Watch the episode or tune in on Ops Cast to find out what the mistakes were and how (if possible) they were resolved.
Avoid your next MOOPs with automated monitoring across your entire marketing & sales tech stack. Learn more.
Episode 12 – Harjot Singh
Hi, everyone. Thanks for joining us for another episode of MOOPs TV. I’m Jessica. And today I’m joined by Harjot. Um, Harjot, why don’t you start with some quick introductions? Who are you, where do you currently work and how long have you been in the marketing ops space?
Sure. Yeah. So my name is Harjot Singh, and, um, I live in beautiful Bay Area and, uh, I’ve been the mark.
I’ve been in the marketing space for almost 10 plus years, but my experience from Hitachi, Uber, Live Ramp, and now at my current role, at Zip, and what I really like about marketing ops is really looking at our customer journeys and evaluating our customer journeys. That’s what my passion is.
I love it.
It’s a great passion. I think, uh, sometimes when evaluating tech people can lose sight of the customer and their journey through the whole process. And I love it. That’s your area of focus. So we’ll get into the specifics of your mistake here in a minute, but at what point in your career did the mistake we’re going to chat about today?
So this really happened around 2019 when I was at live ramp. And, um, we were in the midst of optimizing our landing pages and, uh, getting a tracking in place. So that’s when it happened. Awesome. So optimizing landing pages could be any number of systems. Um, what systems were you working in? And tell me a little bit more.
What it is you were trying to deal. Yeah. So Weaver, um, as a company, if you were expanded, expanding globally into new markets, we were, we were in the us, we were in, uh, UK, France, and then, uh, in Australia and India and Singapore. So a couple of countries globally, and we were creating campaigns and launching campaigns in each of these countries.
And as part of each. Requirement, we would have landing pages for our assets and the challenge there was that each of these, um, regions had their own website and we wanted to keep their landing pages separate, uh, so that they can be localized for the regional leads. And. Also tracked independently. So if you want to be tracking them independently, you needed separate Google tracking codes to make sure that they are being tracked separately.
And we can do the reporting cleanly for each of the regions. Right. And that was the whole process of making sure that we can have the right tracking for the landing pages and all these landing pages were created and more cattle for us. So what, tell me a little bit more about what specifically broke or what the moves was.
Yeah. So from an optimization point of view, what we were really trying to do was we wanted to move away from having to. Add a new Google tag manager code in each landing page. So we came up with a unique idea of using tokens for our, uh, Google tag manager. So we would create folders for each of our countries.
And the campaigns for each of the countries and landing pages would we created under that folder and the tokens would set at the Forder level. So if we would have just one place where we would add the GTM code for each of the regions, and it will just trickle down to each of the assets that are created for each of those regions.
And we would just tokenize. GTM code in each of the landing pages. And that’s it, it was simple. We thought we did it. And it was good. That optimization was successful. That sounds like a really smart way of approaching that problem. Yeah. We are very proud of what we did there. So really, really optimizing our tracking there.
Yeah. But, um, So we really solve for that. Okay. But something else went wrong. I imagined, uh, as part of cracking we’ve we had different tracking codes, which we were integrating with within Google tag tag manager. We had our Google analytics score. We had our Marketo munchkin code. We were using Bizible for attribution.
So that’s what we did, uh, as part of optimization. And the funny thing was what I really did. And this is where the moves part comes in is, um, in order to place the Bizible tracking code, as for Google tag manager, I worked with our rep team, right. And we opened up our request has all marketing ops people.
We worked through project management systems and I opened the request. I added the code directly in the request as a comment and told my web person to update it. Right. And they did. And that’s it. The project was done and recheck everything else. Everything was working, but we did not check the physical tracking was working because Bizible touchpoints take some time to be created.
And I, we just didn’t spend enough time. And at that time we weren’t using Bizible, holistically as part of our reporting. So that essentially slept through. And the MOOCs moment really happened three months down the line. When we started to really look into Bizible user channel reporting, because that’s where the power came in.
