Our guest this week is Mike Brown, Head of Product at GenePeeks. Mike has the unique distinction of being named our very first Product Hero. And he’s also the first one to return as a guest on our podcast – something that we hope to do with many of our past product heroes.
The mission of biotechnology company GenePeeks is to provide better access to better insight by identifying risk that can’t be seen through other methods. Their unique, patented platform enables the rapid and accurate analysis of clinically relevant genotypes and variants to support a clinical workflow and help clinicians, parents, and parents-to-be make more informed healthcare decisions. Believing, like others, that healthcare will increasingly be driven by patients first – something they refer to as patient-led genetic testing – they are combining the science of computational biologists with a web interface that clinicians and consumers can use.
With GenePeeks serving the needs of very different populations, their platform can’t just do one thing and hope that all stakeholders get value out of it. So developing personas was something that was very new and valuable for the team when Mike joined the company.
The platform has expanded a few times over the years, something that Mike and I talked about in our interview. What were the signs pointing to this expansion, and how did they know that the time was right?
Mike and I discussed these and other topics, including:
- Finding the sweet-spot and focusing on the thing that you are really better at doing than other companies
- Combining data and presenting information in a way that’s good for each audience and is meeting their needs
- Developing a problem roadmap rather than a product roadmap. And the value of a “not roadmap”
- The challenge of having a user that may only use your product once
- Sometimes the best interface is no interface
Listen to the Show
- Follow Mike on Twitter
- Connect with Mike on LinkedIn
- Sharing in the Outdoors — lessons learned from a failed sharing economy startup
Heath: You have the certainly unique distinction and perhaps one would debate honor, of being our first ever product hero.
Mike: Woo hoo.
Heath: Did you know that, that you’re the first one?
Mike: I only knew that after you emailed me. Excellent, I love it. Happy to be the first.
Heath: Yeah, that’s awesome. Now you’re the head of product at GenePeeks, tell me about that transition. When did you go from PatientsLikeMe to GenePeeks?
Mike: PatientsLikeMe was an excellent role for me. It allowed me to grow a lot as a product manager, product leader. The company changed a lot while I was there in that the company in size nearly doubled twice. Our revenue model totally changed from doing marketing surveys to our patients for pharma clients to collect data about treatment adherence and things like that. Where that eventually shifted to was those pharma clients were very interested in not just that snapshot of data, but collecting data over time in a structured way, which is what our platform was doing inherently. They wanted access to more.
Once we started licensing data and setting up these data licensing deals with research institutions and biotechs in the area, that really changed the revenue model of the company. We were collecting all of this phenotypic data as reported by patients, symptoms, treatments, side effects, severity, co-morbidities, all these different data points that are really rich. What we eventually realized is that while we were doing a great job on the phenotype side, we were not doing a great job, or we weren’t doing anything yet, for the genetic side. That is the emerging space that we’re in right now, in biotech, is making sense of how your genes are connected to your phenotype.
PatientsLikeMe, while I was there, eventually raised a hundred million dollars to do that. That was a signal to me, as an aspiring entrepreneur product guy, is seeing signals in the noise. So the company had doubled nearly twice, raised a hundred million dollars to do genetics, and that was a signal to me that there’s enough money and interest and potential innovation in the genetic space, that it might be time to really focus in on that almost completely.
I started looking around, and saw a number of different small companies that were doing interesting things in genetics, and eventually met the co-founder and CEO of GenePeeks, Ann Morris, and we got to talking. It was very natural, I wasn’t exactly looking to jump out of PatientsLikeMe immediately because it had a role to play in the genetics integration there, but had the chance to join a ten-person, very small, early start up in purely the genetic space, so I decided to jump on it. Really good opportunity, very, very bright team of people. It had probably been nearly 10 years since I had been exposed to genetics prior to that, just in graduate school at Tufts University. In that last 10 years, it has totally changed and has exploded. The things that they were telling me that they could do, I didn’t think were possible or possible yet, so it’s been really exciting. It was seeing some signal that genetics is really exploding and trying to figure out how to get more in on the ground floor of companies that are doing that.
Heath: It feels like I’ve been in a couple of companies myself where the product was one thing, but what it really was was pharma wanted to buy the data. They seemed to be behind a lot. They get a bad rap sometimes, but it’s been my experience that a lot of what they’re paying for, for data, is to advance the science and for research. It’s not what people assume, it’s like, Oh, they just want to be able to target me and sell me their pills. Yeah, they all have marketing departments, sure. The real nugget that they’re looking for are these insights that they can’t get at without this longitudinal, historical data that they piece together and they themselves are obligated and obviously because they’ve their own businesses and they need data that’s strong across brands to get at that data and that information. It’s interesting that they have their hands everywhere and in some cases no where, and they’re still trying to find that holy grail of data to piece together these questions.
