How to Run a Data-Driven Business with RJMetrics

Nasty Gal built a brand of fiercely loyal customers (the kind that spend seven minutes on a site per day!), Zipcar slashed customer acquisition costs from $150 to $50, and Sticker Mule cut production time in half. You’ve heard stories of companies turning data into business growth, now you’re going to learn how you can do it too.

Join RJMetrics analysts Taryn Cooper and Shaun McAvinney to learn how new and growing companies are using RJMetrics to tap into revenue-generating insights.

You’ll Learn:

  • How to sort through the noise in your data and find the metrics that matter most
  • The 3 approaches to data integration (and why it’s so critical)
  • What advanced analytics like cohort analysis, churn analysis and marketing ROI can reveal about your business
  • Case studies from growing companies on how they’re using RJMetrics to make smarter decisions


Shaun McAvinney: Hey, everyone. Thanks so much for joining us today for our event on running a data-driven business. Before we jump into the material, I just have a few housekeeping notes I'd like to cover. First, we will have a Q&A session at the end of today's presentation. If you have any questions at any point, please feel free to submit them via the chat window. We're also recording this event today and we're going to make sure that everyone gets that within 24 hours of the event. You'll receive this in a follow up email.

Also, just as a quick introduction, I'm Shaun McAvinney, and I'm a Senior Sales Engineer here at RJMetrics. What that means is that I spend most of my day talking to ecommerce marketers, SaaS company founders, analysts, CEOs and COOs about their data infrastructure. My role in today's event will be to talk to you about the tech you need to build a data-driven business.

Taryn Cooper: Hi, everyone. I'm Taryn Cooper and I'm an account manager at RJMetrics. I spend my days working with the individuals responsible for making decisions with data. My role in today's event will be to talk to you about best practices when it comes to making decisions with data, and to share some stories and examples of companies doing this really well, as well as some areas where I see a lot of companies making missteps.

Shaun: Great. Just so you know what you're getting into today, RJMetrics CloudBI is an analytics platform for online businesses. Our platform connects to data sources you already use, like Salesforce, MongoDB, Facebook ads, Zendesk, etc., and consolidates it into a central data warehouse where you analyze it using our chart builder interface. All the things you'll be learning on today's event are relevant for pretty much anyone, RJMetrics customer or not. But if you're interested in learning specifically how RJMetrics can help you do some of the things we're covering today, we'll be showing a quick five-minute demo of the product at the end.

Outside of learning about RJMetrics, you should leave this event with some practical tips that you can put to work today to get more value out of your data. We're going to focus on three core areas today: the metrics you need to run your business and how to identify the metrics that are just right for you, how to get the data you need to monitor your key metrics and answer the questions you have and lastly, how to get started using data, from the basics like tracking KPIs overtime to getting into some of the more advanced analytics. Along the way, we're going to be sharing examples of stories of how small growing and at-scale companies are using data to improve their business results. With that, I'm going to hand this over to Taryn to get us started.

Taryn: Thanks, Shaun. I want to get started today by sharing some guidance on how to identify the metrics you need to run your business. I'll frame this problem with a quick story. A few months ago, I was working with a new RJMetrics client who was just starting out using the product. This is a fairly new ecommerce company, and prior to RJMetrics their access to data was limited to Google Analytics, the limited offering by Shopify, and then occasional data dump from an engineer asked to run a bunch of sequel queries.

On my first call with them after receiving access to their data, they were so excited. It's really powerful for companies who've been in the dark to finally be able to see their business in a new light. They had so many charts they wanted to build: revenue by acquisition source, percent of revenue from repeat purchasers by month, sales by product category. Their first instinct was to start building charts in mass, and that's a really common response. But it's rarely the best one.

I started asking them what problems they're facing in their business. After a bit of discussion, it was clear that there was shared concern around customer acquisition channels working well to bring customers in, but a general fear that these customers weren't ever coming back. Okay. So now we have a starting place. The metric that was most important for their business at this point was number of purchases per customer.

