What's in a price? How to price your products and services
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Serientitel | ||
| Anzahl der Teile | 88 | |
| Autor | ||
| Lizenz | CC-Namensnennung - Weitergabe unter gleichen Bedingungen 3.0 Unported: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen und nicht-kommerziellen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und das Werk bzw. diesen Inhalt auch in veränderter Form nur unter den Bedingungen dieser Lizenz weitergeben. | |
| Identifikatoren | 10.5446/37356 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache | ||
| Produzent | ||
| Produktionsjahr | 2018 | |
| Produktionsort | Pittsburgh |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
RailsConf 20184 / 88
9
14
16
19
20
22
23
26
27
28
34
35
36
37
38
39
41
42
46
47
53
57
60
62
63
64
69
72
80
85
87
00:00
Dienst <Informatik>Produkt <Mathematik>RechenwerkRFIDTwitter <Softwareplattform>FontSystemzusammenbruchSoftwarewartungKartesische KoordinatenNichtlineares GleichungssystemWeb SiteSoftwareentwicklerProdukt <Mathematik>Open SourceProgrammiergerätHilfesystemQuick-SortFreewareDiagrammComputeranimation
03:33
RFIDSystemzusammenbruchThermodynamisches GleichgewichtBildschirmsymbolGeradeWort <Informatik>FlächeninhaltSchreib-Lese-KopfAggregatzustandProdukt <Mathematik>GraphZahlenbereichRechenwerkLokales MinimumKurvenanpassungKartesische KoordinatenLogarithmusNichtlineares GleichungssystemPunktQuick-SortStrategisches SpielEinfügungsdämpfungRechter WinkelMehrrechnersystemDiagramm
07:32
Dienst <Informatik>Elektronisches BuchCASE <Informatik>Arithmetisches MittelKurvenanpassungInformationGraphComputeranimation
08:27
Produkt <Mathematik>WhiteboardQuellcodeMultiplikationsoperatorAbstraktionsebeneOrdnung <Mathematik>Computeranimation
09:02
Produkt <Mathematik>SchaltnetzBeobachtungsstudieDatenfeldProdukt <Mathematik>FlächeninhaltPunktInteraktives FernsehenDivergente ReiheEinfügungsdämpfungComputeranimation
11:02
RechenwerkRFIDSensitivitätsanalyseMeterMaßstabMotion CapturingLokales MinimumSondierungProdukt <Mathematik>DifferenteProzess <Informatik>Quick-SortBeobachtungsstudieComputerspielZentrische StreckungSondierungNichtlineares GleichungssystemResultanteSchnittmengeInformationWeb ServicesDemo <Programm>Arithmetisches MittelInstantiierungLokales MinimumProdukt <Mathematik>MeterSensitivitätsanalyseBildschirmfensterMedianwertComputeranimation
14:59
Produkt <Mathematik>COMFamilie <Mathematik>Poisson-KlammerProdukt <Mathematik>Kategorie <Mathematik>DifferenteCASE <Informatik>Computeranimation
16:11
Produkt <Mathematik>Spannweite <Stochastik>Zentrische StreckungProdukt <Mathematik>InformationExogene VariableComputeranimation
17:01
DifferenteOrdnung <Mathematik>SoundverarbeitungProzess <Informatik>RichtungResultanteRechter WinkelZahlenbereichComputeranimation
18:14
Schiefe WahrscheinlichkeitsverteilungVirtuelle MaschineXMLComputeranimation
18:52
Funktion <Mathematik>DistributionenraumSchnittmengeMultiplikationsoperatorZweiStatistikVerteilungsfunktionFunktionalt-TestWort <Informatik>Computeranimation
19:32
GleichheitszeichenFunktion <Mathematik>CASE <Informatik>Luenberger-BeobachterHecke-OperatorZufallsvariableComputeranimation
20:00
Exogene VariableSkriptspracheLuenberger-BeobachterVariableDistributionenraumZufallsvariableKurvenanpassungProdukt <Mathematik>GrundraumComputeranimation
20:32
GleichheitszeichenFunktion <Mathematik>Lokales MinimumMaßstabMathematikLokales MinimumVariableZufallsvariableVektorraumProgrammierungFunktionalComputeranimation
21:35
MaßstabGleichheitszeichenFunktionalZentrische StreckungGeradeCodeStatistikFramework <Informatik>ResultanteTeilbarkeitLuenberger-BeobachterDämpfungDivisionEinsZahlenbereichComputeranimation
22:45