From what we could tell that, which are the sources, which are driving registration, you don’t really need a key. You don’t need a specific way to track UTMs so Bizible or does that for you. And that’s why we started looking into the channel tracking and build that report. And three months down the line, when we were building that report.
Our marketing manager told us that the Bizible report that you’ve shared with me and the Marketo order board that you’ve shared with me, they don’t match the form fills for these assets don’t match. I’m seeing a hundred form fills in Marketo and. Really no form fills in Bizible. How that’s possible discrepancy of one or two totally.
Within the realm of possibility, right? Like they don’t have tracking pixels enabled on their computer. They’ve got a pop-up or a plugin blocking all that, but a full, there are none would make my eyes go like. Exactly. Um, and that’s what really happened. Uh, we, it was all mysterious. We started digging back into it.
What really the issue is. And we looked at the tracking code, everything looked good from a naked eye perspective. All the code looked good and look good. And our project management tool looked in principle. It looked in Google tag manager look exactly right. No issues right there. Um, so we raised it to Bizible that this is what’s happening and, uh, Bizible came back to us.
And when we copy pasted the Bizible tracking code from Bizible to the editor in our project management tool, for some reason, those double codes actually got inverted.
And to the Bizible to the naked eye, you really can’t tell that. And when we copy pasted that from our project management tool, back into Google tag manager, That’s what the wrong code got copied and was just, just inversed, double coats. And that’s just the nature of editors. They form, it paints a little differently and that’s what really happened.
Such a silly mistake. Very, very costly. Everyone is so familiar. I think. People who listen to this are relatively familiar with Marketo email editor and making sure you don’t or any of the MIP is, and making sure you don’t copy and paste out of a word document that will make formatting and things like that.
But I assume that most people probably don’t think about it when it comes to like a line of text that is the code, um, and how that may also be impacted by the same thing.
Okay. So it sounds like at this point, the air has been going on for three months. So we have three months or so of lost data that we probably can do nothing about. And that the error was found luckily internally, but by a stakeholder who is relying on these reports to make, um, business decisions. Um, what happens kind of at this point, did, what did you do to remedy and how did you acknowledge the miss the issue to the rest of the team?
Yeah, I think, uh, we D we definitely found that issue. Um, I find, I found myself guilty and I feel that that’s good. It’s good to accept the mistake. And that’s where the power is. And, um, We accepted that we accepted that the mistake was made, uh, and shared the whole process and being very transparent and, um, to remedy it, we really couldn’t do anything, right?
Because these are all online touchpoints. What you lose is the referral traffic that it’s coming from, uh, the sources where the leads are coming from, but we sort of had a backup where in Marquetto we had UTM fields where you can source campaign and. We were storing information in that. So we had some level of data and we need to remedy some of the importing or to backfill some of the reporting views, more general data for that three months, timeframe run.
Some of that reporting we could have created. Uh, touch points directly inBizible as a backfill based on whatever the market on farm fields happen. But we decided not to, uh, just to keep our data clean and reporting team moving forward. So that’s what we did, but I think the biggest lesson for me here was that being transparent with the team, acknowledging your mistake and learning from it was really, really important.
Um, so. Yeah, I definitely feel strongly that taking ownership. Um, it’s something that almost everyone that I’ve had on the show has touched on in their own sessions is that big moment of growth is, you know, taking ownership and, uh, not to being things under the rug. You also mentioned learning from your mistake, which is the entire reason why we run this show.
Uh, so tell me a little bit more about what you learned and maybe how you’ve implemented some personal. The process changes after your MIPS? Yeah. I think two things that I’ve learned from here as one, which is more tactical is. Anytime you want to copy-paste any code from one system to another, especially if you want to go through a project management tool, X it’s acceptable, use a text editor like notepad or texted say that as a text file and then upload and share it.