It can be really tempting to just start measuring things once you've got the data, and there's also a lot of advice about things that you should be tracking. For example, "Lean Analytics" has a great framework for how different business models should think about measuring their business. Then, there are the famous Pirate Metrics recommended by Dave McClure, and of course there are tons of blog posts and guides out there for which metrics investors want to see. But if you follow all of the shoulds, you'll spend a lot of time building charts and not a lot of time making decisions.

Less is more, and that's the most important thing, using data to make decisions. Our recommendation based on what we see our clients having the most success doing is to focus on a small number of your most important metrics. So let me talk you through a story of how this work for one of our clients, Sticker Mule. First, I'll give you a quick plug for Sticker Mule. They print custom stickers that are really beautiful. We get our promotional stickers from them, and their work is impeccable. But that didn't just happen. They're constantly using data to improve all parts of their business, from customer acquisition to quality and deliverability. Their CEO followed the "focus on a small number of core metrics" mantra from the very beginning.

When they first signed up for RJMetrics, back in 2012, they were still very new. Anthony and the team had a single-minded focus on sales. So they tracked three metrics, revenue, year-over-year growth, and customer acquisition cost. But things change as your company grows. Only once the business started gaining traction did the Sticker Mule team expand beyond their initial small core group of sales metrics. First, they added manufacturing metrics. But even then, they limited it to only the two most important ones, quality and turnaround time.

Then, as they started adding marketing and customer success team members, they added metrics there as well. The idea was to introduce new metrics relevant to the new roles they were bringing on, but to still keep the team as a whole focused on the top two to three core metrics that never change. Well, one of my favorite aspects of the Sticker Mule story is how they've reduced turnaround time.

RJMetrics makes it very easy to share data across the whole company, and Anthony embraces this transparency. Everyone in the company from the CEO to team members working in production have access to performance data. The manufacturing team was constantly coming up with ideas on how to improve processes, and overtime they were able to cut turnaround time in half.

So, that's incredible. This wasn't some executive mandate. It just speaks to the power of giving individuals access to data on their own performance. What does it actually take to implement a process like this? A tool like RJMetrics certainly isn't the only option out there. So at this point, I'm going to hand it over to Shaun to talk through the different ways companies are building out the data infrastructure it takes to run your business with data.

Shaun: Thanks, Taryn. So, to use data to run your business in the way that Sticker Mule is doing, you need to have all of your data in one place. There are really two reasons for that. First, if you're using tools like Salesforce, Shopify, or even Google AdWords, your reporting capabilities are limited to how those platforms choose to display data.

For example, our marketing team here uses Pardot to run marketing automation and metrics. Pardot has some pretty handy reports. These are great for checking high-level metrics. But if you want to answer a question like, "What's our average email click-through rate," you're kind of out of luck. Or you're going to spend your entire weekend downloading spreadsheets from Pardot and aggregating the data. That's not a lot of fun.

Second, the most powerful business insights often exist across platforms. For example, one of Sticker Mule's key metrics was customer acquisition costs. If you're only running a few campaigns on Google AdWords, then answering that question is pretty easy. But as soon as you're advertising on multiple platforms and running a variety of campaigns, then tracking that question becomes pretty painful. Most companies will settle for answering that question once a month because the cost to analyze the data is just too high.

So let's talk about how to actually make this happen. You essentially have three options. First, you can absolutely use spreadsheets to combine your data. This is a valid choice and works very well for a lot of companies in their early stages. But eventually, you're going to outgrow this. If you want your CAC numbers daily, you're not going to spend one hour a day consolidating spreadsheets. It's just a bad use of your time. Sure, you could hire someone to do it for you, but that's very pricey.

The second option is to build your own data stack, and what's included in a data stack would be three parts. First, you'll need to build data integrations that will connect to your existing data sources, like AdWords, Google Analytics, your back end database. Then, move that data to part two of your stack, which is a centralized data warehouse. That could be MongoDB or Amazon Redshift, or really any other analytical warehouse. Then, three, you'll need a way to analyze that data.

