Data-Driven Work Cultures: Keith Perhac of SegMetrics On How To Effectively Leverage Data To Take Your Company To The Next Level
An Interview With Pierre Brunelle
Measure key performance indicators at a lifetime scale. — Data is reality, knowing that reality lets you forecast out. Knowing ads are bringing in two dollars for every one we spend in the first month is great information to have, but what does that look like long term? Being able to forecast not just the short term return on your efforts, but projecting the long term return is even more important. When you’re looking at acquisition costs, it might cost $100 to acquire a customer, but it may take three months to recoup that investment. You need to be able to forecast that out and project how long it will take to recover the costs on that investment. If you’re spending too much, and it will be eight months before those costs are recouped and the company goes broke in the meantime, that’s not helpful. At the same time if you know you’re going to get that money back you can afford to spend more up front in order to acquire that customer.
As part of our series about “How To Effectively Leverage Data To Take Your Company To The Next Level”, I had the pleasure of interviewing Keith Perhac.
Keith Perhac, the founder of SegMetrics, a digital marketing expert, software entrepreneur and author of Building Marketing Funnels That Convert, A 90 Minute Guide. After growing up in the states, he headed to Japan as a salaryman at a Japanese IT company.. In 2010 he started his own consultancy 2010 to work with startups and digital marketers looking to grow quickly, and moved back to the US in 2016. He founded SegMetrics, an analytics tool that lets you see revenue from the perspective of each touchpoint in your marketing funnel — thereby knowing true points of conversion, ROAS, and more. Working with hundreds of companies, Keith has expert insight into what data marketers of e-commerce brands, digital agencies, online course creators, and SaaS companies need to be measuring in order to authentically convert customers and add to the bottom line.
Thank you so much for joining us in this interview series. Before we dive in, our readers would love to “get to know you” a bit better. Can you tell us a bit about your ‘backstory’ and how you got started?
After I graduated from college, I moved to Japan and became a salaryman, which means I was working from 9 a.m. to about 11 p.m. every day, and I did that for around 6 years. As the company I was working for was in the process of being acquired, I realized I didn’t want to be keeping up such long hours — especially if I wasn’t getting overtime for all those extra hours. My background is in design and development, and back in 2002 when I graduated, that was an anomaly. I decided to go off on my own and everyone thought I was crazy for it. Then came the dot-com boom, and the number one thing the market needed was developers that had any design sense whatsoever. My skillset was in high demand, so I felt rather lucky.
I used that skillset to quit the company when they got acquired. I decided to go off on my own doing conversion rate optimization specifically in the info product space. My first job was with Ramit Sethi at I Will Teach You To Be Rich, and it was just trial by fire. It was a very small team of six at the beginning. I worked with Sethi for a number of years as a freelancer. As I was freelancing with him I was learning so much and at the same time I was using my own marketing knowledge to build up an agency. I had so much work that I hired someone to help, then a second person, then a third and forth, and before I knew it we had a twelve person team. I was giving speeches at different conferences in Japan, hosting workshops, had clients spread out all over the world and it was larger than life.
In that time, we started having issues getting data. With multiple different systems that don’t talk to each other, we were having a hard time getting a global picture of all of our data. Exporting a ton of data from lots of sources was really clunky and inefficient. As an agency, one of the biggest hurdles we deal with is proving value to our clients. In order to be able to prove to our clients that we were doing a good job, we were spending an enormous amount of time pulling spreadsheets, connecting data, and the like. It was 2014, and there were no tools in existence to aggregate all of our data into one spot, so we decided we would build our own tool internally.
Luckily, we had a month where we didn’t have much client work and decided we’d use that time to build our own SaaS. We had always wanted to and now was our opportunity, we are technical people with a marketing background and we built the whole thing in just two weeks. From top to bottom, two weeks out the door. The first version just spit out a text file, but we were able to show it to our clients and they thought it was amazing. We started using it internally and pretty soon our clients were asking for access to it, and the software became bigger than our marketing agency. Eventually, we shut down the agency to pivot towards the software, which we were all very excited about. The migration to SegMetrics began in 2019, we joined the Tiny Seed Accelerator in 2020, and the company has continued to innovate and grow leaps and bounds ever since.
