The metrics snafus have raised concerns for some marketers, ad buyers and publishers. Facebook says the problems haven’t impacted billing. Still, some executives say incorrect statistics can affect how ad buyers allocate budgets, and Facebook has been under pressure to allow more thorough and independent measurement by third-party firms.
+Commentary: This doesn’t directly affect many, and certainly no one reading this blog, but it should make you question whether the decisions you personally make off the metrics they provide are any good? Internal tests conducted by us have proven time and again there are huge flaws in their analytics, their dashboard, and other parts of the reporting process.
So their constant barrage of “Spend $5 and reach XXXX” amount of people that greets us every time we log in is a constant reminder that it probably won’t reach that many people, and we won’t know who to believe. This may seem like a pattern of fraud or as we think of it gross incompetence.
The very fact that these snafus happen prove emphatically that they do not, or won’t allow independent testing of it. That numerous reports already exist on the web of people saying they are rigged and not useful. Yet they continue to bill both the high-dollar publishers above, and small business, solo-entrepreneurs or the microbrands too much for something that may not work.
Without anyone every seriously asking: What do you get for your money? What is the benchmark for success? How do we know it is what they say it is?
What these dashboards, metrics, and other things are supposed to provide you with is business intelligence. They’ve hacked this together, it is an after-thought, and yet it is the primary reason you would want to know how far a dollar goes there. Not very far apparently.
This doesn’t mean you can’t have success, but it is or has been non-strategic, designed more to be a flash in the pan, and raise their revenue, while not providing you with the tools you really need to leverage your limited budgets. Think about it, these are people with enormous budgets whose audit of these things realized that they were getting gypped. Surely they noticed what seemed to them to be lackluster iPhone numbers, and as such could have taken different measures, or made what they did already as a response useless.
The under-reporting has real-time and longitudinal impacts, and if this were a one-time event we might be able to shrug. Yet if we are largely unaware of how any of it works, or if it is even working, or if there are faults we are waiting for them to point them out to us, then we are beholden to them. That an error in one part can make the whole chain suspect, and the data flawed.
It is analogous to TV/Cable running “ads” that they then guess how many people saw them or use Nielsen or some other service. This is the web, these clicks and views are logged, and should show up. That there are bugs that could leave a substantial gap for a month should bother everyone.
This is where we’ll end this particular note, but this ties into a longer piece we are working on about the entire “Fake News” scandal that was supposedly not impactful, but now has 3rd Party verification tools coming soon. These two things are sides of the same coin, and are complexly interwoven into a primary fault with the platform. That both underneath in the algorithmic machine learning depths to the front-end user design, they contribute in outsized ways because of the breadth of their platform.