This approach makes a fair amount of sense for companies that are well-established or have a growing data team. But a lot of companies just aren't ready to make that kind of investment in their data infrastructure. Years ago, building your own data stack was the only option. But luckily, that's not the case anymore. Thanks to option three, you can just buy a complete top platform that takes care of all this for you.

So, let me just clarify that the first option is kind of terrible. As soon as you're outside of the very earliest days of your business, you really should be looking for a sustainable solution to the data problem. The other two options are good and will just depend on the stage your business is in. So, either way, the end result is that you have all of your data in one place, and can spend far less time building reports and more time answering questions.

One of the most common ways our clients use RJMetrics is to analyze marketing ROI, which is what our client Bilna does. When you're advertising at the level they are, they've got four different AdWords accounts, you have to have your data consolidated. But then, before I hand it back to Taryn I just wanted to mention one final benefit of using a tool like RJMetrics. It's a single source of truth. For example, one of the most common themes I hear from prospects is frustration about a question that should be pretty simple, "How much money did we make last month?" That seems like an easy question, right? But not really. Does your revenue calculation account for returns? Is it before or after tax? Are all of your departments measuring revenue in the same way? Hint, they should be.

The beauty of a tool like RJMetrics is that you can build a revenue metric once, and then anyone from any team can use it knowing the definition will be consistent company-wide. Your merchandising team can explore revenue by product category. Marketing can analyze revenue by channel. You might want to see revenue by different regions of the country, and every single analysis will be talking about revenue in the exact same way.

Here's a great quote from our client, Bevel, on why this is an advantage. "Your team is always making decisions based on data, and you want to make sure that everyone is looking at the same numbers." So that's kind of the basics. You know the fundamentals of what it takes to get your business running with data, from choosing your core metrics to setting up the infrastructure you need to track them. At this point, I'm going to hand it back over to Taryn to talk through some of the more advanced analytics.

Taryn: Thanks, Shaun. Once you have your data monitoring in place, you're ready to get started doing some exploring. One of my favorite things about working with clients at RJMetrics is watching how quickly their analytical capabilities grow. It's amazing how fast people get hooked on answering questions with data. One of the first things clients like to do is set up a dashboard that shares their company performance with their investors.

There was a great article on TechCrunch recently by VC David Teten, "How To Run Your Company Based on Metrics". He says, "I review a lot of board decks with a beautifully hand-crafted page with metrics for the company. That makes me nervous. I prefer to see a screenshot of an internal dashboard, not something created for the board, and I prefer to see that same dashboard in the same format at every meeting. The reason is that the real value of the board deck is not, ironically, for the board. It's for management to track their own performance. So, I want to see that management is using a dashboard every day, not just for board meetings."

This exact use case is one of the primary ways that RJMetrics' clients use the platform. Except, instead of screenshots, you can just give your board direct access to dashboards that they can track in real-time. Here's an example of what a dashboard like this might look like for a SaaS company. You can see here we're getting into some analysis that's beyond your most basic revenue numbers. The advantage here is that you only have to do it once. Once you build your metrics for MRR, ARR, CAC, and churn, then you're set. The dashboards will keep updating automatically.

Measuring marketing performance is another major use case for RJMetrics. We already talked about CAC, but another really critical marketing metric is customer lifetime value, or CLV. CLV is the total revenue from a single customer. It's an incredibly valuable metric to evaluate marketing performance, and in this example we're looking at the average CLV across channels. As you can see, customers acquired through AdWords are significantly more valuable over their lifetime than Facebook or Twitter.

So measuring ROI is one part of measuring marketing performance, but another great thing to analyze here is just acquisition trends. Zipcar, who is not one of our clients, used data to identify the areas where usage was high. Then, they doubled down their marketing in those areas, and it worked. They were able to increase signups in those areas, and you can do similar advanced analytics in RJMetrics. Things like identify the different regions of the country generating the most revenue or customers, and then ramping up your Facebook and AdWords campaigns to those specific areas.

