Blog - Entrepreneurs with Denis Oakley

This week we are talking about the Retention Science business model and using the business model canvas to understand the different components of their model. For those of you who don’t know Retention Science is a marketing company that helps brands retain their customers.

This is done through the use of Artificial Intelligence (AI) and machine learning to personalise every single customer’s experience of a brand so that churn is reduced as much as possible.

So who are Retention Science’s customers?

Let’s get the obvious points out of the way first. It’s AI so it needs lots of data – so it’s customers are internet-based companies with a certain amount of scale. For the machine learning to work well we’re looking at 100k customers or more. Because the business model focuses on Retention (Hat tip to the AARRR model typically we’re looking at businesses with a regular recurring revenue model.

That is normally going to be SaaS companies or companies selling subscription boxes. The customer they trumpet most is Dollar Shave club – known for its super cool videos (I’m still on Gillette [sigh] as no one in Malaysia has thought to clone that idea)

So we’re looking at CMO’s of data-driven companies with traction who have got a powerful brand working but need to stop leaks in their retention. That is taking us onto the value proposition.

Retention Science’s value proposition is

We reduce your churn rate by personalising your customers experience in a way that cannot be done manually.”

Problems Solved by the Retention Science Business Model

Marketers have several problems here:

Producing content is expensive and time-consuming.
Putting the content in front of the right segment is difficult.
It’s almost impossible to do this other than as one-off projects without massive resources.

What Retention Science does is to segment customers on their behaviour in real-time and based on that behaviour show them the content that has the most success based on customers exhibiting similar behaviour in the past.

Retention Science doesn’t show their pricing on the website (suggesting that they are not going for the $50/month market) but it’s clear that something like this can automate the work of several marketers if they were hired to attempt to do the same thing. More importantly not only is the cost of hiring marketers to do this cost, but the system continues to get better and better over time and react to emerging patterns of behaviour. So there is a clear ROI case that can be made as you cut marketing heads and reduce churn to increase revenue.

Customer relationships are basically consultative selling. The company uses demos to qualify customers and there is likely to be some integration involved to connect the customer’s web platform and CRM through to the Retention Science’s servers via API. The customer relations can really be seen as a standard SaaS sales, account management and customer support model for larger organisations with little Face to Face time.

Channels

Key Activities

ReSci’s main channels are the web site, where there is extensive content marketing in place and its partnership agreements with agencies. Many customers will be engaging in a larger project to improve website performance and for all its benefits ResSci is not a one-stop solution. The one-stop solution is delivered by agencies, technology vendors and design houses delivering projects for their customers. By partnering with them Retention Science is able to increase the value of their offering and get the partner to do much of the expensive face to face sales on their behalf.

The key activities for Retention Science’s business model are looking at customer data streams and looking for new products that can be sold off the back of them. What can be learnt from the data? How can this learning be transformed into a product? Does this product work and add value? The three current suites of products – email, website personalisation and subscriptions have all been developed in this way.

The other key activity will be the integration and implementation – making sure that customers can be set up quickly – allowing a generic platform to be customised to the client requirements as quickly as possible. This is again likely to be very much a coding activity.

Key Resources

Now that it has been up and running for a few years the key resources in the Resolution Science business model are the data that have from multiple sources and the IP contained in its algorithms. That’s not easily or quickly copyable by competitors. Lesser resources will be the human capital but whilst this is important AI skills are becoming more fungible as the industry grows in size.

Partners

Agencies and technology vendors are key partners as will be the cloud service provider that Resolution Science users. Finally, it’s worth considering the big data/AI community as a partner given the huge reliance on Hadoop and other big data tools

Costs

Product development and design are going to be the largest single cost with a rented IT infrastructure coming in second. The third main cost will be the SaaS sales team. Sales is often a larger component but this seems to be due to relative ease of hiring technical sales as opposed to engineers in Santa Monica. Finally, there will also be sales commissions to its partners.

Revenue

Revenue in Retention Science’s business model is based on a three-year contract implying large front end set up costs which may or may not be clawed back from the customer at startup. Otherwise, the focus is on pure recurring revenue for each of the different products

There have been two incredibly hard things in my life as an entrepreneur. They both begin with F. Failure and Fear.

Failure

Success is commonly seen as creating a unicorn and having a bank account bulgin with billions. If you don’t achieve that, if you stumble and fall along the way then it sometimes seems if you are a failure. The mistake here is that if you take this view then you are relying on your achievements as myour measure of success. Achievement is fickle and rarely under our control. The activity that you put into creating achievements or otherwise is a far better measure.

Much of my best work never delivered. The market changed. The company went bust. We pivoted. Was I responsible for the market changes? No. If it hadn’t changed I’d have a shiny medal. The reality is that I don’t. The reality is that as an entrepreneur this is really really common. We look at the stars and expect dreamlike rewards. The truth is that we mine diamonds and we need to shine and polish them as they are the true measure of our value.

What we do is incredibly valuable; then, we do it again and again.

Fear

Or we try to. Fear is a brilliant tool. It is super power at stopping us doing stupid things. If it hurts, fear kicks in and stops us doing it again. Great for teaching us to stay away from cliffs, bicycles and rivers in flood. The truth is that as entrepreneurs we need to go to these places. We need to go back to them time and time again and the fear makes it harder for us to do so. Worse we get platitudes that tell us to overcome fear but don’t tell us how to cut through the scar tissue that forms over time.