GraphResultanteOrdnung <Mathematik>KurvenanpassungART-NetzStatistikGraphComputeranimation
23:16
KurvenanpassungCASE <Informatik>MehrrechnersystemMultiplikationsoperatorMAPServerComputeranimationDiagramm
24:28
HyperbelverfahrenProdukt <Mathematik>ComputeranimationDiagramm
25:06
SchwellwertverfahrenGraphKurvenanpassungZahlenbereichMAPWechselsprungProdukt <Mathematik>ComputeranimationDiagramm
26:03
GraphXINGNim-SpielGlobale OptimierungGraphResultantePunktGlobale OptimierungLokales MinimumComputeranimationDiagramm
26:39
RFIDProdukt <Mathematik>VerschiebungsoperatorArithmetisches MittelSchwellwertverfahrenKurvenanpassungDiagrammComputeranimation
27:07
KurvenanpassungGraphLokales MinimumKurvenanpassungGraphDruckverlaufInformationDifferenteGlobale OptimierungPunktComputeranimationDiagramm
27:37
PunktKurvenanpassungPunktRandverteilungComputeranimationDiagramm
28:09
PunktDifferenzkernZahlenbereichRechter WinkelSpannweite <Stochastik>TeilmengePunktGraphDiagrammComputeranimation
28:39
GraphKurvenanpassungMenütechnikEntscheidungstheorieGraphKurvenanpassungSpannweite <Stochastik>VektorpotenzialMedianwertMultiplikationsoperatorEntscheidungstheorieTeilmengeInformationComputeranimationDiagramm
29:28
Lokales MinimumDienst <Informatik>Wort <Informatik>ComputeranimationXML
30:24
MaßstabSondierungZentrische StreckungComputeranimation
31:10
GruppenoperationWeb SiteUnternehmensarchitekturGruppenoperationWeb ServicesSondierungMathematische LogikDifferenteResultanteSensitivitätsanalyseBeobachtungsstudieComputeranimation
32:04
SondierungOrdnung <Mathematik>Deskriptive StatistikResultanteWärmeleitfähigkeitWort <Informatik>Inhalt <Mathematik>ErwartungswertTabelleSondierungComputeranimation
32:56
RFIDProdukt <Mathematik>Produkt <Mathematik>Physikalisches SystemLeistung <Physik>AbstraktionsebeneMultiplikationsoperatorBeobachtungsstudieDifferenteMultiplikationSoftwareentwicklerInformatikBildverstehenWort <Informatik>Computeranimation
35:19
COMRFIDComputeranimation
35:57
COMp-BlockDatentypKartesische KoordinatenMultiplikationsoperatorComputeranimationXML
Transkript: Englisch(automatisch erzeugt)
00:12
I'm gonna get started. So the title of my talk is What's in a Price? How to Price Your Products and Services. My name is Michael Herold.
00:21
I'm the lead application engineer at a company called Flywheel. We help creatives do their best work. The way we do this primarily is by having a world-class WordPress hosting experience where we make it a delight for designers and creatives to host their WordPress websites.
00:41
We are hiring like everybody else. We are hiring product managers, scrum masters, developers of all kinds. If you're looking for a job, come talk to me or any of my friends here that are in their Flywheel shirts. If you have any questions, please tweet me at mherald during the talk. If you tweet me, I will try to reply to your comment.
01:05
All right, so as an outline for the talk, I'm first gonna tell you a story. After that, I'm gonna give you a quick crash course in economics. I know economics isn't something that most programmers have a lot of experience with, so I think a quick crash course
01:22
will help you get your feet under you when you're thinking about pricing. Then we're gonna talk about a model that is really gonna help you do your pricing from now on. It's a very simple model to use and I'm gonna walk you through how to use it, when to use it, and why. And then I'm gonna talk about how the partnerships that we can make
01:40
through pricing and through the mindset that we gain through doing this sort of work can help you be better at what you do. But first, the story. So we were launching a new product and I heard that we were planning to price it at $99 a month. And I innocently asked, how did we decide on that price?