Cohort analysis is a pretty advanced analysis that becomes super-easy when using a tool like RJMetrics. This particular example is so valuable in being able to track performance overtime. As a growing business, you're constantly making improvements. You're adding new product features, you're improving email copy and making changes to your website.

What you want to see is that these aggregated efforts are creating more valuable customers over time. Chubbies has been an RJMetrics client since 2012, and they're a great example, like Sticker, Mule of a small company that embedded data-driven thinking from the beginning. Cohort analysis is advanced stuff, but it's so powerful and there's no reason only the big companies should be able to take advantage of the insights it offers.

Now, this is the second example we're looking at that's using the customer lifetime value for the analysis. Let me just say a little bit more about this one. We do a lot of benchmarking research in the ecommerce industry, and one thing we found is that the top 10% of customers are worth 6 times more than the average customer. That is incredible.

CLV is the metric that can help you identify those customers and make absolutely certain that you don't lose them. Nasty Gal is an ecommerce company that, very similar to Chubbies actually, has built their brand on a foundation of extremely loyal customers. In June of 2012, when Nasty Gal was on the cusp of its explosive growth, half of their sales were coming from 20% of their customers, and this top-of-a metric was translating into other metrics as well.

Twenty-five percent of their customers were visiting their site daily, spending at least 7 minutes there and their most engaged 10% of visitors were on the site more than 100 times per month. All that to say, when you're building a business it's invaluable to know who your most passionate fans are. The ability to identify those customers and analyze the buying patterns unique to them can be a very powerful growth driver.

Churn analysis is another really important metric. If you're a SaaS company or have any kind of recurring revenue model, then this analysis is a must-have. You want to know every month what your customer and revenue churn is. But a lot of our classic ecommerce businesses track churn as well. If you know that your average customer makes a purchase every six months, it's not hard to build a churn metric that will allow you to see how many customers have recently crossed that threshold. Or even more actionable, who the customers are that are nearing that threshold and might need a reminder to come back to your store.

So that wraps up your crash course in advanced analytics. At this point, I'm going to hand it back over to Shaun, and Shaun will walk through a five-minute demo of how you can build some of the charts we just looked at in RJMetrics. After that, we'll be opening up for Q&A. So start asking your questions now, and I can see that we've already had a couple of great ones coming in.

Shaun: Great. Thanks, Taryn. I think everybody should be seeing my screen now. So this is RJMetrics CloudBI. It is a full-stack business intelligence platform. I'm just going to skip the whole part of connecting all the data. Let's assume we've already replicated all the relevant data from various data sources into the data warehouse, and I'll be walking through some of the analysis that many of our clients find valuable.

So, on this dashboard where should I increase spend? What we're looking at here is there's a lot of information here. But what I'm going to really focus on is this one table and this specific row of this table. This is our ad campaign statistics. We can take a look at this first row, Celebrity Testimonial Shares Shirts. You can see we've pulled in all the impressions, all the clicks, all the spend, the new customers, that cost per customer, and the first order of revenue. Really, if you guys are keen Google Analytics users, you probably know that you can get a lot of this, or all of it, in the GA interface currently.

So let's take a look at the ROI for this campaign. Right now, we're spending $381 to acquire a customer, while our first order of revenue is only $55. So that kind of seems like pretty terrible ROI, and it is. But it's because we're only looking at the first purchase. So what we've done in this next column is we've actually calculated the average revenue a buyer brings in in their first 90 days. This is $619, and that's pretty positive ROI. So, by extending the payback period to 90, we're able to uncover that Celebrity Testimonial Shirts' campaign gets us customers that spend a lot of money in repeat purchases.

