Speaker 0: We’ll wait till we get to eleven here, a couple seconds.

Speaker 1: I’ll share my screen now.

Speaker 0: And we’re at the time here. So hello, everybody. Welcome. Super excited to have you joining us today. I am Kirk from Sercante, and I’ll be moderating today’s session. Uh, yes. These sessions are going to be record recorded so you can see it after we’re completed here. Uh, we will have time at the end. So if you have questions for Kira, uh, we can certainly ask them at the end here. Now let’s go ahead and get started. I’d like to introduce you to our speaker, Kira, who has an awesome session ready today. It’s all about five strategies for a cleaner account engagement database in preparation for AI.

Speaker 1: Cool. Thank you. Thank you for the introduction. Um, you kind of on the bit that I was about to do, but yeah. So hello, everyone. My name is Kiera Hodgson, and I’m a marketing automation consultant at Nebula Consulting. And today, I’m gonna go through five strategies for cleaner account engagement database in preparation for for AI. That was quite a wordy title, but we are gonna break that down today, um, during this presentation. But first and foremost, I just wanna say thank you to our incredible sponsors that makes this great event happen. So there’s great content happening over today and tomorrow, and I’m really excited to see the other talks happening. And a big shout out as well to the Maa Dreaming organizers giving me the opportunity to speak today. And thank you to you guys as well for taking time out of your days to join this session. So to start off, I just wanna have a look at a objective of this presentation today. So by the end of the presentation, you’re gonna be able to understand the importance of data. So understanding why data is important, um, in relation to AI but as a whole, and have the tools to implement five strategies to clean your database. So these are five things that you can do today straight after this call. You have the tools at your disposal that you can implement with the overall objective to be better prepared for the future of AI. Now AI is a big conversation, um, happening at the moment, not even in just our industry, but across industries and how the introduction of chat g p t and generative AI is really changing the landscape of how we work. And, um, the point of this talk today is to really look at the relationship between how data and AI, um, link. But first, the big question, what does clean database have to do with AI? And I’m just gonna go through a few points here. So the accuracy of AI models rely on data to learn those patterns, um, predictions. Personalization. As marketers, we know how important it is to create personalized experiences for our prospects, And data about those prospects is essential for tailoring those experiences. Predictive analysis relies on data to make those accurate predictions and recommendations. So the more accurate your data is about your prospects, the more accurate those predictions are gonna be. And segmentation. So segmentation relies on data to ensure segments are being accurately defined. But don’t just take my word for it. There was a report done by Salesforce that I really think articulates exactly the relationship between data and AI, and it quotes the CEO of AI at Salesforce, so Clara Shee. And this quote really represents what the relationship is. So data is the fuel for AI. Without data, AI would not have that foundation to do all of the amazing things that it can do, And it goes back to the quality of that data. So without high quality trusted data, they’re not gonna have the results that you expect to get with these AI tools. And another thing from this report that took took my attention was that 67% of marketers say that their company data is not actually properly set up for generative AI, but a similar number of 63% saying that trusted customer data is important for the success of generative AI at work. So here you can see a, um, a similar amount number of people who are saying that data is really, really important for AI to work, but then equally putting their hands up and saying we’re not ready for AI to start using because our systems aren’t prepared for it. So that’s what I’m hoping I can address today during this presentation. So I’m gonna be going through a five step strategy that we’re gonna look at different points of your data and how you’re handling it to follow best practices to get you better prepared. So first, we’re gonna be looking at the data capture. So this is how you are capturing data and entering it into your system and looking at best practices on how to do this. We’re then gonna be looking at governance, which is might not be the most exciting topic, but it’s equally as important when it comes to handling data. So asking those important questions to see if you are being compliant with your data handling. We’re then gonna be looking at a validation process. So, um, if you have validation in all process or if not, how you can steps that you can do to start creating one. So looking at ways to start capturing invalid data and defining what data that you want to enter the system. We’re then gonna summarize this altogether and look how we can leverage automation tools in account engagement, to start doing the work for us when it comes to actioning that validation process and making sure that our data is cleaning in the background. And then finally, we’re gonna be looking at how to educate your team. So bringing all of this together, it’s really great to have a strategy in place to get ready for AI. But if you’re not spreading that word about why we’re doing this and educating your teams when it comes to data, it really makes a difference when it comes to getting ready for AI. So first, looking at data capture. So with this, I’m looking at when I refer to data capture, I’m thinking about your forms. So this is the first, um, touch point that your prospects have when it comes to entering data into your database. And what I’m gonna be going through here is looking at examples how to make your forms as user friendly as possible to ensure that your prospects are submitting the right type of data that you want to be submitted. So on the example in the screen here, we’re gonna just have a look at the descriptive placeholder text. So in addition to having the field label, we have a bit of information about the field and what sort of information we require from the prospect. So this is really handy when it comes to defining what you, um, what formats that you require for the field, what explicit information. You’re letting the prospect know without them having to guess. So an example of this is using, um, the placeholder text for the format required. So in this example, we can look at the contact number. So if a prospect encountered this form without this placeholder text, they would be unaware that as a company, you want them to use