02:02
Can you guess what the answer was? We're a very intentional company, so I expected there to be a great answer. And in this case, there was actually not really an answer at all. What's missing from this picture? What's missing from the lack of answer
02:21
to why are we pricing our product at this price? We're an experienced company. We have several products that we have intentionally priced in the past and we still made this mistake. What if you have a side gig? What if you're a solopreneur? Pricing your first product is scary.
02:43
It's scary to put your work out there and ask for money for it. Heck, it's scary to put your work out there for free. Ask any open source maintainer. But having money enter the equation makes it even scarier. Pricing your first product is also hard.
03:02
It's likely that you've never thought about pricing before. I know I hadn't really before I got into doing this. How do you do it? What if you do it wrong? What if you overprice? What if you underprice? Makes the pricing your first product, it's full of angst.
03:20
The fear and the uncertainty in what you're doing, it causes a lot of angst and you're worried about what if I embarrass myself? Will it cost me my job, my reputation? What if I fail? It doesn't have to be like this. How can we make it easier?
03:41
How can we go from screaming our heads off to a nice zen state where we feel comfortable with what we're doing? So that's my story. I'm gonna give you your quick crash course into economics now. So economics, this is a nice icon that I found
04:04
representing economics as a whole discipline. It's really interesting that you can sum it up in four lines. So what's in a price? When you hear the word price, what do you think of? A price is just a mutually agreed upon construct for doing a bartering situation.
04:21
We happen to use money now, but in the past there wasn't money. It was all bartering. Now there's money to make transactions easier. We don't have to agree on the value of a cow versus a chicken. We have dollars or other currencies that we can agree upon the value of based on its relative price to other things.
04:43
So in economics, everything starts with two axes. The vertical axis is price. The horizontal axis is quantity. The price axis is based on like price per unit of things that you're selling and the quantity is the number of things
05:01
that you're selling. When we graph supply, supply curves are always up and to the right. They start at a point where you have to cover your costs unless you're doing some sort of weird pricing strategy where you're taking a loss on a product. So it starts above the end of the pricing axis
05:21
and then curves off in a logarithm as you approach your ability to produce what you're making. This is generally a technology problem. We can't infinitely create things. So it's always a logarithm.
05:42
And then we graph demand. Demand is the consumer side of the product or the equation. When things are cheaper, we tend to prefer more of them so it's a downward sloping curve. Down in the lower right, you can buy a lot of things for a very little price.
06:01
In the top left, if you're buying a car, you don't really wanna buy very many cars because they're very expensive. Where the two lines intersect is what we call equilibrium. Equilibrium is based on all of the knowledge that consumers have and all of the knowledge that producers have.
06:23
There's an equilibrium price and there's an equilibrium quantity. These two things are trying to match each other. The people that are producing things don't wanna produce more than they can sell, then they'll have to sit in a warehouse. That's inefficient.
06:42
The people that are buying only really want the amount that they want at that certain price. There are two more things that we can garner from this graph. The first is this red shaded area. In this area, we have consumer surplus. Because you know that you're willing to spend
07:01
this amount of money for this quantity of things, if you get a better price than that, you're happier. You're a much happier person as a consumer because you haven't spent the maximum amount that you're willing to spend. The other thing is the other side, which is producer surplus.
07:21
Producer surplus is roughly equal to profit. There's also opportunity cost. I'm doing this thing now so I can't do this other thing that's factored in there, but you can roughly think of it as profit. So what does this graph mean for us? Supply is largely controlled by us,
07:41
the people that are making things. In the case of if you've written a book that you wanna start selling as an e-book online, supply, your supply curve is gonna be relatively horizontal because your only costs are the ability to host your e-book. In the case of large services like GitHub where they have lots of hosting services,
08:02
they do lots of things in the background, they have a lot more costs involved, the supply curve looks different, but it's still largely controlled by us. Demand is largely from the consumer or the customer. They know what they want, they know the information that they have,
08:23
and they will do what makes them happiest. In order to price well, since we know the supply side, we need to figure out the demand for our products. Without figuring out demand, it's like throwing a dart at the board. You don't really know where you're aiming
08:42
and it can end up anywhere. That's the source of the angst when you're doing your first pricing for your first product. You don't really know what you're doing, you don't know what you want to sell things for and what's appropriate, so you want to figure out the demand for your product.