For me the secret lies as much in sidestepping the issue as in facing it directly. Does it scare me? Yes? Will it kill me? No (if yes – such as downhill biking I put on the brakes). Then try it. Push to overcome that discomfort at every moment. I’m always in awe of how Napoleon and other generals ignored canon and musket fire, appearing unconcerned as they got pelted with dirt and had horses shot from around them and aides mutilated. Fight the little battles first and then big ones are easier. The epiphany in my life was mundane. Cold water. I used to scrunch myself up before having a cold shower. it was a horrible experience. Now I just walk into the cold stream of water and accept the sensation.

Each battle with fear gives us confidence

One of the things that I do is growth hacking for solopreneurs. Growth hacking is an overused phrase. To me it has a great resonance. I live on the very edge of one of the largest untouched jungles in SE Asia. Seen from a distance it can look beautiful and your path can seem clear. It’s not that easy. Pretty soon you have no contact with the outside world, visibility reduces and it is quite simply trackless.

My parang (the Malay word for a machete) is a vital tool and I use it to help make and find my way through the dense vegetation. In the same way I help my clients find their way through the dense and confusing marketing options that are available for them and craft a strategy that lifts them up into the sunlight.

I’m feeling rather poetic today!

Growth Hacking for Solopreneurs – The Process

Like most good things it’s pretty simple. The following are the key stages in growth hacking for solopreneurs

Understanding the business model and the value proposition
Considering the ideal customer
Thinking about the many marketing channels available and brainstorming ideas for each of them
The prioritising and taking the best ideas to run simple low cost experiments

The first stage is simple. I need to know and understand your business. Many of my customers have weak target customers. They may say small businesses or women. These are huge audiences with lots of distinct segments within them. Trying to market to them en masse means that your bland boring message disappears and gets lost.

creating a persona when growth hacking for solopreneurs — *first thoughts as I create a persona*

So often one of the first tasks is to figure out who the ideal customer really is. Sometimes this means deep dives into Facebook or Google Analytics trying to put a personas together. Other times it is reading amazon reviews or listening to you talk about your first or second customers. Whichever it is I then create beautiful detailed profiles of ideal customers which we then use as the baseline of the marketing experience. These ideal customers are detailed enough that they almost seem real. I can tell you what Emilia’s handbag looks like, or what type of glasses Virgine D wears – even though I have never thought about that before – that’s because they are real people who you want to sell your goods or services to. (They are both favourite persona’s I’ve done for customers)

Once we know who we are selling to we then start looking for them. I have a list of 19 generic marketing channels – ways of reaching customers. We quickly think about how easy it is to reach the customer using that channel and how to do it. There are lots of millennials on the internet and because every marketer and her pet cockatoo are chasing them the obvious ways are expensive. That’s cool if you have unlimited resources and nerves of steel. Really it is. The only problem is that Google and Facebook will get rich on your money.

There’s a better way though. That’s to look for trails that lead us easily through the undergrowth and avoid consumer’s resistance and price wars with other marketers. Some of the approach is based on systematic processes, parts on stealing other good ideas that have not been worked out and some is based on imagination and daydreaming. Whichever part comes to the fore it is all aimed at finding some way that will effectively allow you to acquire customers at a much lower cost than through frontal assault.

Unfortunately it’s not that easy. We don’t know whether the approach will work, nor do we really know whether consumers will love your product or service and how we deliver it.

So we never use big fancy marketing campaigns. These take lots of time and money and if it goes wrong you haven’t got very far. A far better approach is to do lots of small experiments to test and learn. How can we spend $100 on a SEO experiment that will make a big impact? We think about it and have a hypothesis. Then we spend the $100 and see what happens. It’s ll data driven and if we see no results we’re a bit disappointed but the world hasn’t ended. On the other hand….

If it’s good we don’t celebrate too much. We go back and spend $500 to see if we can reinforce the success and do it on a larger scale. When that works. We spend $2000, or $5000 confident now that the approach will bring you customers and sell your products.

Why does it Work?

Small cheap steps that you repeat until we find something that works. Its systematic, effective and far far less risky than amost any other approach

That’s the story of how I do growth hacking for solopreneurs

Why am I good at it. I’d like to say it’s because I’m good at hacking through jungles. It’s probably because I have run so many small scale frugal campaigns over the years for tiny struggling startups. That’s given me a wealth of experience. When you combine this with super curiosity and a love of awesome stories you often get results.

If this sounds like it would help you – send me a mail – denis@denis-oakley.com

BY TAYFUN USLU – FINANCE EXPERT @ TOPTAL

#Freemium #NetworkEffects #Pricing #SaaS #Software-as-a-Service

Pricing is one of the most important financial levers that companies have at their disposal to influence the financial success of their business. However, it is not an easy task.

Firstly, pricing affects multiple stakeholders, both inside and outside the company. Experimenting with pricing is therefore not a task that should be taken lightly. Secondly, finding the right price is notoriously hard. Customer preferences are hard to gauge ex-ante, and internally, it can be difficult to foresee the effects of pricing changes on the financial performance of a company. And finally, of course, pricing doesn’t happen in a vacuum. In any competitive market, pricing changes can often lead to retaliatory actions which end up canceling out the intended effect of the pricing adjustment.

With this in mind, techniques and methods for getting around these challenges are extremely useful.