their country code. But if you apply that in the placeholder text, they can see here that, oh, they want the place up. They want the country code. So you’re kind of helping the prospect submit that correct data before it even enters the system. And to echo that again, having the additional information evident on the form, so leveraging that below copy just to indicate any other things about the form. So letting them know that the asterisk means that those fields are required and that it could they cannot be left blank. So here, what you’re really trying to do is encourage prospects to submit that correct data that you want as a company before it even enters the system and really leveraging your UX here, your user experience for your prospects is gonna maximize those efforts. And we can take this a step further by, um, doing advanced form styling, leveraging JavaScript. So in this example, very slowly, you can see here when a value gets entered into this field, the first letter of the value is being capitalized. And the reason for this is that we are follow we’re making sure that the data that’s being entered is being validated the standard that you want. And this is really handy when it comes to things like personalization. So using your merge merge tags when it comes to your emails or other assets that when you are referencing the prospect’s data of their first and last name, you’re confident to know that it’s in the right format that you want it to be intended. Rather than if a prospect was filling out a form, they might be doing it very quickly. They might not bother to capitalize last name. But when it comes to creating that personalized experience for your prospect, these little details become really effective in how your message is being delivered. So I’ve just spoken about how you can tailor your So I’ve just spoken about how you can tailor your forms to help your prospects submit data, but there’s also things that you can do on the back end to protect your forms against bots. Now as marketers, we know bots are the biggest enemies when it comes to our database. And if you do these steps to minimize the amount of bots activity and data that’s coming in, it would really help when it comes to AI looking at your database. And with this, I’m referring to enabling recapture on your forms. So doing this prevents that invalid data entering your system. Um, then going back to the slide from the Salesforce report referencing the inaccurate data that goes into these systems, the inaccurate results. And we don’t want bot data bloating out our database, feeding these AI tools, and giving us information that don’t represent the real prospects, the real people that you want to market to. So having these things in place really minimizes that risk. So for the second point, I’m gonna be talking about governance. So it’s not as exciting as the prettiness of our forms, but equally as important when it comes to handling data. So I’ve got a series of questions now that I’m gonna ask you. Obviously, you can’t reply, but it’d be great for you to consider these and really see if you know the questions to them right now. So if I was to ask you right now, could you tell me who is responsible for data ownership in your team? And the reason why I ask this is because there should be an idea of who has the ownership and accountability when it comes to handling data. And this doesn’t necessarily mean it has to be a single single person’s responsibility. I believe that this is something that all the team should be aware of. But if any issues or any questions come about around data handling or how that your processes are, there should be a clear direction who is responsible in your team. Another question is how is data being integrated in from different platforms? So we’ve looked at how data is being inputted by the prospects through your forms, but that’s not always the only source of data. We could have form handlers, um, from external forms submitting data back into the system through imports, third party, um, platforms that are pulling data through. And all of these data points, they should be able to align with a certain quality and integration process. So all the efforts that you’re doing for your form, that’s also being mirrored for the other data input. Another question. Are you keeping explicit consent for data collection and processing? And you can probably tell what I’m alluding to here, but are you being compliant? And with GDPR, this is essential for every single business. You do have a certain responsibility when it comes to handling prospect data, and being able to have a audit trail of explicit consent from your prospects is really, really important. And this doesn’t necessarily just have to be if the prospect is opting in and out, but also having custom fields to say if they’re eligible for marketing or the time and date that they gave that consent. That additional information makes you even more compliant, um, and having an audit track record is really, really important. And then finally, how long are you storing data for? It is sometimes very hard to say goodbye to prospects, but having a data retention and deletion policy is really important when it comes to making sure that your database, again, similar to the bots, it’s not having inactive data that is skewing your AI tools. So if you have done your reengagement campaigns, you’ve done everything as marked as you can to try and reengage inactive data and they’re still not budging, it might be time to say goodbye because we don’t want that data to then be feeding into these AI tools, and not giving you the results of your active real prospects. So for this third point, I’m gonna be looking at the validation process. So I’ve broken this down into three steps that you can do to start to identify, um, your data and start to have a validation process. So the first one I’m referring to, a data dictionary, and this doesn’t necessarily have to be the term term a data dictionary, but I really like this term. And it’s something that’s come about with the increase of data cloud, customer data platforms that there’s a real attention around having a track record of what data that you hold as a company. And your data dictionary will consist of things as anything related to data. So your fields, the values that relate to that fields, the formats, and also things like the source of truth when it comes to sync behavior. So knowing all about the information of the fields and having that source of truth of where you go to when you have a question about data is a really good starting point. Because once you have that information, you can then start to capture invalid values that don’t align with what you’ve defined. So now you have clearly defined exactly what data that you accept and want as company, you can now start to see the invalid values that don’t align with this coming into the system. But then finally, what is equally