09:00
So we're developers, it's time for an abstraction. Talked about economics, now I'm gonna introduce the model that will be most of the remainder of the talk. So economics is applied in many disciplines. Economics is applied in finance.
09:21
When a lot of people hear economics, finance is what they think. If you think of the stock market, finance is the stock market, studying that, predicting that. Economics is applied in accounting. Accountants do a lot more than just bean counting, counting your profits and losses and everything like that.
09:42
They often have to get involved in the forecasting for your company and there are even accountants that specialize in doing post mortems of failed businesses. Why did this business fail? These are all applied economics.
10:00
Economics is applied in political science. There's an entire field called political economy that studies markets and their interaction with law, custom and government. Those three areas are the whole point of political science and the combination is called political economy. Economics is applied in neuroscience.
10:22
Interestingly, neuroscience is the study of the human brain but there's a subdiscipline slash interdisciplinary field called neuroeconomics that explains models, that uses models to explain phenomena in the brain. None of these really seem like they apply yet, however, there's another field that uses it.
10:42
Economics is applied in market research. Ah, that sounds like what we're looking for. Market research uses economic principles to build models for understanding customers. There are many different models for doing this. I picked one of them and it is called
11:03
the Van Westendorf Price Sensitivity Meter. This was originally outlined in a paper in 1976 for the European Society of Opinion and Market Research and it's still used widely today and that's the model that we're gonna use to model this demand side equation
11:22
for deciding how to price your products. The price sensitivity meter is easy to use. It doesn't require any special skills or preparation. It's easy to understand. The results of the model that when you apply it give you a very easy to understand set of information.
11:45
It's lightweight. If you have an audience, it's quick to get in front of them and there's not a lot of upfront preparation that you need to do and there's not a lot of afterward study that you need to do. It's based on surveys.
12:02
I think almost everybody here has probably done a survey in their life. They're used for all sorts of different things and this model will be no different. All right, so we have the name of the thing. What does it actually look like?
12:21
Gonna talk about the process for the Van Westendorf model. First, you need to decide on your pricing scale. You might be saying, wait, wait, wait. I thought I was learning how to price from this model. Why do I need a pricing scale? Well, presumably you've done some sort of market research to figure out whether your idea is actually going to work at all.
12:41
The pricing scale is based on that market research. You don't have to worry about the exact price but you have to know some sort of window that you're thinking. You use that as the basis, probably the median or the mean of the scale that you're gonna create and then you want to aim for 25 to 45 steps in this scale
13:02
so you have different buckets in the scale like zero to five dollars for instance. You want to aim for 25 to 40 steps. That gives you a great amount of fidelity in the surveys that you're gonna gather
13:20
and you want to try to capture both impossibly high and impossibly low values within this scale. You want to see an example? Think of pricing a granola bar. I had one of these for breakfast. The minimum step, we can set it say 50 cents.
13:41
50 cents is possibly an impossibly low value. You might need to go lower for this but for the example, let's say 50 cents. Can anybody here imagine paying $10 for a granola bar? I know I can't. So you have the bucket from zero to 50 cents
14:00
and the bucket from $10 and up are the two ends of your scale and then you have the scale in between. After you have that pricing scale, you need to survey your audience. When you're doing this, you want to start with a detailed explanation of the product or service. Actually, better yet, you really want to start
14:21
with a demonstration of the product or service. If you have a product already built and you're just wondering how you want to price it, that can be an actual demo of the product. If you're just getting started on a project, you can use the mock-ups that you've started and give a demonstration. The demonstration is just a much better way
14:40
to communicate what you're actually selling and I've found that using an explanation alone, people get confused about what it is you're actually selling. So I recommend using a demonstration. In the survey, you're gonna ask four questions. All of these questions start with at what price do you begin to think the product?
15:03
The first one is, is so inexpensive that you would question its quality. Would you buy a car that was brand new and priced at $100? I don't think I would. I don't know that I would trust that it was safe enough for my family. At what price do you begin to think the product
15:20
is a great deal for the money? It's positively worded but this is really trying to capture what they think is a cheap price for the product. This place is the happy place for the consumer. They think I'm doing so well, I'm a great shopper, I'm awesome and being in this pricing bracket is important for the customer.