In my experience, having co-founded a developer tools startup that uses a SaaS model, I have come to appreciate the SaaS business model as being, amongst other things, very pricing-friendly. The intrinsic characteristics of SaaS and its delivery mechanism have important ramifications related to pricing which, in turn, can be extremely useful in the financial management of your company.

In this article, I explore in more detail what these are and how they can benefit your business from a financial analysis and management standpoint.

What is SaaS?

Software as a Service (SaaS) is a software licensing and delivery model in which software is licensed by a third party and delivered to clients through the internet. Compared to locally hosted software models, SaaS clients do not have to install the software, update it, maintain it and integrate it. The vast majority of technical aspects are “taken care of” by the SaaS provider so that the client can start using the SaaS product with little effort.

The ramifications and benefits of SaaS versus other types of services are displayed graphically in chart 1 below. As you can see, with traditional on-premises software, the client has to manage most of the activities related to setting up and running the software. At the opposite end of the spectrum, SaaS takes care of all of this on the software provider’s side. There are shades in between; for instance, Infrastructure as a Service and Platform as a Service.

The pros and cons of SaaS are fairly straightforward. On the one hand, SaaS is much easier to set up and run. It doesn’t require local servers, storage, management, etc. On the other hand, it doesn’t allow the same level of customizability that on-premises software can provide.

The change from locally hosted software to SaaS has been happening for some time, and is part of a more general shift in the IT industry to cloud-based applications. According to FTI Consulting, 69% of businesses today use at least one cloud-based application.

The shift to SaaS from on-premise services can be seen nicely when one looks at Adobe, one of the industry’s most well-known players. As can be seen in chart 2, courtesy of Tom Tunguz, Adobe’s non-SaaS product revenue peaked in 2011 at $3.4 billion and halved to $1.6 billion in just three years. Such a drop in revenue would normally have huge ramifications, but Adobe has also been offering a suite of SaaS products, which have increased in revenue five-fold, from $0.45 billion in 2011 to $2.1 billion in 2014. Adobe’s changing product mix is indicative of the overall industry’s shift away from locally hosted software towards cloud-based software provision, such as SaaS.

All indications are that this trend is likely to continue. Gartner estimates that SaaS application software was a $144 billion market in 2016 and that, by 2020, businesses will be shifting to cloud-based software to the tune of $216 billion a year.

SaaS Is a Pricing-friendly Business Model

As mentioned, the reasons for this shift to cloud-based applications are manifold. Lower setup and maintenance costs are of course a big factor. However, one important benefit that I have found of SaaS-based applications is that they are very pricing friendly. This is win-win both for clients as well as for the businesses providing the service, and I believe it plays an important role in explaining the overall industry shift.

In particular, there are several intrinsic operational characteristics of SaaS as a business model that in turn have implications on pricing, and specifically create and magnify the number of pricing levers that businesses have at their disposal. In particular, I highlight four main pricing-related benefits of SaaS products:

Flexibility: Because of the web-based, real-time nature of the delivery mechanism, SaaS products are very flexible and allow for constant iteration. This affects pricing as well, since changes to pricing can be relatively quickly implemented, and the impact of these changes can be immediately measured. One can keep iterating until the desired price is reached.
Real-Time Usage Tracking: Since SaaS products are in the cloud, usage of the software by the client is tracked in real time on the service providers’ side. This means that SaaS companies are more able to charge on a per-usage basis than traditional on-premises companies. This in turn has several important benefits detailed below.
Network effects: The delivery mechanism and frictionless implementation of SaaS products acts as an enabler of network effects. This has important pricing ramifications in that it makes it easier to both create and charge for these network effects, thus in theory maximizing monetization.
Freemium: Due to near zero marginal costs of delivery and product hosting, SaaS companies are more able to allow clients to use certain products for free for months or even years until they upgrade or cancel the service. Freemium prevents the product pricing from being a barrier to users trying or adopting the product, thus increasing the number of potential clients.

Benefit No. 1: Pricing Flexibility

The first and arguably most important financial benefit of adopting a SaaS model is the inherent flexibility that the model provides. In particular, flexibility refers to the following attributes:

Fast iteration: Because of the delivery mechanism, SaaS allows you to iterate quickly on pricing changes so that you can test price points and packages until you reach the optimum mix.
Tiered pricing: SaaS lends itself very well to tiered pricing, meaning the ability to price different product offerings differently to address multiple customer sub-segments.
No middleman: Since SaaS is delivered right to the client, bypassing any middlemen. This removes a key stakeholder from the equation and again gives the company greater flexibility to test different pricing points and packages

A practical example of this can be observed by analyzing Zendesk’s pricing evolution. Zendesk is a very successful customer service software company that generates over $300 million in yearly revenue. It offers different features of the same product to different customer segments.

As seen in the example below, in January 2015, Zendesk offered five different plans for five different segments. Its cheapest plan was available at $1 per agent per month while its most expensive was at $195 per agent per month. It also directed the users towards the “Plus” plan by highlighting it in green.

Exhibit 2: Zendesk January 2015 Pricing Options

Source: Archive.org

In January, 2016, Zendesk must have felt that there was room for improvement in pricing; they decreased the total number of plans from five to four. They also made changes in the pricing of plans where the least expensive plan increased from $1 to $5 per agent per month while its most expensive plan decreased from $199 to $99 per agent per month. The names of the plans were also changed (e.g., “Starter” to “Essential”) and the users were directed to the “Professional” plan by a lighter highlighting technique.