Mike: Absolutely. One of the first questions I asked the co-founder of PatientsLikeMe when I was interviewing there was, It sounds like your model is to collect data from sick people and sell it to pharma. That’s potentially an uncomfortable topic. His perspective is that all of these companies that are out there developing drugs, are doing so without a lot of real data on how that’s actually affecting people. The studies that they do are quite limited because it’s usually limited to their own particular drug.
But in reality, people suffer from multiple conditions. They have multiple co morbidities that they’re dealing with, on the platform of PatientsLikeMe, patients were taking, some people were taking dozens and dozens of treatments every day. Which range from things like chemo to drugs for MS and a number of different things. What are those interactions like? It democratized the data that these companies are using to make decisions. Would you rather have people developing drugs in a silo with no context to real-world data, or would you rather give them access to do R & D with data that’s accurate to people’s day to day experience? The openness policy of the company was really upfront and center, which is really honest.
Heath: It’s not unlike a simpler, straight forward example of user testing your software. You can create the artificial laboratory conditions, to try and determine whether this design works better than that design. But it’s better if you can say, Look, we have this user history out there and we can combine it in multiple different ways to figure out what combinations work and what don’t. No matter how good your test plans were, QA is never going to find everything. As soon as you release something, your users are going to find something you didn’t uncover because you didn’t realize that was a use case. That’s an important thing to learn, you’re just applying it to health care.
Mike: Exactly. If users are good at anything, they’re good at breaking the stuff that you’ve built.
Heath: Exactly. Let’s assume there is a typical day, as head of product, what does that look like. How do you divide your time, your thoughts, your day as the head of product?
Mike: It was kind of interesting coming into the company, it was about 10 people big, mostly computational biologists and bioinformatics and a few software engineers. The thought coming in was, Who’s this product guy and what is he even going to do here. It was kind of an interesting question from the people who I was starting to work with. What I found, moving in there, was that we had just expanded the platform from looking at how someone’s genetic combination with a sperm or an egg donor might affect the inheritance of disease in their offspring.
As I came in, we expanded at that same time, to all reproductive scenarios. So, husband and wife, or two females and a donor, two males and a surrogate, single parent, all these different kinds of combinations and so the use cases really expanded quite a bit. What we started to do, which was something that was new, was how do you really combine the science that our computational biologists are doing with a web interface that consumers can use. We had taken this approach that health care would be driven by patients first. What we talk about is that it’s patient-led genetic testing.
One main difference between us and a lot of other companies out there doing genetic testing, is that we are ordered by your doctor. It’s a clinical test, we give you data back about the inheritance risk for having your future child with your partner, whomever that may be. That’s ordered through your doctor, the results first go to your doctor. We also give the results directly to the patient too. You have your own user name and password for the GenePeeks portal, where you own and get access to your data, but it does have to go through the regulatory pathway of your doctor first.
A lot of times it’s led by the patient. There are a lot of interesting companies out there trying to make that more robust, where the patient says, “Hey, I want this test done. I think it’s useful for me and my family.” Then working with a network of doctors, or your own specific doctor to sign off on that test and deliver the results in a way that not only your doctor’s involved and your doctor’s in the room, but you get access to the data too and it’s not buried, either in a paper chart, which still exists, or your EHR, which is also sometimes hard to get access to the data too.
There’s a couple of different layers. There’s a layer for our internal research team. It’s very, very deep and in the weeds, as in the weeds as you can get. Then there’s a layer for molecular, clinical geneticists who are making sense of your DNA data and all the potential damaging variants that you have in your genome. Then there’s the layer that is for your actual clinician, somebody who’s maybe a reproductive endocrinologist, or primary care doc who is at a different level in genetics, in terms of understanding what the impact is and what the risk is.
We have a few different layers of information presentation. It’s like, how do you keep it high-level enough so that people understand it, but detailed enough that you know what’s going on and we’re not hiding anything in the background. How do you open up that box without it being too confusing or complicated? A lot of what we do is try to figure out how to take data that already exists in different research databases that are used for protein modeling or allele frequency in your genome and how do we combine those in a way that visually makes sense to people and combine them to get a result about how damaging a particular area of your genome is and just describe that in a way that’s good for each audience and its meeting their needs.
