Video: The Effects of Bad Data

AKA The Damage Done by Data Sea Monsters! When the six main causes of bad data combine, you can see some real problems. Elizabeth Crinejo, Director of Client Operations, shows 4 real-life cases of data-driven concerns.

From the session “Drowning in Data, but Thirsty for Insight?” given by Elizabeth Crinejo and Melissa Mines at ConnectToConvert, August 21, 2017.

Transcript

So, let me give you a couple of examples of this out in the wild. So, this is in that world of having your data enculturated, having conversations about your numbers be a part of the company. So, this was a client that we had that was switching to a new website, and these are all quotes that I heard at some point on the phone or in person. So, the 83 percent bounce rate was not good news. It was on the new site, so we were at 75/25 moving over to the new website, and all of a sudden we see this bounce rate. Now, this was a big deal.

So, there were high-level people on this call. The marketing team was on this call, and we had to do a lot of our digging amongst all of us to figure out what was going on. So, what happened was they had no fewer than 10 different tools that were hitting this site with a variety of different types of tests. So, here this tool was definitely doing what it was supposed to be doing, and that’s what it knew it was going to cause 100 percent bounce rate, but this conversation wasn’t enculturated.

So, one side wasn’t talking to the other saying, “Hey, we’re about to go from 75 to 100 percent. We need to make sure that everything looks good.” “Oh yeah wait. I was about to run a test. Should I wait?” That wasn’t the conversation. So, the site move was delayed another 48 hours while we figured out what all this stuff was. So, don’t have that happen; keep it part of the conversation. This also lands in the world of enculturating your data.

So, this was something I heard from a client not too long ago. “We don’t have keyword level data.”

Which it stopped me in my tracks because we had worked with this organization, and it’s a company—they have a very long funnel that can go anywhere from three months to two and a half years, and we had worked with them really closely to build out keyword level data, which they have from the leads all the way to closed/won, and there’s revenue dollars on the other side of these numbers as well.

Now, the problem was two people had left the company, and they hadn’t kept the culture of data alive. Those two people were the stewards of the data conversation we found out later, and so when they left, so did this level of granular market, and it definitely hurt them. So, the thing for you guys to think about is make sure you’re keeping the conversations alive.

Here’s another one. This was a prospect, and we do audits with potential customers who are really interested in making sure we can make a difference for them, and, you know, if their agencies are good, we usually recommend them staying. So, that’s what our audit helps us figure out, and in this one we always ask the goals. They said, “Okay, our CPL is really solid. We just need to grow. We just need to scale.”

So, when we actually dug into the data, we saw well that’s not really the case. Their whole portfolio, as you can see here, at that $77.23, their whole portfolio was relying on the performance of their brand, but the rest of their portfolio was fairly weak when they were playing against $110 CPL cost per lead here. So, with this we would recommend, no don’t grow yet. Build your foundation, make it stronger in your non-brand categories so that then you actually have something that can really build and scale without the integrity that you need to keep that $110 in play. So, just remember this is the world of like having the fullest picture.

Then another, just to address the vanity data issue. I’m hoping this one doesn’t happen as much as it used to. This is really the legacy world of billboards and eyeballs and all of that. So, this is just an illustration of how vanity data can really cost you money, and this is, you know, those calls, those emails like, “Impressions are way down. What’s happening?”

So, this was a prospect, and we did an audit of their account. They were in the electricity market back east where it’s deregulated, and they were like, “We just want traffic.” So, they got traffic.

Now, they have broad match keywords like coal, gas, wind, electric, electricity, and power. So, they were really on point, $23,824 spent helping people find gas the lowest, the cheapest gas price near me. They could have also gotten electricity if they wanted to. I don’t know why they clicked the ads, but they did, and it cost them almost $24,000. Other things were Power Rangers, Transformers. I’m pretty sure these were like 12 or 13-year-old kids because we’re not talking about like transformers on the power line. We’re talking about like Decepticons and those guys, yeah.

Volts. They were trying to give a lot of information to people about voltage for different things, and that was—I mean well it’s only $186 dollars. These are all clicks that are not likely to turn into customers. We could have a whole session on why people look for one thing and then click on your ad. If you have that problem, it can be really fascinating and worrying, but $40,000 and change went to this type of traffic, and I would assert that the line share of this—you know maybe someone who’s looking for the benefits of coconut oil probably might also be upset about their electricity bill. I don’t know, but I do know that none of this money converted. So, not all traffic is good traffic. Any questions?