Exhibit 3: Zendesk January 2016 Pricing Options

Source: Archive.org

In 2017, pricing changed again and the number of plans increased back to five. While the least expensive plan stayed at $5 per agent per month, its most expensive plan increased from $99 to $199 per agent per month. By removing highlights, Zendesk stopped directing people to any of the plans.

Exhibit 4: Zendesk January 2017 Pricing Options

Source: Zendesk Pricing Page

Zendesk unfortunately doesn’t disclose the specific effects that its pricing changes have had. Nevertheless, the more general point that flexibility is useful is summarized nicely in this post: “Finding the right balance between value and revenue – your ability to help customers and be fairly compensated for that help – will make or break your […] company.” No doubt Zendesk’s changes are a reflection of this point.

To sum up, SaaS, as a delivery mechanism, allows for a great deal of flexibility and iterating capacity when it comes to pricing. As Lincoln Murphy of Sixteen Ventures puts it, “Pricing (like pretty much everything) is never a set-it-and-forget-it situation,” so the ability to iterate, test, learn, and improve is a very useful feature of the SaaS business model.

Cohort Analysis

If the results of pricing changes cannot be measured in a definite way, there is no point in iterating. So it is important to set up appropriate data collection and measurement processes and tools from the start.

One of the ways of measuring the results is through a cohort analysis. In cohort analysis, customers are put in different cohorts (i.e., groups) based on a characteristic, typically a selected time period. All of the new customers signing up can be divided into cohorts based on the date they become a customer. For example, there could be different cohorts as customers who joined in January, February, March, etc.

Analytics tools like Mixpanel readily offer cohort analysis based on the data they gather. As seen in the example above (Table 1), retention is being analyzed to measure how many people keep using a fictitious product as time passes. Going into the details, the cohort of April 27, 2014 has a retention rate of 23% at week 6, meaning that 23% of people who signed up in the week of April 27 2014 are still using the product at week 6. The cohort of July 13, 2014 has a retention rate of 9.6% in week 6, indicating a significant drop in retention rates, which means a major problem with this cohort or an action taken within that timeframe.

The critical aspect of cohort analysis is availability of data, so it is crucial that the data that will be the basis of the cohort analysis is collected correctly and available when needed.

Less Is More

In a well-known study conducted by psychologists Sheena Iyengar and Mark Lepper, it was showed that excessive choice can produce “choice paralysis.” As explained in this HBR article:

“[…] shoppers at an upscale food market saw a display table with 24 varieties of gourmet jam. Those who sampled the spreads received a coupon for $1 off any jam. On another day, shoppers saw a similar table, except that only six varieties of the jam were on display. The large display attracted more interest than the small one. But when the time came to purchase, people who saw the large display were one-tenth as likely to buy as people who saw the small display.”

Due to tiered pricing, the absence of intermediaries, and the fast iteration capabilities of SaaS, a product could be offered in hundreds of different ways. Nevertheless, one has to be careful not to create an instance of choice paralysis; thus, it is paramount that pricing be approached from a design perspective as well.

A practical example of this can be observed with Hubspot. Hubspot is an inbound marketing and sales management platform that automatizes and makes email marketing, analytics, and SEO more effective. PriceIntellingently does a nice job of explaining how although Hubspot’s pricing could be displayed in a variety of ways (e.g., a moving bar that shows the increase in pricing as the number of contacts increase), only three different plans were created based on three different value metrics. This makes these plans easier to understand and track.

Benefit No. 2: “Per-usage” Pricing

The second key benefit of SaaS products is that because of the delivery mechanism, SaaS lends itself much better to per-usage pricing. This is because usage by the client can be tracked and measured appropriately since the software is run on the service provider’s side.

Per-usage pricing is a very useful pricing tool for the following reasons:

It creates a lower barrier to customer adoption.

Users face a lower up-front financial barrier to signing up for the service since they know that their expenses will be in line with their usage. In contrast, traditional on-premises software usually charges a flat up-front fee based on an estimated assumption of usage. This estimate may not necessarily correlate with the specific needs of the client in question.

As outlined by Reason Street in this article, “Because of low or no cost startup fees, enterprise customers that typically take six months to a year to make a decision about enterprise software often leap frog this decision and simply begin to try the service. Customers then increase their use based on actual need of the company (vs. paying up front for licensing fees per seat).”

Costs can be better aligned with revenues.

Per-usage pricing also allows the service provider to better align its own costs with its revenues, since the revenue it receives is tied to the costs it incurs to charge the service. As Reason Street again points out, “For software startups, particularly in the SaaS sector, once the initial minimum viable product is built, companies can scale revenue closer to the scale in costs required to deliver the service. Some SaaS companies prefer pay-per-use, because the value relationship is made very clear to the customer, and they are able to prioritize feature and service development based on the customer’s expressed needs.”

Per-usage pricing leads to greater stickiness.

Per-usage pricing often leads to greater customer stickiness for two important reasons.

Firstly, since the customer’s expenses are spread out over a great period of time, their costs at any given point in time will be small compared to other costs that their businesses incur. As a result, they may not be as tempted to cut these. Of course, if the client were to view the costs as a whole (or over a longer period of time) they may well notice that the total cost is much larger, but human psychology doesn’t always follow logic.