Coming in and developing personas with something that was very new and very valuable. Who are we talking about? Are we talking about the 40 wealthy single mom? Or are we talking about a clinical geneticist? Those are two very different populations, and so how do you serve their needs in a useful way because you can’t just do one thing and hope that everybody gets value out of it.
Heath: If you test two random people on the street, their combination is going to have a lot of potential results. The question is, which ones are of significance.
Mike: Exactly. What we do is filter out all of that noise for people. What has been happening previously is people will go out and do a carrier screen and say, Hey, I might have some family history of cystic fibrosis, for example. I’ll go out and do a carrier screen. That’s interesting, and it tells you that you’re a carrier of a disease, but it doesn’t necessarily tell you what your partner has. If you go test your partner, that’s interesting too, but that is too simplified for how genetics actually works so the actual combination of their DNA is fairly sophisticated.
We model that very closely and look at what is the actual combination of two people’s genomes and what is the resulting genetic risk? It’s interesting, we filter out all of the things that aren’t going to affect your future child.
Heath: It sounds like you’ve got at least two users, multiple personas on each, but you’ve got the clinician and you’ve got the patient consumer user. How do you manage them? I’ve been at companies like that where you’ve got two. One is paying, one is non, maybe, or one is just the conduit to get the data, but they’re the user, not the customer kind of thing.
Mike: Absolutely. Right now, our tests and our analysis is all paid for by patients directly. It’s not yet covered by insurance, although that’s something that we’re actively trying to unlock for people. The money comes from the patient directly. Who you get paid from is very important, but then the person who holds the keys to that analysis actually being delivered is also very important. There’s a number of different features that we had to build for the two different audiences to make sure that the workflows that they need to do are smooth and in some cases, making assumptions that, Oh, we need an interface for a doctor to be able to sign off.
Well maybe sometimes the best interface is no interface in that there are cases where doctors just say, Yeah, I want to sign off every time. So, they have a signature on file or they just order the test automatically because they’re comfortable with what the test is, who the test is coming from, and what the results look like. In a lot of cases, the doctor has to do very little, if they understand exactly what’s going on really well.
Heath: So did I hear you correctly in that originally GenePeeks was focused solely on female egg, male sperm, and then your expansion was to include a whole additional set of combinations for that, or was an expansion of diseases or?
Mike: Yeah, it was more to the broad reproductive market. We spent a few years in the donor space proving out the technology and then in the last year and a half doing work in the broader reproductive market and now we’re actually growing a lot faster than we anticipated and we’re launching into diagnostics also. Which is more broad, so instead of being pre conception for family planning, we’re moving into now you have an affected individual, an individual who’s sick, and often times you don’t know why.
A lot of the diseases are rare, and so the incidence in your population is very, very low. The having a rare disease is not rare. It’s between five and eight percent of people have some sort of disease and there are percentages of those that are very rare, and are undiagnosed. We’re working with a number of organizations to actually try and diagnose people for whom we can’t figure out what their diagnosis is. A lot of people, for the really hard cases, people are turning to genetics to try and figure out, What variants did they have inside their genome and are those variants significantly affecting this individual or not?
We make that determination really well, which is really interesting. There’s very few companies who are doing that interpretation side really well. Just looking at the time line of genetics over the last many years, the human genome was only sequenced in 2003, fast forward another ten years, people got really good at trying to figure out, given this human genome, how does that differ from a reference genome? All those differences they call variants. Here’s a variant in a gene, maybe there’s many variants in a gene, the genetics industry and community has gotten really good at being able to tease out those variants. In some cases, within an hour, you can look at somebody’s genome and say, yep, you’ve got these variants in your genome that we should take a look at.
Then around 2009, 2010, when GenePeeks was founded, people were really struggling with, now you have all these variants, how do you interpret them? What does it mean? It means nothing to have these variants. Are they important, are they not important? Hard to say. GenePeeks made this bet that the interpretation side of genetics would need a solution that’s automated and software-based. Since that period of time, over the last six or seven years, that’s what Gene Peak’s been developing, is the interpretation side of those variants and doing that in a computational automated way.
What people are doing today, in a lot of cases, is they get these variants back and they’ll have a team of genetic counselors and microbiologists and geneticists, pour through all of these variants, sometimes a dozen, sometimes two dozen variants, and in a lot of cases, it takes weeks to sift through that. They’re looking at research, they’re looking at publications, they’re looking at databases and doing a lot of it manually. Sometimes it’s been semi-automated scripts and things like that, but a lot of it’s automated. We’re seeing a lot of groups where it’s taking two to three to four months or longer to do the interpretation side of it.