Now, I'm going to show you guys how we can build a chart like this, or a little bit similar to this, and to do that we're going to open up this Report Builder. This is the main way you'll explore data in RJMetrics. The first thing to do is add a metric, so I'm just going to go down and add that revenue metric. We talked about this a little bit earlier, but one of the key parts of RJMetrics is making sure that everybody is on the same page. So I'm just going to tighten this up to show revenue from the last quarter, and view it by week.

We can see that in the week of November 8th, for instance, we had $187,000 in revenue. One of the first things that many of our clients want to see as soon as they start using RJMetrics is the difference in first-time versus repeat revenue for the company, and we can do that pretty quickly by adding in a filter here. So this is actually a calculated dimension in the data warehouse that just orders each users order. So by setting this to one, we're going to be looking at all first-time revenue here.

In that period, we had $9,000 in first-time revenue. So how does this compare to repeat revenue? I'm just going to duplicate this metric and change this user's order number to be greater than one. Now, in just a couple seconds we can see the comparison of first-time revenue to repeat revenue. So, first-time revenue is pretty low at $9,000. But repeat revenue is doing pretty well for us, $177,000 in repeat revenue for that week.

I can go another level deeper than this. Let's say I want to identify where we acquired all these repeat buyers. So, using this "group by," I can segment out using any demographic, geographic, or acquisition source data, or really any data that you have in your database. I'm going to group this out by the user's UTM campaign. Very quickly, we're looking at the original acquisition source for the users that are spending all this repeat revenue.

We've got this big peach-ish group here, and I bet you guys can see where we're going with this. This is Celebrity Testimonial Shirts again. We already know that they are positive ROI after 90 days. We can also see that they just drive the majority of our repeat revenue. In some cases, visualization might not be the best way to absorb the data. So I'm just going to collapse this chart and bring in revenue back into here. So you can build out some really robust reports in here that would take hours in Excel or in SQL, and you could continue to layer on additional metrics in here to just go deeper into your business.

Cool. So now, to close this out, I kind of want to close this out by highlighting a cohort analysis that is crucial for your business. It really exposes how changes in your signup flow and marketing cadence can have serious implications. So what we're looking at... I'm just going to bring this up full-screen. So this chart is titled "Time to First Order Cohorts." What we're doing here is cohorting or grouping users by the date that they signed up on our marketing site.

What this chart is kind of answering is what percentage of each cohort made at least one purchase in each month following signup. Again, this is a crucial analysis for any company. You can see that our best cohort was the week of October 11th, 2015. So people that signed up in the week of October 18th are our best cohort. You can follow along in each month what percentage of that cohort had made at least one purchase.

Kind of what we're looking for here is to kind of identify that changes that we're making in our signup process, in our marketing cadence, are positive changes. Actually in this Vandelay data set, it looks like the changes we made might actually be negative. So there might have been something we changed in the post-signup form, or the initial marketing cadence that we send out to try to get somebody to make their first purchase that was a negative change. The only really well-done way to figure this out is to cohort your data together so you can identify those positive or negative changes quickly, and then either iterate on it or revert those changes back.

Actually, let me get back to this. So those are just a few of the analyses you can run on top of the data you already have. But you really need it to be consolidated together. So at this point, we're going to open up the Q&A section. We've had a lot of great questions, so let me just start up at the top here. While I'm pulling those up...I think my computer went to sleep. While we're doing the Q&A, I'm going to leave this poll open so you guys can just answer questions, if you'd like us to follow up with you after the event.

So, the first question is, "So, I'm a new business running on Magento. What should I be looking at?" Taryn, I know you have a lot of Magento clients, I'm sure. One thing I did want to plug is we do have a really great blog post about, I think it's the top five analyses that you can't get out of the Magento interface currently. Magento is great. We love Magento here at RJMetrics. They have a pretty standard way to run an online business. But what they don't have is the reporting capabilities.

Because they're so standardized, it's really easy for us to get a client up and running quickly with RJMetrics, and provide a lot of value with that robust data set that's there. Taryn, do you want to add anything to that that you've seen successful for Magento clients?