Secondly, since per-usage pricing leads to lower barriers to adoption (as explained above), customers are more likely to give the service a try. As they continue using the service, the software may become more embedded in the everyday processes of the company, making it more difficult and costly to switch providers. Andreessen Horowitz sums this point and the related benefits up nicely:

“Once a SaaS company has generated enough cash from its installed customer base to cover the cost of acquiring new customers, those customers stay for a long time. These businesses are inherently sticky because the customer has essentially outsourced running its software to the vendor […]
It’s very difficult to switch SaaS vendors once they’re embedded into business workflow. SaaS customers, by definition, made the decision to have an outside vendor manage the application. In the perpetual license model, in-house IT staff managed all software instances and thus could incur the internal costs to switch vendors if they so chose.”

Per-usage pricing allows for subscription pricing, which helps with financial planning.

To be clear, per-usage pricing is not necessarily always good. There is a clear cash flow tradeoff from switching from up-front licensing to per-usage pricing. This effect is again illustrated nicely by Andreessen Horowitz:

“In the traditional software world, companies like Oracle and SAP do most of their business by selling a ‘perpetual’ license to their software and then later selling upgrades. In this model, customers pay for the software license up front and then typically pay a recurring annual maintenance fee (about 15-20% of the original license fee). Those of us who came from this world would call this transaction a “cashectomy”: The customer asks how much the software costs and the salesperson then asks the customer how much budget they have; miraculously, the cost equals the budget and, voilà, the cashectomy operation is complete.
This is great for old-line software companies and it’s great for traditional income statement accounting. Why? Because the timing of revenue and expenses are perfectly aligned. All of the license fee costs go directly to the revenue line and all of the associated costs get reflected as well, so a $1M license fee sold in the quarter shows up as $1M in revenue in the quarter. That’s how traditional software companies can get to profitability on the income statement early on in their lifecycles.
Now compare that to what happens with SaaS. Instead of purchasing a perpetual license to the software, the customer is signing up to use the software on an ongoing basis, via a service-based model — hence the term “software as a service”. Even though a customer typically signs a contract for 12-24 months, the company does not get to recognize those 12-24 months of fees as revenue up front. Rather, the accounting rules require that the company recognize revenue as the software service is delivered (so for a 12-month contract, revenue is recognized each month at 1/12 of the total contract value).
Yet the company incurred almost all its costs to be able to acquire that customer in the first place — sales and marketing, developing and maintaining the software, hosting infrastructure — up front. Many of these up-front expenses don’t get recognized over time in the income statement and therein lies the rub: The timing of revenue and expenses are misaligned.”

This effect is displayed graphically in Charts 3 and 4 below:

Nevertheless, as can be seen clearly from the chart above, the major benefit of subscription pricing is that it provides a much steadier, more predictable cash flow cycle, which in turn can be of great help when it comes to financial planning. This helps to explains why the vast majority of SaaS products are priced on a subscription/monthly basis.

Benefit No. 3: Create and Charge for Network Effects

Another important pricing benefit of SaaS products is that, since these lend themselves well to network effects, this has useful implications for pricing. To be clear, when we say that SaaS “lends itself well to network effects,” what we mean is that a) SaaS can be an enabler for creating network effects, and b) SaaS as a delivery mechanism is often more practical for activities that benefit from natural network effects already. Before moving on to looking at the pricing implications of this point, we delve a little deeper into how SaaS lends itself well to network effects.

SaaS as a Powerful Enabler of Network Effects

As mentioned, the specific attributes of SaaS products often mean that these can be powerful enablers for network effects. Specifically, there are three reasons:

1) SaaS makes for good collaboration software, since it is delivered over the internet.

By using the internet, rather than locally hosted delivery, SaaS software allows people to collaborate in real time (e.g., Google Docs) and view things simultaneously. It also avoids interoperability issues between different users (perhaps due to their operating systems), again making for easier collaboration.

As pointed out here, the ease of collaboration for SaaS products is “why collaboration is one of the most heavily populated and fastest-changing SaaS categories.” Examples of companies in this space include services like Campfire for team collaboration, Basecamp for project management, AnyMeeting for web conferencing, Vidyo for video conferencing, Box for storage, and Google Apps for productivity.

2) SaaS makes it easier to work with external sources, since there are fewer interoperability issues.

Most businesses have to interact with multiple third parties, be it vendors, suppliers, freelancers, etc. When software is used to work together, interoperability of this software becomes an issue. SaaS solves this by removing interoperability (aside from interoperability at the browser level, which is fairly minimal in most cases). As outlined here, “there is the obvious one which is actually connecting businesses to each other. For instance, if your are in purchasing management and you connect companies to possible suppliers, such as Kinnek does. This creates market place like network effects where more suppliers create more value buyers and vice versa.”

A corollary to this is that, since SaaS is relatively easy to set up and get running, it adds less friction when convincing other parties to run the same software as you, which again makes it easier to create network effects.

3) SaaS allows for data-enabled network effects.

