 5y ago
You can stop advertising for a while and compare the outcome. Or take 2 similar markets and only advertise in one of them. Online tracking enables fine-grained analytics, there's little left to guessing - especially for small or medium businesses who do not (yet) have high brand familiarity.