Can you share a story about the funniest mistake you made when you were first starting? Can you tell us what lessons or ‘take aways’ you learned from that?
On one of my first jobs, I was tasked with creating a popup that would appear when a new person came to the website. The goal of the project was pretty straightforward, when people came to the website there would be a pop up that said, “Hi! Welcome to (insert company name here), you should sign up for our newsletter!” They would then click yes or cancel and it took them to the main site. Easy enough. I was so nervous I tested every facet of this pop up on our test server for two days to make sure there were absolutely no bugs. I spoke to our internal tech team and had them test it as well to make sure there were no issues. I was living in Japan at the time, and my client was in the U.S. I set it up to go live at 11 p.m., so those in the United States were seeing it mid-morning and then I went to bed. About three hours later I woke up with a very uneasy feeling. I checked my phone and had a dozen missed phone calls. No one knew that the live site had a caching mechanism, and the test server did not have this function. So what ended up happening is that if one person saw the popup, it would then show it to the next 20,000 people, regardless if they had just seen it. No one on the team knew this existed, the test server did not have this function, only the live site.
I successfully took down this site for about six hours on my very first big project. The major lesson I learned from this experience is never to deploy anything right before I’ll be away from the computer and won’t be able to “babysit” it for at least the first four hours. I’ve never launched something new and walked away again!
Is there a particular book, podcast, or film that made a significant impact on you? Can you share a story or explain why it resonated with you so much?
My favorite book, and probably the one that has had the biggest impact on me, is called Illusions by Richard Bach. I read it for the first time in high school. It’s a parable about the nature of reality. Doing right and being good, and how that influences our lives. It’s a way of looking at reality not as a set thing, but how our actions and attitude affect our reality. The idea that your outlook towards the events that are happening to you is more important than the actual events themselves. It’s all about that perception, and how perception shapes your reality. This is very important from a business perspective. A good friend once said to me, “The defining characteristic of an entrepreneur is to be able to see how things are now, how things should be, and then see the path from one to the other.” A lot of people see one or the other, they either see how things are now or how they should be, but they don’t see them both together and the steps they need to take to get from here to there. This book really opened this concept up to me. It’s not enough to see that the world should be one way or the other, you have to understand it on a real, tangible level in order to move forward from your current reality to the reality you seek. It also consequently made me very superstitious. I’m like an athlete with my superstitions. The one that annoys people the most is that I won’t talk about projects or say that things are going to happen until they do. I don’t want to jinx things, and am afraid that if I mention things publicly they will fail.
Are you working on any new, exciting projects now? How do you think that might help people?
My full focus is on SegMetrics. The fun thing about running this business is that we have about 80 different projects within the company that we’re looking to accomplish at any given time. I’m very excited about that and I think not only will these innovations mean a lot for our customers, but to the marketing community as a whole. Everything comes under that SegMetrics umbrella, which is why I am so excited about its future. With the agency we were doing a little bit of work for a lot of clients and with SegMetrics there’s a lot to do, but it’s all for the greater good of our company. It’s all for this greater vision, but going back to my superstition, I’m not going to share specifics quite yet!
Thank you for all that. Let’s now turn to the main focus of our discussion about empowering organizations to be more “data-driven.” My work centers on the value of data visualization and data collaboration at all levels of an organization, so I’m particularly passionate about this topic. For the benefit of our readers, can you help explain what exactly it means to be data-driven? On a practical level, what does it look like to use data to make decisions?
First I think it’s important to talk about what data driven is not, because when people think of being data driven they often think of taking all the art and creativity out of the process. That’s what you should not do, the data is there to guide you, but it’s not the end all be all of what you need to do next. You have to make decisions and use your own creativity. The data is not going to tell you what to do, the data is just going to tell you about reality. It tells you where you’re doing well and where you’re doing poorly. Then you as a business owner need to make a decision about what that data means for your business.