Since SaaS allows for real-time data tracking, it can also lead to so-called data-enabled network effects. In this fascinating blog post by Tomas Tunguz, Venture Capitalist at Redpoint Ventures, he outlines this point (specifically referring to SaaS Enabled Marketplaces) as follows:

“Because SEMs deploy SaaS to both the supply side and the demand side, these companies can develop an exceptional understanding of their market. Access to supplier data and consumer demand provides four key advantages to SEMs.
First, SEMs understand the supply/demand curve at every second […] Consequently, if there is a supply/demand imbalance, the SEM can [step in to] balance the market place.
Second, SEMs understand the operational excellence of their suppliers […] The SEM can better qualify and match supply and demand.
Third, the accumulated data on both consumer demand and supplier performance compounds into a unique data asset that erects a moat around the business.
Fourth, because of the visibility into supply/demand, a deep understanding of supplier excellence, and the ability to identify the right types of buyers and sellers, SEMs benefit from a more efficient go to market.”

The Pricing Benefits of Network Effects

The pricing benefit of the above is that SaaS makes it easier to price products so that network effects are at first encouraged, and later taken advantage of. Specifically, because of the delivery mechanism, SaaS makes it easier to implement pay-per-account or pay-per-user pricing systems.

The benefits of per-user pricing with relation to network effects can be summarized as follows: Since on a per-user basis the cost is relatively low, this introduces little friction for companies to try out a service. If they do so, and assuming they like it, this in turn kicks into gear the creation of network effects. Once the network effects take hold, the companies roll out the SaaS product to the wider organization, which in turn allows the per-user pricing model to “ride” the rollout and capture maximum revenue. In many cases, no discount for large volume is offered because of the “addiction” to the service that the network effect has created.

A good example of the above is JIRA. JIRA is project management and issue tracking software product provided by Atlassian, a $6 billion Australian SaaS products company. JIRA allows developers to document and review bugs and other issues, as well as to perform project management functions. The more people who use JIRA, the more valuable JIRA is to each user.

As a result of this, JIRA’s pricing reflects the network effects. As the number of users using the product increases, the per-user pricing actually increases. For a small team of ten people, per user pricing is $10 while at fifty people, per-user pricing increases to $50, which is a fivefold increase (Table 2).

While the exact effect of this pricing is impossible to know, a quick look at Atlassian’s annual report (page 44 and 49) shows that subscriber revenue increased by 1200% ($12.24 million in 2014 to $146.66 million in 2016) while its customer numbers increased 80% (37,250 in 2014 to 60,950 in 2016). A simple math shows that if pricing stayed same, Atlassian’s revenues would be only $22 million due to an 80% increase in number of customers. In other words, thanks to changes in its pricing, Atlassian generates an additional $126 million ($146 million – $22 million) per year!

The overall point of charging for network effects is summarized nicely in this academic paper: “The presence of network effect also favors pay-per-use over perpetual licensing; if the network effect is strong, pay-per-use will always dominate perpetual licensing regardless of the inconvenience cost or the potential piracy. With more heterogeneous consumers, higher potential piracy, lower inconvenience costs, and stronger network effects, pay-per-use licensing yields not only higher vendor profits but also a higher social surplus than perpetual licensing.”

Benefit No. 4: Freemium – Pricing as a Marketing Strategy

Freemium is a marketing strategy where a part or all of the product is offered with zero pricing for unlimited time. The “-mium” portion of the word “freemium” refers to “premium,” where the user of the free product could upgrade to a premium product, i.e., pay for the product. Freemium is offered in the hope that some of the users of the free product would upgrade to premium, but this offer is for unlimited time, so the users can upgrade whenever they please.

The benefits of freemium are summarized nicely in this Harvard Business Review article:

“Several factors contribute to the appeal of a freemium strategy. Because free features are a potent marketing tool, the model allows a new venture to scale up and attract a user base without expending resources on costly ad campaigns or a traditional sales force. The monthly subscription fees typically charged are proving to be a more sustainable source of revenue than the advertising model prevalent among online firms in the early 2000s. Social networks are powerful drivers: Many services offer incentives for referring friends (which is more appealing when the product is free). And freemium is more successful than 30-day free trials or other limited-term offers, because customers have become wary of cumbersome cancellation processes and find indefinite free access more compelling.”

The pertinence of this to the SaaS model is that freemium only works if the cost of delivering the service for free is marginal. If not, freemium is unlikely to work from a financial standpoint. Fortunately for SaaS, this is one of the inherent benefits of the delivery mechanism (Chart 5).

As such, SaaS lends itself very well to offering freemium pricing. This point is summarized nicely in this post,

“For the freemium model to work out, one specific product attribute must already be in place – low marginal distribution and production cost. Only if you can keep the cost as low as possible, an additional free user will cost you nothing more than a database entry […] this is an inherent characteristic of SaaS products”

A company that has a great freemium SaaS offering is Slack, which has over 4 million daily active users at the moment. Out of these users, 1.25 million are paid but the remaining 2.75 million are using the product for free. Pricing does not prevent them from using the product and makes them potential paid users in the months and years to come. In other words, many leads are acquired thanks to zero-dollar pricing.

Slack seems to be benefiting from its freemium structure, with a 30% claimed conversion rate. In other words, of the 2.75 million users who are using Slack for free, 800,000 are expected to become paid users in the future. That is lots of additional recurring revenue; to be somewhat precise, at $6.67/month/user with its cheapest paid plan, that is an additional of $5.3 million in revenue per month.