Taryn: Yeah, absolutely. Shaun, as you mentioned, we have a ton of Magento clients and we see everything from their merchandising team is using it to track inventory and product performance. I have bookkeepers using the Magento data and comparing it to what they see in their financing database as well. So really, we have a lot of experience with Magento and we can definitely help you out with that.

Shaun: Great. Thanks, Taryn. Another question is, "Can you say more about cohort analysis? What can it actually do for me?" I think part of this whole webinar is, it's hard to say just point blank what cohort analysis or really any analysis could do for a person until you discover what your needs are in the company, where your analysis is lacking. Again, we showed those slides earlier about the Pirate Metrics and the various metrics that investors say that you should be looking at.

But one of the things that is really valuable is to identify what you need to be improving today, and then do the analysis that is relevant. But I don't know if...Taryn, do you have anything to throw in there about why cohort analysis is valuable or how it can be used generically?

Taryn: Yeah, absolutely. So, as I think we mentioned earlier, our clients really are focused on showing that their business is improving overtime and that the customer lifetime value is improving as well, so more recent cohorts are performing better than earlier ones. I know that I have an example of a client who recently changed their product offering because as they got to know their customers better, they understood what types of products that the customer is really interested in.

So they found that by making this change to their product offering, they hypothesized that they would eventually have better customers who were spending more money with them overtime. They were able to then show their investors that their original cohorts of early on customers maybe didn't perform as well. But once they were able to really hone in on the types of products to focus on, that they improved their customer lifetime value and they were able to show that through the cohort report.

Shaun: Great. Thanks, Taryn. Let's see. Another question we have here is, "How do you recognize when you're tracking too many metrics? What number of metrics should be considered core?" That's a really great question. I know it's something that I think every company struggles, especially when they get started using BI platform, or the latest and greatest analysis tool. What I found, it really depends on your ability to keep your goals in focus.

The bottom line is that you don't want to be looking at too many numbers and moving the needle on too many areas. If you feel like you're starting to lose focus on your KPIs and objectives, you should probably scale back. You've probably all heard of the term "analysis paralysis." Coming from a very data-driven company, I know that we've experienced that here a couple times ourselves where we just feel like there's too many numbers flying around, and really we've got to scale back and just focus on what actually matters.

So one other question we have is, "How do you know it's time to start expanding your core metrics?" Taryn, do you want to take that and kind of jump in there and answer based on some experience?

Taryn: Sure. Absolutely. So, again, a lot of clients will come in and they want to tackle so many things at the very beginning. Once we're able to scale back and focus on what's most important, then sometimes we have to think about, "Okay. Now, when do we expand and when do we grow that number of metrics that we're looking at?" Really, it's based on the individual client. But typically what we see is that as the team grows and as you add new departments, maybe you're adding sales members, maybe you're adding a marketing team, then you expand the metrics.

So you want to be able to provide those specific teams with metrics that are important to their particular performance while at the same time, continuing to keep the focus on what the core KPIs are for the overall business.

Shaun: Excellent. Great. One other question we have here is, "What are some of the dangers of not having a single source of truth?" Yeah. So, I talk to a lot of companies, as I already said, and one of the big problems is that when you're...let's say, for instance, you're growing a sales team, and you're growing a marketing team at the same time. When there are different teams that have the same goals or kind of the same goals, or hand off on the same goals, you really should be using the same metrics to drive those goals.

For instance, we use Salesforce here, and one of the main goals that we use is opportunities. So, opportunities created for sales reps, and then those sales reps take those opportunities to closing. There are a lot of companies out there that are not using the same metric on their marketing team for opportunities as they are in their sales team. That's a huge issue when one team relies on the other for their leads and for their sales leads, and then the original team relies on their sales team for hitting their conversion goals. If they're not using the same metrics, that can cause serious breakdown in relationships between teams and kind of questions around, "What number is correct?" So, if everybody starts from the same source of truth, there's this implicit trust between the teams that we are all working towards the same goal, because the goal numbers actually reflect the same.