For example, we have an acquisition funnel, with copywriters and designers coming into the funnel. Copywriters make up 90% of the people that come in, but they only convert at 1%. Designers make up the last 10% that come into our funnel, but they convert at 80%. That’s what the data says, but what you do with that is the art of business and marketing. At this point you have options, you can stop targeting copywriters or you can pivot and change your marketing to serve them better. This is one of the main problems that people have when they talk about data driven decisions. They think that the data makes the decision and it does not. The data lets you know what reality is, and then you have to decide if that reality matches the reality you want for your business. You have to ask yourself how you are going to change what you’re doing to get the current reality to match the one you want. You have to get creative.
Any organization, even a small one, can be quite complex. One person cannot understand every single technical aspect running through it–how your sales team reaches out to leads, ads, CRM, email marketing, and processing. You have all these different data points that co-exist but sometimes they don’t talk to each other or don’t have the right data. Or they may have holes, and it’s impossible for any one tool to see the other holes that may be breaking for other teams. A data solution like SegMetrics lets you smooth that all out, so that anyone can see how they fit into the bigger organization. That’s incredibly important, to not just see what you’re doing, but to see the greater whole of that organization’s ecosystem.
Which companies can most benefit from tools that empower data collaboration?
To be perfectly honest, I believe all companies can benefit in some way. We specifically are focused on marketing and marketing departments, so that’s where my brain always goes. Every part of a fulfillment, procurement, marketing or sales funnel, everything that includes a process needs data to be able to understand it better. As an example, I did a lot of work for Toyota. They have sixteen giant monitors at headquarters showing the exact flow of every single car from order to fulfillment. It was incredible to see all of that data in one place. Something that I feel very strongly about is that the things that are measured get improved and those that are in front of you every single day get the most attention. This is why I like dashboard systems like the one Toyota had. Knowing that you’re 20% down from the previous week, or that cars are currently stalled in Sheboygan, is information that we need to know because as we’re watching it happen in real time it remains top of mind. That’s how we are improving the business in every aspect.
We’d love to hear about your experiences using data to drive decisions. In your experience, how has data analytics and data collaboration helped improve operations, processes, and customer experiences? We’d love to hear some stories if possible.
When we looked at our own numbers, we had a very good conversion rate, about 50%, from trial to customer. So 50% of the customers that tried our free trial turned into customers, which was fine, except we weren’t really getting the results we desired and we wanted to improve that rate. So we asked ourselves, “how do we make it better?” We started diving in and examining our trial process. Everyone that starts the trial gets an offer for a free kickstart call. This is a one-on-one call that is essentially a training–we teach people how to use the software, how to improve their marketing, etc. The call is personalized and tailored specifically to their business. We found that people that participated in that call converted at 90%, those that did not converted at only 30%. How many people took advantage of the call? Only 20%. So 80% of our free trials were not taking advantage of this free call that was a sure fire way of converting them.
Again, here’s where data and creativity merge.
We had two choices, we either needed to create better DIY services or we had to have better messaging to get customers on the call. However, if we had just looked at that top line data — everything would have looked great! On the surface 50% conversion from trial is actually great. But we would have only been looking at the beginning and end of the data, we would not have seen the data that makes up that journey between and we never would have identified the customer segments that were performing better or worse. Then we went one step further and identified which industries the customers that participated in the calls were from and which industries typically did not partake. We started to find out that some customer segments had no interest in the calls. We were then able to devise a funnel that would allow us to speak to them specifically and have the same success as the kickstart calls without ever speaking to them one-on-one.
You can see by this one example, that by not seeing our actual customer journey — we’d be leaving a lot of money on the table, and a lot of potential customers without a solution to help them optimize their marketing as well.
We took something that was so wide and average, narrowed it down, found the outliers and improved our messaging to reach them in order to make sure they get where they need. It all ties back to seeing the reality as it is, then identifying the reality that we want, and finding the path to get to that reality.
Has the shift towards becoming more data-driven been challenging for some teams or organizations from your vantage point? What are the challenges? How can organizations solve these challenges?
