Small businesses beat large advertisers at digital marketing

I have said often over the years that small businesses running their own digital marketing do better digital marketing than large advertisers. Put simply, if a small business (SME) spends $100 on digital ads and doesn't get a return, they don't have another $100 to spend. So they are maniacally focused on real outcomes, business activity, and sales all the time.

Whereas, the largest of advertisers hand their hundreds of millions of dollars of digital ad budgets to agencies to spend for them. These agencies' top priority is to "spend it all" so they can maximize their own revenues and profits, to please Wall Street each quarter. Spending it all may require them to buy fake ads because there aren't enough humans on earth to generate the 100's of trillions of bid requests flowing through programmatic supply paths every month. Conveniently, these agencies and large advertisers use legacy fraud verification vendors to cover up the fraud, because these vendors always report around 1% IVT ("invalid traffic"). They show these reports to their bosses and proclaim, "there's low to no fraud, let's keep spending." All too often, how much they spent becomes the advertiser's badge of honor and was the campaign "delivered in full" is the goal of every large agency. Why don't they proclaim whether the campaign drove any sales for the advertiser instead?

Vigilantly looking at analytics and outcomes

Let me share two stories of small business owners successfully beating ad fraud and optimizing their own digital campaigns, by constantly looking at analytics and outcomes.

The first small business owner sells his music and merchandise online. He had been running Facebook advertising for five years himself. When he first started, he was seeing very good returns -- i.e. people clicking through on his Facebook ads, arriving on his website, and making purchases.

During those five years, something changed. The number of impressions he got for roughly the same monthly budget continued to go up. But his sales continued to go down. He was even getting more and more clicks and higher click-through rates, according to the campaign interface. But, he observed that the discrepancy between the number of clicks reported by Facebook compared to the number of visits that arrived on his site got so large - 90% - that he reached out for help. For every 1000 clicks reported by the Facebook campaign interface, he only saw 100 arrivals on his website, according to Google Analytics. Something was terribly wrong.

Within five minutes of hearing his case, I asked him one question: Was the checkbox for Facebook Audience Network checked or not checked? He said it was checked (it is checked by default). This meant his ads were shown on all of the sites and apps outside of Facebook. I told him to UN-check that checkbox to turn off Facebook Audience Network (FAN). Within the week, he saw the number of impressions plummet and the number of clicks on his ads also dropped through the floor. But the sales of his music went back up. What happened?

As it turns out, over the course of the five years, the ratio of his ads being shown on Facebook itself versus on the audience network outside of Facebook had completely flipped. In the beginning, most of his ads were shown on Facebook; by the end, most of his ads were shown outside of Facebook.

And that was the problem. The audiences outside of Facebook were not humans. The Facebook Audience Network generated large quantities of ad impressions, and some bots even clicked on the ads, resulting in higher click rates, but none of that resulted in actual sales. By forcing his ads to show only on Facebook, this small business owner cut out the vast majority of the ad fraud and saw his sales per-ad-dollar-spent go back up.

Our next small business owner ran campaigns in Google Adwords herself. She ran many campaigns over the years, but the recent ones caught her eye. When she turned on a campaign and checked her Google Analytics, she saw an 118,600% increase in Android devices hitting her site. But there were literally no additional goal completions. And her conversion rate dropped dramatically because none of this additional traffic “converted” (made purchases). ?

Digging into the analytics further, we could corroborate that this was fraud -- a botnet pretending to be Android mobile devices was clicking through on her ads, instead of humans genuinely interested in her product. Analytics revealed the time on site all averaged less than two seconds; bounce rate spiked upward; and pageviews per session were 1.2 (not sure how Google Analytics calculates that). She turned the campaign off right away and even got a refund for the entire campaign after showing the analytical evidence to Google (no special fraud detection tech needed).

"Common sense, and vigilance in looking at your own analytics will enable you to reduce or eliminate ad fraud in your own campaigns, with no fancy fraud detection tech whatsoever."

Leaning on crutches that let you down, literally

So how did digital advertising get so bad for large advertisers for so long? Simple. Everyone wanted to believe what they wanted to believe, and eagerly paid for the next shiny objects that were sold to them by adtech-oil salesmen (the new "snake oil salesmen" as it were). Even after I showed them most of it was fabricated out of thin air, via 600 articles over the last 11 years, large advertisers, trade associations, and hold-co media agencies doubled down on using legacy verification vendors as their crutch to cover up the fraud. They knew there was more than 1% fraud, but they continued to cite legacy vendors' low fraud numbers to justify continuing to spend in suspect channels. There was no other way they could "deliver in full" on campaigns where advertisers demanded irrational quantities of ad impressions at irrationally low CPM prices.

They also wanted to believe that they could show the right ad to the right person at the right time, because the adtech vendors selling those services told them they could. They eagerly paid for the magical shiny objects of "precision targeting" and audience "segments of one." Even when they observed high bounce rates, low time on site, and no activity on the landing pages, they chose to believe it was working -- "oh, we were doing branding, so we don't really expect users to do anything on our sites anyway." Advertiser and agencies alike leaned on the crutches of legacy fraud verification that ultimately let them down, by not actually protecting them from any of the fraud and waste. Everyone has claimed for years, they've been paying for fraud detection and prebid fraud protection so they must have been protected from fraud, right? Advertisers and agencies are realizing they have not been protected as well as they had assumed.

So, where do we go from here?

I am highly encouraged to see a shift in the tide, a swing back the other way in the pendulum, etc. The largest of advertisers and some hold-co media agencies are finally opening their eyes to the reality of digital marketing. They could not see clearly before when using the legacy verification tools. Their business decisions were negatively affected by the legacy vendors' failure to detect fraud and waste. But new analytics tools are available -- like FouAnalytics. Advertisers and agencies alike can take back control and not only measure where the ads went, but also whether the clicks they got on the landing pages were human or not, attentive and valuable or not.

If advertisers became more like small business owners, and vigilantly checked their analytics and outcomes metrics, they too can do better digital marketing.

I have the tools and the experience to help you. The question is when would you like to start using FouAnalytics to "see Fou yourself"?

Further reading: Analytics that report bots on your site - FouAnalytics