Great. One of the other questions I just wanted to highlight real quick is, "How would we apply this to email marketing? So how would we apply kind of bringing together disparate data sources to email marketing?" One of the dashboards I didn't show that is in the demo relates to this specific question. One of the things that we like to do here, I kind of eluded to it a little bit, is look at lifetime value metrics. So, either CLV, customer lifetime value, or repeat purchase rates, or quarter analysis, things like that.

One of the things we see our clients doing all the time is building out dashboards that identify the conversion rates from first-time purchaser to repeat purchaser, or second-time purchaser to third-time purchaser. What we normally see is the more a person purchases, the more loyal they become, or the more likely they are to make that next purchase. In the same vein, we also filled out charts, or clients can fill out charts that show the average time between those purchases.

Similarly, the more a person purchases, the less time they spend in between those purchases. So they're buying more frequently. So by using those two charts together, you can really identify the best time to reach out to those users for a follow up. Whether you're going to give them an incentive between the first and second purchase to try to get them to become a more and more loyal customer, or just identify the optimal time to reach out to somebody based on their purchasing.

Then, kind of to close the loop on that, you can bring all the data from your email marketing system back into your centralized data warehouse, and run analysis on those campaigns to identify, "Was this actually valuable for me to do? Is this working out for me?" Either by cohort analysis or simply by identifying the campaign and seeing these first-time purchaser groups that I'm sending emails to are really kind of converting more people to become second-time purchasers. Now they're 50% likely to make a third purchase, so on and so forth."

One other question that I want to follow up here that's just pretty cool, "Are you able to join from multiple data sources to generate new data sets?" Absolutely yes. Whether you're doing this in a homegrown system or in a platform like RJMetrics, that's one of the core features that you would need to do. Bring in those disparate data sets. Right now, let's say it's Salesforce, Pardot, Marketo, whatever it is, MailChimp. If those systems are disparate, now the only way to do that is to write code that's middleware to kind of pull from those APIs, hit up your database and then kind of build out one static report, which could take a significant amount of time.

But by replicating the data out of those data systems into a centralized data warehouse, you'll be able to join across those disparate data sets pretty quickly. Taryn, you might have something to throw in here about how often you see clients adding new systems or adding new data sets because they want to get new analysis.

Taryn: Yeah, absolutely. So it's very common for my clients initially to start out with just their main database, and then as time goes on they'll add Google Analytics for example, Google Ecommerce data. They may start bringing in data from their helpdesks, from things like Zendesk or from their CRMs, like Salesforce. So, this is incredibly common. It's incredibly powerful. You're able to actually combine that data into the data warehouse, and then create layered analyses.

So, in the same report you can have metrics coming from all different data sets and be able to segment in ways that you would never think possible before.

Shaun: Yeah, awesome. So, one of the final questions here is, "What are some of the pros and cons of giving company-wide access to your metrics?" I'll hand this one over to Taryn. I think she's probably got the best experience long-term of companies doing this.

Taryn: Absolutely. So, I am a huge advocate of data transparency, and it might make some business owners a little bit uneasy to be so open about their KPIs. But we've seen that overwhelmingly, businesses that share their performance progress open themselves up to more ideas about how to move the needle. They are able to get their teammates to really rally and push in a specific direction when one metric is under-performing. So there's just huge benefits there to sharing some charts publicly, and then also keeping some private.

So, you don't want every team member necessarily to have access to everything. But we find that keeping the main core KPIs public and open throughout the organization really is motivating to the teammates.

Shaun: Okay, great. That pretty much wraps it up for our webinar today. I just want to thank you all so much for joining us today about learning how to run a data-driven business. I hope that this was educational and fun. I wanted to also thank our team here so much for all the hard work they did getting us up and running. So for me and Taryn Cooper, I'm Shaun McAvinney, thank you so much for joining.