Not all products are created equally; thus, each product could have a different freemium offering and conversion rate. For example, while Spotify has a 27% freemium to premium conversion rate, Dropbox has a rate of 4%. Therefore, each company should figure out whether freemium is doable. Nevertheless, as Ben Chestnut, co-founder of MailChimp, puts it, Ben & Jerry’s free ice cream day showed that freemium could be quite powerful and has been for MailChimp since they started offering it, eight long years after they founded the company.

N.B. In this article, Freemium is defined as offering a product to a customer for unlimited time. A free trial is defined as offering a paid product to a customer for a given time. A free trial is not considered freemium, as it cannot be offered indefinitely. To further read the implementation of freemium and free trials, please check out this article from Lincoln Murphy.

On Cloud Nine

As more and more companies switch to cloud-based provision of services, the shift to SaaS from On-Premises software is likely to continue at a steady pace. And we think this is a good thing. Startups will find it easier to offer the same services that larger, more established companies have only been able to do thanks to large investments in in-house capabilities. Consumers will have greater flexibility to find products that balance their needs with their wallets. In other words, SaaS creates win-wins across the spectrum.

As SaaS technology and our understanding of its benefits develop, we expect to see more pricing models emerge to take advantage of this unique model. The examples above are not exhaustive, and are just a reflection of the types of pricing implementations used in the SaaS world, but I hope this article may serve as a guide for continuous pricing challenges and opportunities.

Article via Toptal.

Startups have many problems. Across all my channels I probably look at between 5 and twenty startups in the earliest stages of their lives each week. In many cases the entrepreneur has given me a brief that they want me to do X and they have given me some information about the company. In a very large proportion of the startups it is not clear what the problem is that they are solving, who the customer is or how they are going to make money.

Since I started working in startup back pre-2003 I have spoken to thousands of entrepreneurs and this appears to be a pervasive and nontrivial problem. There are many many entrepreneurs who do get it. In part this is because they have nous that the rest of us like, or because they have gone through on e of the many amazing training courses available – whether from Ycombinator, Udacity or a huge range of startup blogs like this one.

At the same time there are so many entrepreneurs who have never come across this idea and so they build a product to service an idea not a customer with a problem.

So being really clear, the first things that you need to be sure of as an entrepreneur are:

Is there a real problem?

If you can’t give me 10 real people’s names and what the problem is in their words I’m going to be a bit cynical. In a lot of cases if you can’t do that you haven’t got out the building and started trying to sell the idea. You haven’t seen your idea splinter as it comes into contact with reality. It’s better to do it now than it is later when you have spent thousands of hours and a shed load of money and are into serious panic mode.

Do the customers care?

Some problems are so small, so trivial or just not worth solving. You may have proved it but if the customer is going to pay you peanuts and there aren’t enough customers out there to justify buying enough peanuts to make a jar of peanut butter then why bother? In many case a lot of people cannot accurately describe their customers (partly because they have never met any)

How hard is it to solve?

It’s amazing building mousetraps. Seriously. There is nothing better than to sit down and make or build something awesome. I built a sandpit once for my kids. It was made out of IKEA slats from an old bed. It was glued together and had a waterproof inner line. It was a work of art. They played with the cheap plastic sandpit we got from Tesco. In that case I supplied a wonderful but over engineered solution that cost massively more – both in cash and time – than they wanted to pay.

So What to Do?

Get out the building! Talk to customers as early as possible and build what they want. It’s actually far less scary than sales because you aren’t actually selling anything to anyone. All you are doing is asking

“Do you have this problem?”

That can be put into conversation in so many ways. As you do it hundreds and thousands of times as you get into your startup you get a real sense of who the customer is. Then they tell you this is the problem I really have and this is how you can solve it for me.

Without even sounding a teeny bit salesly you can then ask

How much would you pay for that?

And they tell you. Bang! In just a few hours a potential customer has said if you build me this I will buy it from you. Then all you have to do is jot down their email or phone number and move onto the next person.

“Do you have this problem?”

And pretty soon you have a bunch of answers that you can categorise and derive the feature sets that customers really want. And you have a nice big email list of people who can can say to

You said you had this problem and that you would buy the solution if it cost $X. Well guess what we have got a solution for the problem and it costs $X. Press this button to go to the checkout.

This converts. It is far easier than building a business in any other way.

In a nutshell this is what I do for my customers. I help them to get from the idea to delivering the first products to customers

What’s a business model for? As a business strategist I use models day in and day out. They are the tools of my job in the same way that a carpenter uses an adze and saw or a farmer uses a plough and drill.

Let’s play with that analogy a bit. Can a farmer plough the field and sow the seeds without machinery? Sure he can. He can get rid of the tractor and replace it with a hoe and a basket for the seeds. In some places where the soil is thin or rocky that may actually be a more productive approach; not usually though. Can he do without any tools at all? Again it’s possible. Circumstances, where it will work, are more limited. You probably need to find a fertile river valley recently flooded, without too many birds and then scatter the seeds by hand. Primitive.

Trying to be a carpenter without tools….. I am sure there are some very talented people out there who can do amazing things with wood and their bare hands, but they are probably not carpenters.

So a business model is a strategist’s tool. Often people hear the word strategist and they throw their hands up like “Wooooh, How many degrees do you need to do to call yourself that?” Really though a strategist is anyone who thinks about how something is to be done. It’s about where the ship should be heading (to jump from land to sea) not making sure that the ship is moving. And so just like a sailor or a captain, the strategist needs tools to navigate successfully.