We do that within a day on the computational side. That has been really powerful. While we had previously been really focusing on developing the tech for the reproductive space, because we are much faster and look at a lot more of the genome, that has become a really good sweet spot for GenePeeks compared to what the standard of interpretation today. It’s been really interesting to be part of that and to really focus in on what the thing is that you are ten times or a hundred or even multiple hundred times faster than other people. At a small start up, you focus on exactly the thing that you really the best in the world at, and then interpretation side is really where we’re light years where many other people are.
Heath: So this is way cooler than I thought it was, originally when you started out. I thought you were hinting at the expansion, well we started off male female sperm egg, then we expanded to recognize really what’s going on in the world around us to other forms of reproduction. But really what you’re talking about is much bigger than that and it’s really interesting.
Mike: That was the dream of GenePeeks, is to move into these much larger, not just reproductive but diagnostic.
Heath: So someone thought in this direction beforehand?
Mike: Oh yeah. It was just about when, really. It turned out that just happened a lot faster.
Heath: So what determined the when then? What were the signals? What caused someone to press the button and say, Now go.
Mike: There are a number of different things. We realized that our, there’s a whole bunch of things that GenePeeks does. We work with clinics, we process data, we do this variant calling that other people are already doing and we deliver reports sand we work with doctors and patients and things like that. Lots of companies already are kind of doing those functions, the one thing that GenePeeks is doing that other people are having a hard time with is that interpretation part, where we have a very automated and a lot of other groups don’t.
That was the light bulb, aha moment, for the team, for our collaborators, for our board, investors, etc., that the thing that GenePeeks is, very truly unique in the market, is the speed, accuracy, sensitivity, specificity of doing the interpretation part of your genome. Once we realized that the market was ready for that, and that we were also ready for that, we jumped in full into the interpretation side. It’s really interesting working at a relatively small company, how do you do everything, especially when you’re competing against a number of companies that are better funded, much larger, all these things, you really have to focus the team on what is that specific thing where you guys really are ten times, a hundred times or more, better than everyone else. That’s what we’ve done, is really the resources of our company really focused on that interpretation side and making sure that’s automated and accurate.
Heath: Which user base would you say is more difficult to solve for or meet the needs of, is it the clinician side, is it the patient side, or is it there really is no distinction?
Mike: Everything is a problem, all the time. From the clinician side, every clinic wants something different. Some clinics only want to fax, some clinics only want to do it over the phone, some clinics want it integrated with EHR, some hate the EHR and want to use your platform, many clinics abroad would prefer that it’s translated into their local language. There’s like all of these needs that come through from the clinic side. It’s trying to figure out how do you prioritize a focused feature set for your product and not do all things for all people, which sometimes means you alienate some parts of your user base by not providing them with exactly what they want. That’s a constant challenge.
On the patient side, it’s a challenge too because the GenePeeks test is unique in that it’s not just a test for one person, it’s a test for two people and sometimes more. So we can look at your family for example, sequence a child who’s sick and also their two parents. In the current clinical world, it’s mostly one-to-one. You do a blood test for a person, urine test, genetic test, whatever, but it’s one-to-one. Because we’re now one-to-one, one-to-two, one-to-three, or more, how do you then deal with things like data rights, data access, consent, informed consent, research consent, all these different problems are now a little bit new.
I didn’t fully appreciate that when I dove into this world, and it’s much, much more complex than I anticipated. Just the fact that you have, for example let’s say in the two parent’s case we were doing a pre-conception analysis for them. One person buys the test. They’re the patient of their doctor who signs off on it. But that second person in the analysis, maybe doesn’t go to that doctor and they also didn’t pay for it. How do you return results to them? How do you get them involved? How do you do that on the web and how do you do it legally and ethically? So there’s a whole slew of problems around how do you deliver, how do you collect data and deliver results to multiple people involved in analysis. It’s not something that the clinical world has yet contemplated.
Heath: You still have stories of patients simply trying to request access to their medical record, forget a case like this, you get multiple combinations, multiple people rights to consider, and it takes now 30 days, paper-based, to get access to my record, if that. You’re talking about something inherently more complex.