15:44
Different people have different preferences though so you wanna ask this third question of whether or at what price you think the product is getting expensive but you still might consider it. In the case of a car, people want the car to be safe for their family
16:01
so they're probably willing to spend more for the car even if it feels a little expensive if it's demonstrably better in the safety category. The fourth question is the end of the scale and you wanna ask at what price is it too expensive to even consider?
16:20
Talking about cars, Mercedes are kinda too expensive for me to even consider. They're just out of my price range. People think, oof, really that's the price? Yes, and then they don't buy your product. These four questions gain us four pieces of information from each respondent.
16:41
The price that they think something is too cheap, the price where they think it's cheap and they think I'm awesome for buying this, the price where they think it's expensive but they might still buy it because it's worth it to them and the price where you've just priced them out of it. I'm not made of money.
17:03
You wanna ask the questions in both increasing and decreasing order between different participants. There's a psychological effect called anchoring where when you list a first number, say one dollar, if you're asked to list a number higher than that,
17:21
you get psychologically anchored to one and your number will be lower because that is a low number. There are people that talk about using this trick for job negotiation where you just say, man, did you see the million stars last night and that anchors people at the higher value. So you want to eliminate the bias here
17:43
by asking people in both directions. You're gonna ask your friend Tom in increasing order and then your friend Sally in decreasing order and the idea is that the bias in their answers is going to balance each other out. So in the second example, you're gonna find out
18:03
too expensive first and then expensive and then cheap and then too cheap and hopefully doing that right will eliminate the bias in your results. If you get impossible answers, the original paper says just correct them. If somebody says that it's too expensive
18:22
for them to buy at $30 but that it's getting expensive but that they would still consider it at $50, the original paper says just correct them. And they were just mistaken in the way that they spoke. If anybody here is in machine learning, that might sound like a failure but they say that the, a failure in the cleaning
18:41
of the data but they say that the skew that is introduced by this is about 3%. So you can either throw out the data or you can correct it either way. All right, so we've surveyed a bunch of people and we have a data set.
19:00
What do we do when we have a data set? It's now time for some, brace yourself, statistics. I used to teach students and Kermit here was the exact opposite when I said the word statistics. Yay, statistics! Statistics can be intimidating to some people
19:20
so I'm gonna go through the one piece of statistics that you need. Cumulative distribution functions. The idea here is that you're gonna come up with a function where the right-hand side is equal to the probability that a random variable is less than or equal to x. Have I lost you yet?
19:42
Let's break this down and start to pick it apart. What's the first thing in here that pops out? What the heck is a random variable? The random variable in this case is the observations that you picked up from asking people the four questions like the people's preference on what cheap is for them.
20:02
Respondent one might have said that cheap is $1.22. Respondent two might have said $9.99. Respondent three might have said $4.88. This is your random variable. It's sampled from some distribution of the actual cheap curve in the universe for your product. That's all it means when it says it's a random variable.
20:22
So if we're gonna represent this in Ruby, if you need to write a script to do this for you, cheap is just the observations that you have of your variable. All right, so that's the random variable. What can we pick apart next? How about this x?
20:40
People often get intimidated by seeing variables when they're talking about math. So what is x? X is the maximum values of each step that we made in our pricing scale. Let's go back to our granola bar example. In this case, x is the vector of things on the right-hand side here. It's 50 cents, and then $1,
21:02
and then $1.50, all the way up to $10. And you're looking for the probability that the thing that they gave you is less than each step in this. So we can just name that as the x. In Ruby, we're just gonna generate the array of our steps and that will be all we need.
21:24
So we have the random variable and we have x. What's next? Ah, a function. I know what a function is, I'm a programmer, right? We use those in programming. All right, so we know we're gonna apply the function to every step in the scale.
21:40
So it's called a CDF, so I'm gonna name the function CDF and I'm gonna pass in my observations. Well, we need to define the CDF function. What's it gonna do, though? Let's go back to the definition of what CDF is. So it's a function where the right-hand side is equal to the probability that it is less than or equal to x.
22:03
So let's write that. We have our framework here. The first line here, when you're calculating a probability, is you're just gonna take the ones that match and count them. So in this case, we're gonna select all the observations where they're less than or equal to the x that we passed in and count the result.
22:23
To get the probability, we're gonna divide by the total number of observations. You need to do it as a float because otherwise it's gonna be an integer division. And that gives us everything that we need for CDF in what, eight lines of Ruby code. Guess what?