important is making sure that you have this consistently checked. It’s all well and good having this information, having a resource, being able to capture those invalid values. But if that’s not being regularly checked to see if that incorrect data is coming in, there’s there’s no action from this, and it goes back to the governance. This could be something that aligns with the person’s responsibility when it comes to data to be able to check this the values that are coming through to the system. And you might be wondering, we’ve now captured those invalid values. So what can we do? And that’s where we start to talk about leveraging the automation tools to really maximize those efforts of cleaning that data. So you’ve started to now gather those invalid values that are entering your system. Um, and I’m gonna run through an example that you can do leveraging dynamic lists and engage and the engagement studio to clean that data and do that process on your behalf. Um, and for this example, I’m gonna be using the country field. And the reason why I’m using the country field is because it can be an important field that has other dependencies on. So, for example, your lead assignment might be based of country to allocate leads. So it’s essential that the value that is being submitted into the system in lies with the value used for those other processes. So we have a prospect who submits a form, and this form doesn’t necessarily have to be a account engagement form. I actually think it’s this process works well when a prospect might submit a external form, and then that data is being fed back into the system using form handlers. And the reason why this is a good example is because we can’t always control what format the data is being sent back into the system from form handler submissions. So having this process in place to capture that is really gonna help clean that data. So they submit a form, and with this form submission, a field value is entered. And going back to that country example, let’s say they submitted UK, so an abbreviation of The United Kingdom. But you’ve defined in your data dictionary that as a company, you only accept United Kingdom as a as a, um, valid value. So So what you’re going to do now is create dynamic lists that will match that value when that gets entered into the system. And what the dynamic list is gonna do is have the rule logic of the prospect field. So the prospect default field country is incorrect values, and this doesn’t necessarily have to be that this one example. If you’re noticing that people are submitting UK, Great Britain, GB as that, um, same value for United Kingdom, you can list them here. And this doesn’t necessarily just have to be for The United Kingdom value, but you could create another dynamic list that looks for The United States. So prospects might be submitting US and America, but you want United States, so you would create another dynamic list to capture those incorrect values. So once you have created these dynamic lists, they’re then gonna fill feed an engagement program. And what this engagement program is gonna do, it start it’s gonna start automating correcting those values. So once these prospects have entered the program, there’s gonna be a rule node that’s gonna evaluate what pros what member of list is prospects. So are they a member of The UK list? Are they a member of The United States list? And then if they match that list, we can then have an action node that is gonna change whatever the incorrect value is to The United Kingdom. So what we’re doing here is that we are creating a process where when data is being, um, incorrectly submitted into the system, we’re capturing it, we’re segmenting it with dynamic list, but then we’re also feeding it into a program that can change that, um, value for us and then feed into the other processes. So if you had a lead assignment that’s looking at The United Kingdom, even though the prospect has actually submitted that incorrect value, by the time they’ve run through this program, they can that will then match the next lead assignment process. So there isn’t an error in the, um, the way that your business is operating from an incorrect value. But one consideration for this is making sure that the program is repeatable. And the reason why this is important and could cut you out is because prospects might make this error a few times. You might have other data entry points that aren’t following the same system. And if there is a scenario where that prospect has that value submitted again, you want them to be able to match this program so it gets corrected. And then finally, we’ve spoken about all of these things that you can do to really help clean your data in your system, but I think there’s no point doing all of this if you’re not spreading the word of why you’re implementing this strategy, why AI is exciting, how that’s gonna change, and to do that is to educate your teams. So one way of doing that is having documentation and resources. So this is like our data dictionary. Doesn’t just have to be a data dictionary. It could be procedure docs for imports. It could be best practices, user guides, but just having a resource that your team can go to when it comes to handling data. You’re giving them the stepping stones to make sure that they’re following things properly, and you’re following those best practices. The next point is having role specific training sessions. So depending on who the person is in your team, they’re gonna have a different relationship when it comes to data. For example, your sales team will have a different relationship to your marketers, and being able to identify this and accommodate your training sessions for your team will really enhance their engagement and empower them to really take on board what you’re saying and then follow these best practices. And then finally, I really like this term, um, having a culture of accountability. So going back to the governance, there might be somebody who’s more of an expert when it comes to data, but this doesn’t mean the responsibility stays with them. It is a whole team effort when it comes to handling data and making sure that everyone knows what what role they play in this strategy really leverages the success of being able to clean your database and be better prepared for AI. So to summary, we have all of these really exciting AI tools that are coming in the future, but they all have one thing in common. Well, probably more than one thing in common, but for this instance is that they are all needs data to operate. And depending on what data that you’re putting into the system is really gonna depend on how they can really leverage your marketing and success as a business. So I hope that you have some takeaways from this session today, and I really appreciate everyone joining this in your busy day. Um, and, yep, we definitely do have time for questions. So thank you again for joining.