Mike: Yeah, and doing it on the web, too. That has actually been a huge problem to solve of which we’ve done, I think, a decent job solving it but I don’t think any of us really appreciated at the beginning how challenging that would actually be. None of the use cases for our personas are easy, it’s just about figuring out which ones are the biggest pain point. What we look at, at the company level, you try to do the thing that is solving the number one pain. At the product level, when you’re actually looking at your road map and what are we going to do with ten engineers, a few designers and a product person, how do you spend those resources wisely?
You really just look at what are the biggest things that are in the way over the next six to eight weeks? What’s keeping us from moving forward? It’s being very clear about what your goals are, which sometimes from the executive level can be many goals. As an executive, you want to have your cake and eat it too. You want everything to be done yesterday so do your investors and your board, so a big part of my job is to take all of those needs, desires and long term vision, and make those goals crystal, crystal, crystal clear and when they’re not crystal clear, work with groups to force that.
Then when you bring that to the team that’s actually responsible for doing that value creation, all of the engineers, designers, scientists, etc., that six to eight week set of goals should also be very, very clear. Stitching that together is a big part of what I spend my time doing. We have a barrier’s road map. We’ll have what our goal is over the next three to six months, then how do we just squash barriers that are in our way between now and then. Everything that we ship, and everything that we push across the finish line, has just removed some barrier for us, making it easier for us to get there.
We take this problem of it’s not necessarily like a product road map, I don’t really like that term because it makes it sound like the product person says, This stuff’s important, do it. We refer to it as a problem road map. We’re here to solve problems, people love to solve problems, these are experienced people, they’re solving workflow issues, how do people do things in an easy way. Everything is a problem all the time, so you’re always trying to fix it. Same on the engineering side. We really are just trying to solve problems for the company, for our users, for everybody in that chain.
It’s about figuring out what the most important ones are. That’s super challenging, and you’re always in some cases, making people unhappy because where you sit in the organization or where you sit as a user, for you, that pain point might actually be your biggest pain point. It’s about looking at everything, and deciding, Where do we spend our resources? If we’re spending tens of thousands of dollars over the next one to two months, how do we make that a really good investment? Not just work on all the things, all the time. Which is just too diluted for a very small company that’s expected to work fast and do big things, like solve undiagnosed diseases.
Heath: I like that, the problem road may you called it?
Heath: What I like about it is it certainly ties it more closely to the user, because your users don’t have products, they have problems, and that’s what you’re trying to solve for, is their problems. I dig that.
Mike: I made that mistake many, many times where you just build stuff. It’s easy to build things. People want to build things. The people you work with want to build stuff, but framing it in terms of solving problems for people is, I think, more gratifying and more focusing than just saying, Let’s build some features that we think are neat or helpful. What is the real problem and stay away from the edge case stuff that is nice to have, maybe someday you’ll do it, but solve a real problem and solve it fast for people.
In addition to the problem road map, I also have what we call the not road map. It’s all the stuff we’re deciding not to do and it’s years long. There’s stuff that we’ve come up with as really excellent ideas and interesting things, but we’re very firmly deciding to not do them. I think it’s really important for the team to understand we are not in the business of doing this, whatever that happens to be. Maybe later, maybe if the company grows a lot more, etc., but right now we’re not doing that.
Heath: That’s an important, we’ve always said that’s an important part of your road map. It’s what you’re going to do, and when in broad terms, but also what you’re not going to do. You’ve got to clarify that. Don’t just assume you’re passively not doing something, you’re actively not doing that and there’s a reason why. It may be that we’re not in never, or we’re not now.
Mike: It’s helpful to publish that kind of stuff with inside the company so that way, in three months, when a stakeholder is maybe upset that something is not done, you can say, Well, back in July we decided not to do that and that’s why it doesn’t exist. It’s making that known to people throughout your company that these are the decisions we’ve made, they may not be the best ones, but we did make a decision about what’s important, what’s not yet important and making sure that people understand that is super important.
Heath: So did we cover the pre-conception screen product?
Mike: What exists today, for the most part, is a lot of people will get tested after they’re pregnant and then you’re faced with the decision of, You have a sick child, what do you do? What people have been doing, prior to GenePeeks, is they’ll do carrier screening, so ahead of time they can say, What’s the risk for me and my spouse? For specific ethnic populations, that’s really, really common. The more closely knit you are in your ethnicity, the higher the incidents of disease. So, there’s some communities where it’s really prevalent and people just get tested a lot.
Heath: I’m from the south, so I can appreciate that. We always say, We have very few branches on our family tree.