22:41
We just did some statistics. All right, so we have the results of our doing statistics and we now need to start graphing CDFs. Humans are an inherently visual species so it really helps us to see the results of the data
23:03
in order to understand what's happening. So first we wanna graph the expensive and cheap curves and we wanna look for the intersection. So let's do that. In this case, we have the intersection
23:23
where the orange cheap curve and the blue expensive curve happens. This represents what's called the indifference price. The indifference price may be the median price in the market if it's a well-known market. For example, if you're doing,
23:41
if you work for DigitalOcean and you're doing hosting for servers, people generally know what servers cost. So that might be the median price for what you're asking in the market. It may also be the price of an important market leader. In the case of the granola bar, I had a kind bar this morning because they're really trendy right now.
24:02
That might be the market leader. It might be the price of the market leader that you're measuring here. Indifference price also indicates price consciousness. We're not very intentional about how we think about prices a lot of the time but we have this concept of price consciousness where something feels like it's too expensive or too cheap.
24:22
It can also indicate the level of price consciousness in your population. So let's look at the graph again. In this case, the intersection shows that at $80 a month, 46% of people
24:41
think that the product is either expensive or cheap. That means that only 8% of people think that the $80 a month is normal. This is pretty high when it comes to the indifference price and it means that there can be some confusion about what your audience is actually understanding when you're asking them.
25:00
I'll get into that a little bit later and why we ran into this problem. Next, you want to look for psychological thresholds. When we look at this, we're looking for the steep curves in the graph. So in the cheap curve, there are large jumps in the $50 to $60 slot,
25:21
the $100 to $110 and the $150 to $160 slot. This means that people are very sensitive to the price at this level. You lose a lot of people thinking that it's cheap in these jumps. Threshold values, these psychological thresholds, are why things are often priced at 99 cents.
25:40
It's easier to think that something is 9.99 than $10. When we look at the expensive curve, we see the same kind of behavior. At 90 to 100, 140 to 150 and 190 to 200, we see large jumps in the number of people that think the product is starting to get expensive.
26:02
All right, so we've looked at these two curves. Next, we want to graph the too expensive and too cheap curve. Here's the results. Again, we're going to look for the intersection. In this case, it's the optimal pricing point that we're looking at. The optimal pricing point indicates
26:21
that there's minimum resistance to change. The fewest people drop off when you change from this value. If we look at what we're graphing, it's the too expensive and too cheap. So at this level, 26% of people think it is either too cheap or too expensive.
26:41
Small shifts from this point mean that you won't lose very many people to thinking that your product is too cheap to be quality or too expensive to be worth it. Again, you want to look for the psychological thresholds. In this case, we've got three of them on the too cheap curve and a bunch of them on the too expensive curve.
27:03
Just helps you get a better feel of what people are thinking. Next, we want to graph all four of these curves together. We have a kind of messy graph, messy looking graph here, but we still have the same information. If you have a wide difference
27:20
between the optimal pricing point and the indifference price, it can indicate that people are sharply sensitive to pressure in the market. This generally means that the cost of your market has changed sharply recently and people are scared of pricing. Next, we want to create the not expensive
27:42
and not cheap curves by inverting their counter points. We take the expensive curve and the cheap curve and we invert them to get not expensive and not cheap. Then we combine them with too expensive and we combine them with too cheap. Just like this.
28:02
These intersections are your point of marginal cheapness and marginal expensiveness. I love that word, expensiveness. Looking here, we have marginal cheapness on the left and we have marginal cheapness on the right. Both of these are the equal number of people think it's either not expensive or too expensive
28:22
or not cheap and too cheap. These points form the range of acceptable prices is what they call it in the literature. So when we're looking here, the acceptable prices are between $50 and $100 on this graph.
28:41
All right, so now what? You want to graph all six curves together and look at what you have. You have your range of acceptable prices. We know the range that the customers think is acceptable for your pricing. You also have the potential median price for the market and you have the minimum resistance to change.
29:03
Now it's time to make your decision. It's still gonna be a little hard but at least you have information about what to do now. You can use your reason and think about what you want to optimize for here. Now I have a question for you. Are prices inherently adversarial?