Yes, there are a lot of people who have challenges. The main challenge being, everyone knows they need data and thinks they are data driven, and nine times out of 10 they aren’t. They either don’t believe the data, they don’t want to look at the data because it’s overwhelming, or they just don’t understand the data. Being data driven is not for everyone, some people’s brains just don’t work like that. I think that there is this view that the manager or CEO of a company has to understand the data, but that’s not always the case. They may be a great leader and have a great vision, but being able to parse the data themselves may not be a skill they possess, however you do need someone within the organization that does have that skill. You must have someone that can look at the data, surmise what is not ideal and make recommendations to management.
For example, when you’re working with ads people, they may see an ad that is down 20% and one that is up 50%, so they decide to take down the campaign that’s doing “worse.” That’s technically data driven, because you’re using data to make a decision, but it tells you nothing about what’s causing it or what should be done next. Someone needs to be able to look critically at that data — and not just parts of it, all of it. If an ad is turned off because it’s not visually appealing, but it was bringing in a lot of traffic to the site and now that’s dropped off — the ad needs to be re-worked visually and re-started. It’s the same as the difference between saying that your return on ad spend is up 8%, which is good news, but designers only convert at 4% from your ads, that’s bad news. The first stat didn’t give any actionable steps, the second stat provides an opportunity to make a data driven decision. It lets us know that not only do we have a problem, but it gives us a hint at a solution, which is that designers are not converting and something needs to change.
Often when people think of data they think only of key performance indicators (KPIs). They think of numbers going up or down, but that doesn’t hint at what action needs to be taken. You’ve got to have someone who understands how to look at the data and interpret it appropriately, there are people who enjoy deciphering data, find them and you’re set.
Ok. Thank you. Here is the primary question of our discussion. Based on your experience and success, what are “Five Ways a Company Can Effectively Leverage Data to Take It To The Next Level”? Please share a story or an example for each.
Here are the five ways I believe a company can effectively leverage data to take it to the next level:
Measure key performance indicators at a lifetime scale.
Data is reality, knowing that reality lets you forecast out. Knowing ads are bringing in two dollars for every one we spend in the first month is great information to have, but what does that look like long term? Being able to forecast not just the short term return on your efforts, but projecting the long term return is even more important. When you’re looking at acquisition costs, it might cost $100 to acquire a customer, but it may take three months to recoup that investment. You need to be able to forecast that out and project how long it will take to recover the costs on that investment. If you’re spending too much, and it will be eight months before those costs are recouped and the company goes broke in the meantime, that’s not helpful. At the same time if you know you’re going to get that money back you can afford to spend more up front in order to acquire that customer.
Uncover true return on investment.
There is more to customer acquisition than just your ad costs. There are ad costs, producing content, salespeople’s time, how many calls they get on, how much support they require etc. Being able to put all of that information into a data flow and see that every call they get on costs you $100 is incredibly useful. This not only shows you the cost of acquisition, but also the cost of conversion. You can then compare the lifetime cost of a customer to your overall revenue. Being able to see that certain customer segments are bringing in $1,000 on average, but costing an average of $1,500 in effort is very useful information. If you just look at your ad costs, it may look like you’re acquiring customers at $200 a piece, which seems great. But you’re not factoring in the long term costs, the time that the various support teams and other departments may spend on that effort and such. Being able to understand the true cost and the true return will give you a more accurate depiction of your overall return on investment. There’s no way to calculate this unless you’ve got a data funnel system in place.
Discover new niches and target markets that are underserved in your marketing.
No matter what business or industry you’re in, you’re always going to have niches and submarkets within your audience. You need to be able to identify the people either by demographic, industry, or action that indicate if someone is more or less valuable. There’s always this idea of “raising your hand” when talking about customer acquisition, you wait until someone raises their hand and then you reach out to them. But, what are the signs that someone is about to raise their hand, what is the correct combination of signs and people that show that they are ready to be sold to? There’s a reason that retailers like Target purchase lists of who was born when, because they have an instant target audience to market to for diapers, formula, food, and the like. They pay a lot of money for these birth lists because they want to be able to market directly to new parents with the products they need, they know that finding that sub market, that niche is key.