Mike: I think it’s similar in the far north, also. So people are doing carrier screening, and that has become very ubiquitous, but as all of those tests get better, what they’re finding is that the positive rate of how many people are carriers of diseases, the more diseases you look at the more you find that everybody’s a carrier of the disease. So those positive rates, you can see on the trend line, they’re approaching 100% because it is a fundamental truth that you are a carrier of diseases.
Moving beyond that, once we understand that everyone’s a carrier of disease, what do you do next. Where GenePeeks has filled a niche and solved a dramatic problem is looking at not just people in isolation, but looking at both people. We’ll sequence mom, we’ll sequence dad, and then using our platform and algorithm, digitally combine that DNA in a way that simulates reproduction and then make a risk assessment based on that. That’s the whole pre-conception value prop is we’ve gone way upstream in the process and let you know ahead of time, inform your reproductive plan ahead of time, as opposed to after the fact, because there are clinical interventions that you can take.
Heath: What are the biggest challenges associated with being in the business that you guys are in? Is it clinical regulatory? Is it boring software challenges, they’re all the same, it doesn’t matter?
Mike: That’s really interesting. The biggest challenges are day to day figuring out what’s the most important stuff to work on, especially with a young, startup company. How do you make sure you’re spending those resources wisely? That’s the kind of thing that keeps me up at night. That’s at a company level, but in terms of things that are the most challenging, at a business level, it’s interesting, in that your customers will generally only use you once. If you’re targeting patients, for example, they’ll buy your test, they’ll use you once, and then, unless they’re having a second kid, even if they’re having a second kid, they still may not use you and that’s many years down the line, typically.
To have a user that only uses your service once, is really challenging. Then you have to reframe yourself and say, Alright, where is the volume coming from? Who is the user through which they’re pumping volume into your system? For the pre-conception product, it’s clinics. Reproductive endocrinologists clinics, so they have people coming through the door who are even maybe having a hard time getting pregnant or are previously known to be at risk. Those groups are constantly pumping our funnel with patients that we can run analyses with. It’s challenging to figure out at first, do you focus on the one off patient or do you focus on the clinic who’s bringing you many, many, many patients a day? That was really challenging to figure out how to approach a business where people really only need you once or twice.
Heath: It’s all in how you define success and value. At the end of the day, if you’re delivering value to the user, that’s what matters. It behooves you to continue to focus on defining and delivering value that’s appropriate to the user.
Mike: One thing I found at every problem that I’ve tried to solve generally is that if you require your user to spend more time to do a particular thing, they just won’t. If you can, conversely, if you can save your user time, they’re move willing to, more likely and more willing to use your product but they’ll also spend more money on it. I noticed that myself as a consumer, but then in every project that I’ve worked on, the projects that have gone really poorly are the ones where we’ve actually placed additional burden on the user, even if we somehow convinced ourselves that was necessary at first. It would flop because we required them to spend more time to do a particular thing. The takeaway for me is always try and save people time.
Heath: I think healthcare is the worst at this, we make our clinicians spend more time, it’s not that they don’t want to spend no time in these digital tools, they just want to spend appropriate time and they want to be efficient. If you can deliver a tool that makes them more efficient, they will adopt it. They’re not techno Luddites, we’ve made physicians and clinicians out to be technophobes, they love technology, but they love technology that’s designed for and works for them. That’s certainly the key take home message and it’s really appropriate today is now we have all these digital tools in the workplace and we can measure how much time is spent in them and we can ask users, Hey, by the way, how do you feel when you use this? Do you enjoy using this? No, I’d rather not frankly but I have to use it. Well then why don’t you design it so that it delivers joy to them. There’s no reason why, okay fine I guess a time and materials system may not deliver joy, but at least you can make them not unhappy.
Mike: Our head of user experience has a quote where he’s constantly saying, Sometimes the best interface is no interface. What that means is sometimes you just don’t need somebody to come in and go through four or five screens. Maybe many of those things we can actually do in the background, give us some limited set of information, and we’ll create all of our models and things like that on the fly so you don’t have to manually go through everything. And how to aggregate information such that you’ve maybe removed a number of interfaces. You’re just removing the number of things that people have to do and have to go to. So don’t always assume that for this task we need to build something. Maybe we can actually remove something. That’s a really interesting framing, so we’ve tried to do that also at GenePeeks, how do you make work flows very easy for people and sometimes it’s removing them completely. Always ask yourself that question, is there something we could actually remove to make this easy for somebody.
Heath: I dig it. Well, thanks for coming back to Fresh Tilled Soil and thanks for coming back to The Dirt.
Mike: Absolutely. Pleasure to be here.