It’s much more safer way to share that. Uh, the second thing is. QA. I think that’s what it goes down to is that we just didn’t see way the change. Um, so it’s just very, very important to make sure that you QA any changes that happen. And that’s something that. Built in my process, moving forward, any changes that you make, having a QA process next to it has been really helpful.
There are some technologies that I looked into after that, to us, that sort of do that QA check for you, uh, through and through, um, for your processes as things change. So I think that’s, that’s really good. Next step with technologies that are coming up with stuff. Yeah. I think I know a number of people who’ve been guilty of.
Oh, it was simple. It was a simple change in this thing takes forever to propagate. I know Bizible touch points. Um, personally I’ve found them to be occasionally a little slow, but I think reiterating the importance of QA is always a good learning experience. Also. Great tip on attaching the text file. I’m not sure I would’ve thought of that, but it definitely makes a lot of sense, um, as a safe way to pass that over and ensure that.
Copy and paste errors aren’t made. I know I recently was working with my web team on deploying a C name entry, and they said the same thing. Like, can you send me the file? I don’t want to miss a character and totally mess up the C name entry, um, or something like that. Yep. So final thoughts. Um, what would you say to somebody who has made a similar mistake?
It’s okay. That’s, that’s what I would say that, uh, if you’ve made that mistake, just learn from it just don’t make that mistake again. And I think that’s, that’s a good observation for team culture too. Is that when those mistakes happen? How do you react? How does your team react? How does your team support?
That’s that’s just a very good way to even sets out the culture of the team to us that, and I’m going to potential here, but it’s very, very important today is that are you in the right culture are missing. Acceptable and they should be, but they should not be repeated same mistakes, but they should definitely be acceptable.
So that’s why I would say it’s okay. I love you talking about it from the perspective of the team culture. I know as a hiring manager, I often ask candidates to tell me about a mistake they’ve made. Um, one to talk about learnings and two to. Express that they haven’t been able to learn from those. But I think as a candidate, the next time I’m in the candidate.
I’m curious to ask kind of the inverse. Tell me about a time, somebody on your team or recent times, somebody on your team made a mistake and how that was approached and how you handled it. I think it says so much about the culture and I’m going to start asking that if I’m the next time I’m looking for opportunities.
I did exactly that, and that was very enlightening for me. I love that. Yeah. I think that’s a, such a good question. I’ve got them, I’ve seen one or two people. I’ve heard one of them. He was like, I’ve never made a mistake. And I was like,
um, and I think similarly, if a team is, oh, no one on our team makes mistakes. I would be. That would definitely give me some pause as a candidate. Right. Um, what other advice do you have for fellow marketing and revenue ops professionals? Yeah, I, I’m just going to go back to mistakes and I’m just going to just one that is that mistakes happen.
They were built to make us stronger and better. That’s their whole purpose. So don’t shy away from time trying something. In our case, we were trying to be creative by having a more tokenized approach for our tracking. It was a new, creative way to do it, and we were successful, but we made mistakes. We learned from it.
But if you haven’t tried a new way, we might not have made these mistakes. So make mistakes. That’s, that’s certainly the key takeaway and, um, try something new. And I just end up with, it’s going to end up with. Uh, the song that’s in my head by Kelly Clarkson as what doesn’t kill you makes you stronger and I I’ll appreciate it, Jessica, if he was singing along with me, um, I don’t think our listeners would appreciate hearing my singing.
They appreciate mine, but I think let’s do it together. 1, 2, 3. What doesn’t kill you makes you stronger.
That’s such a good, uh, piece of advice. I think. Um, you know, all across the board, the smartest people I know in this space have learned from mistakes they’ve made and it becomes when they’re trying new creative and innovative approaches to things. And, uh, if you aren’t making the occasional mistake, you’re probably not pushing the limits of your own professional creativity.
Yes. Yes. That’s perfect. Any last words of wisdom or anything else you want to share with the audience? I think that’s it. Um, I think we shared and, uh, it was just great to share this mistake. It’s just, it’s just always been learning with all the mistakes, so, well, thank you so much for your time and thanks everyone for joining us for another episode.