Speaker 0: Thank you, Kiara. That was a great session. And like you said, we do have time for some questions. One that just popped in the chat, uh, was can we create any engagement studio rules or functions to take care of duplicate prospects in account engagement?

Speaker 1: That is a very good question. However, it kinda differs from what we’re talking about today in terms of correcting values. Um, and the top of my head is probably definitely can doable for the engagement studio. It has many capabilities that go beyond just email nurtures, but it would require a little bit more development. But with this example, we’re really looking at how we can update values rather than duplication of prospects.

Speaker 0: Thank you, Kira. Another question that came in. How frequently should you update your data dictionary?

Speaker 1: I would say that you would wanna update it anytime there is an update to your data. You never want it to be outdated. And with data, it is some things that is constantly changing. You might have constantly have new fields that are being added to your forms, and being able to align that back to your date and dictionary is keeping it most up to date. So whenever there’s a change, um, and that doesn’t necessarily mean that solely when you would need to update it, and also having maybe monthly reviews just to make sure that there’s nothing, um, that that’s misaligning, uh, and also encourage your other team members to review your work as well, I think, is really important.

Speaker 0: Yeah. I think that makes sense. Thank you, Kira. Another question that came in from Jeff. Could we create a JavaScript to standardize phone numbers?

Speaker 1: Potentially. I I guess I I think with the dev work, I’m not a developer, so I couldn’t give you a yes or no answer here, but, um, JavaScript is really powerful. So that if with further development, I’m sure there might be a way to do that.

Speaker 0: Yeah. Any other questions? Just drop them in the chat here, and Kira will answer them to the best of her ability. We have five minutes left. And give a couple minutes here. Okay. One more just came in from Kate. What is the best practice time frame for cleaning up inactive prospects, and how long do we let them be inactive before we clean them up?

Speaker 1: I think that really depends on the business, really, um, and aligning with your selling cycles as well. So if you have subscriptions, yearly subscriptions, aligning it with that. But once you identify who is inactive, maybe using a six month time frame, quarterly time frame, depending on your business cycle again, you can then identify them, um, and start to I really recommend engagement pro reengagement programs where you can reach out to your prospects, see if they’re interested in giving them the opportunity to engage again. It’s a really good starting point, and then that should give you the information and the reporting to start actioning the cleanup.

Speaker 0: Yeah. Definitely. I always put them on a dynamic list. Right? That way, you can always actively review that and have that as recipient list going forward. No. Totally agree with that.

Speaker 1: Alright.

Speaker 0: We’ll wait a couple more minutes here, see if anybody else has any questions. Again, thank you everybody for joining us. Thank you. Yeah. You’re welcome, everybody. Alright. I think that probably wraps it up unless another question comes in as we’re wrapping up here. Uh, that concludes today’s session. Again, thank you, Kira. Thank you to our sponsors. Without them, none of this would be possible. This will shut off if we don’t end before the three minutes automatically. Make sure you check up the agenda for your next session. And remember, this is recorded, so you can always check it out on demand at some point. Thank you, everybody, and enjoy the rest of your time. Take care.

Speaker 1: Thank you.

Speaker 0: Bye. Oh, Kira, we just get one more. Oh, she left. RJ, I’ll make sure oh, Kira. Oh, we got it. I saw it. We’re good if everybody’s here. Um, Regarding the the data forms, an example was The UK to United Kingdom. Are there histories of those changes for troubleshooting incorrect logic?

Speaker 1: I think it it I think it depending on what field that you’re gonna be targeting, but I think using the country field, you can kinda gauge which those incorrect values are. Um, so with UK abbreviations, but you could also have fields that are industry, for example, that you might have a certain format that you could but I think being able you need to be able to identify these changes to start creating the process. And, yes, Zoe. It’s correct. Thank you, Zoe.

Speaker 0: Alright. You’re welcome, RJ. Alright. We’re gonna try this one more time. Everybody have a great day, and, uh, hope you enjoyed the session. Take care, everybody. Thank you. Bye.