Discover holes in the process.
Shoring up things that are broken is incredibly important. Sales processes are complex, they go through a number of different technologies and any one of those failing can cause a problem later on down the road. The systems from different departments don’t necessarily talk to each other. Your sales department may have their section buttoned up and working perfectly, but they aren’t sending people to the marketing team. So you’ve got a sales pipeline that’s not getting sent to the marketing team which creates this giant hole where you have 10,000 people that are never getting an email. Or maybe on the surface level everything looks fine, but there’s a stop in one of your funnels that is causing people to get overlooked. There is so much complexity in modern marketing, customer acquisition, and fulfillment funnels. All of these processes are so complex that if you’re not measuring the movement of pieces from one data silo to the next then you are definitely missing out. A hole in any one of those can cause irreparable harm to your business.
Quickly find the things that work and improve on them, cut the things that don’t.
Find the things that work for your business and do more of that. Find the things that don’t work for your business and either stop doing them, or change your business so that it does work. It’s just that simple. Identify the things that work and do more, identify the things that don’t work and do less. The one interesting thing that people talk about in marketing is that you’ve got to have multiple channels, but in reality most businesses have one acquisition channel they go full force on. You don’t know what that acquisition channel is until you start trying a bunch of them, but the faster you can find that channel the more efficient your company will be. Because if you’re messing around with seven different acquisition channels for years you can’t really tell which one is doing the best, you’re never going to succeed.
The name of this series is “Data-Driven Work Cultures”. Changing a culture is hard. What would you suggest is needed to change a work culture to become more Data Driven?
There are a lot of theories, but for me the most important is to have data be in front of people. The biggest problem I see with companies that want to be data driven but aren’t is that it’s hard to get the data. When it is easy to get the data, people become data driven because they’re looking at it regularly and understand it. This is why each department should have their own KPIs, they should have their own things that they’re measuring, not things that aren’t in their purview. The level above them will manage all the departments. The other most common problem is that the data we do get isn’t applicable so we start ignoring it, which is worse. If you have an alarm that goes off every couple of hours, you get used to it. When a real alarm goes off, no one pays attention to it. It’s why we have a data dashboard, if you’ve got the data front and center for your team it stays top of mind. Making data readily available, easy to access, and act on is crucial. If it just creates extra work, no one’s going to do it or pay attention to it.
The future of work has recently become very fluid. Based on your experience, how do you think the needs for data will evolve and change over the next five years?
We are already at a saturation point in our current climate where we have more data than we know what to do with, so we’re having to get creative to get the value out of the data because there is so much. However, I do believe the need for data is going to continue to increase. Looking specifically at the marketing side, one of the issues that we have is that customers are looking to see the value of every single event that they provide. The question I ask is, what are you going to do with that information? How will it influence your decision making? Some people have a very valid answer, but some people don’t know, they just get that data because they can. These are the people that are trying to be data driven, but don’t understand the question that they are asking.
We have so much data available now and we’re going to have so much more available in the next five years. What’s going to be important is not the data collection, it’s going to be asking the right questions to make sense of the data. Asking the right question is not something that machine learning can do, it can help you get the answers, but at the end of the day you have to be the one asking the question. You have to know what question it is that you’re trying to answer to obtain the data set you need. That’s the art of business. Business is not just science, there is an art to it, it’s understanding and intuition born from data.
Does your organization have any exciting goals for the near future? What challenges will you need to tackle to reach them? How do you think data analytics can best help you to achieve these goals?
SegMetrics grew 12x in just 18 months, and we’re looking to do it again. Analytics is at the core of how we were able to do that the first time and it will be the core of how we do it again. It’s understanding our customers–who our best customers are, not just those we adore working with, but also those that are most valuable. Analytics is really at the crux of every decision we make and not just the monetary side, but also evaluating industries to bring in the type of customer we want to work with.
How can our readers further follow your work?
Find out more information about SegMetrics at https://segmetrics.io/ and follow us on Instagram, Facebook, and Twitter.
Thank you so much for sharing these important insights. We wish you continued success and good health!