29:25
Often it feels like companies are trying to extract the maximum value out of every customer but pricing doesn't have to be about that. Often it works better if you price for some consumer surplus.
29:42
Happier customers spread word of mouth and are more fun to interact with. If you have a happy customer and they have a problem with your service, they're gonna come in less hot when they have to write a support ticket to you. Customers can often become your greatest champions.
30:00
Some companies spend zero dollars on marketing and rely entirely on word of mouth to drum up services or to drum up more business. They do this through the goodwill that they get through pricing via consumer surplus. People think that they're great at shopping and that they're awesome and that makes them feel good which makes them want to represent your company.
30:25
All right, I mentioned before that there's some gotchas with this that we encountered. The first thing that we did is we forgot to actually define the scale before we surveyed. You really wanna do this. It requires some upfront work where you need to have some idea
30:41
what you're gonna price but it really helps. In our case, we had one person that we surveyed that said that they're willing to spend $10,000 a month. That's not our market. That's not the market that we're going after and that ended up just being noise for us. Defining your scale beforehand allows you to say
31:01
the maximum thing that we wanna charge is say $500 and you won't get impossible values like $10,000. That wouldn't be realistic for the rest of your market. When possible, you wanna segment your participants into logical groups. In our case, we have three populations that we really market to.
31:21
We have individual designers that are looking to host a WordPress website. We have design agencies that have a bunch of designers and then we have enterprise customers that are large customers that expect a lot of service. If you include all of them in the same survey and all of the results in the same data set,
31:41
then you end up muddying the results of your study because all of these groups have different price consciousness and price sensitivity. The enterprise customer is willing to throw money at the problem because they have money. The individual designer who's just scraping by or is just getting started doesn't have that kind of capital available
32:02
so you want to segment them. Also, remember within each segment, you wanna segment it further and ask in both the increasing order of prices and decreasing order. And again, that's to eliminate that anchoring bias that I spoke about.
32:21
Make sure that you're very, very clear with your description. You don't really scratch that. I said this earlier but I wanna say it up again. I wanna say it again. Conduct your first surveys with a mock-up if you can, if possible. If you're writing an ebook or something, maybe give them a sample and say, based on the table of contents and the sample,
32:43
what are your expectations for this? You're gonna get better results this way because words are confusing. Wording is hard and you wanna eliminate that problem. Okay, so we have a model now. Let's talk about what we learned
33:02
from studying this model and applying it. What kind of partnerships would we be looking for? Remember, this is one simple model, just one. If you go and read the paper, there's actually lots of different things that you can do with this model.
33:20
You can measure the power of your brand. You can measure the power of your discounting strategy, like how much should I discount something to increase sales? And you can also measure the value of bundling things together. If you're selling multiple things, you might wanna bundle them together and you can measure how you can do that. One model, it's from a discipline that's outside our own.
33:43
We're all software developers. If you recall David's opening keynote yesterday, he said that we need to liberate the best ideas and learn from what others are doing. In this case, I went and learned this. But how cool would it be
34:00
if you could get somebody who had this experience, teach them how to be a developer, and let them live their own destiny and make a new friend? You can use this for intentional design. We're great at abstractions at developers.
34:20
Computer science is all about abstraction. It's the study of abstraction. But that abstraction sometimes gives us tunnel vision when it comes to modeling the humans within our system. I recoil every time I hear the word user in my system because it doesn't really adequately explain what the human is doing for your system.
34:43
So using models that are better equipped for handling humans allow you to design more intentionally. When you design more intentionally, you can help your customers be awesome. Being awesome is one of the things that we all want. Everybody craves to feel like they're great
35:03
at what they're doing at any given time, whether it's grocery shopping or your actual work or being a parent, being a sibling, being a child. Everybody wants to be awesome and you can help people be awesome through doing this.
35:20
Consider thinking of your customers as partners instead of customers. Having partners that are willing to go to bat for you, as we heard in the last talk by Todd and Justin, they get most of their business from people that used to work for them
35:40
or from their current customers, just sending business their way. If you think of your customers more as a partner, you can get that value as well. All right, hopefully I've left some things for you to think about. Again, my name is Michael Herold. I am the lead application engineer at Flywheel.
36:01
If you think you want to help creatives do their best work, please come talk to me or anybody in a Flywheel shirt. We have a bunch of people here. Thank you for your time.