In some cases, tools are common sense. In many other cases tools have been derived from lots and lots of data. In almost all cases we aren’t taught ways of making great decisions at school or university. And that is another definition of what a business model is. It’s a way of helping you to simplify masses of information and use it to make better decisions

Pragmatism and Business Models

The pragmatic strategist, often working fast on a variety of small projects or startups, does not have the luxury of using large complicated frameworks. One of Leander’s projects for this year is to work out how to take the mass of startup models that he has in his toolbox and build them into a coherent system. The problem that he faces is the huge variety of problems that need to be solved quickly.

Someone once described being in a startup as being like a car driving fast on a potholed road. Sure you know that there are lots of potholes ahead – but all you care about is the pothole right in front of you now. If you don’t miss it, you’ll break your axle and never get that next few hundred metres down the road.

So unlike Shell with its famous 5 year plans startup models have to be usable out the box and quickly. If you can’t explain it, use it and understand the results in a few hours the time cost of the model is too large for most startups. If someone needs specialist training the time cost is too large. If the model is proprietary information or buried deep inside and inaccessible book the time cost is too high.

Any model that we use has to deliver results quickly and in a super digestible form that makes it clear what we need to do next.

One of the biggest benefits of the Business Model Canvas is that once you have done it you pretty much have your pitch structured and mentally prepared in your head.

These are my customers
This is their problem
This is how I will reach them
And interact with them
Using these resources
etc…..

Idealism is not the Enemy. Really

On the other hand, there is a case for idealism. We do get engrossed in the details and too often cannot see anything but details. I once worked for a COO who in the 8 months that I worked for her kept on saying ‘I’ve got to be more strategic’ and I think I had over 20,000 WhatsApp messages from her as she delved down into the details and we didn’t make the hard calls about which ways trade-offs should play out.

So flying up in the air and looking at the idealised company and how it should work and where we should go is important. Whether we use Porter or SWOT or Lean Canvas or anything else stepping away and taking an idealised approach on how to get to our vision from where we are now is very powerful. Of course, it will get trashed as we start getting blasted with WhatsApp messages again but the idealist has the deeper understanding that acts as a bulwark against the thousands of small decisions and tradeoffs that she has to make every day

Business Models Give Results

For a startup, it all comes down to results. Once a long time ago I worked for a very large multinational and spent months designing and polishing a business continuity plan. It was for an event that never happened, and I guess that if it has been maintained it would still work. Probably. The point is that we used a bunch of models but the net impact of all the work was a cost to the company. Sure it mitigated against a risk that never materialised.

In contrast, a startup needs results quickly. If we spend the time ‘playing’ with this business model what is the opportunity cost? There are always more tasks to do than seconds in the day. Will this model or tool mean that tomorrows tasks deliver more value? Next months’ tasks deliver more value when I complete them. In many ways, this is the inverse of the accountants’ net present value. Instead of worrying about the future value of money we are worrying about the future value of our time.

A business model by this understanding offers a map and a shortcut. It is a heuristic, it can be used wrong, it can be executed wrongly, but by and large, it saves us time and makes us more money tomorrow than we lost today.

Understanding Comes Faster with A Business Model

It does that through understanding. Almost everything we do is a level of abstraction. Our brains abstract the information that comes through our minds into comprehensible constructs. When we use maps we strip away vast amounts of information in order to produce something that is quickly and easily understandable. The way that we use business models is to structure our understanding of the business problem so that we have a level of simplicity that gives clarity.

Some may say oversimplification. I think not. Oversimplification does exist but one of the consequences of that is to trigger debate and greater group understanding if it is not agreeable to all. It’s probably less of a danger than under simplification but there is a huge tendency among people for their minds to skate over the detail. So we reduce complexity until the model is at the point where we can understand all the components of the model without our minds skating or skipping a bit.

It is surely no accident that most of the best models have 4,5 or 6 components. Others may have a few more but once you get beyond a dozen models start to creak under their own weight and collapse.

Business Models Give the Benefit of Systematisation

So what is a business model for. It’s like a stepping stone. Each gives you enough intellectual stability to stand safely, whether the water is rough or smooth and step confidently forward to the next in pursuit of your vision.

Strategy models bring some system, some order, to moving forward in a chaotic environment where you have too much or too little information.

Worth the price

One of the truisms of business is that it is easier to keep a customer than to get a new one. Actually it may be it’s cheaper to keep a customer. In a lot of modern marketing, especially for startups there are so many channels available that it is often easier to just keep pumping money into them and get an endless stream of new customers. So long as the costs aren’t too high and you are getting growth…. what’s the problem with that? This article is about sing cohort analysis to help with this, but first some reasons why churning customers is a bad idea.

Well there are a number of problems.

Wasting money
Lower marketing ROI
Poorer customer satisfaction
Limited customer pool

Shallow Pools of Customers – However Large the Market

I’m not going to spend any time on the first three. The last one is interesting. When you start off and define your niche it’s fairly easy to define who your ideal customer is. They are X, with Y job and have these values and these hobbies. You may do some demographic research and estimate that there are 100,000 of them in your city. When you are on customer zero, 100,000 customers is a pretty huge pool for you to swim in and catch fish.

Move forward six months or a year and if you are even moderately successful you are going to have reached a lot of those potential customers. There will be awareness even if they haven’t tried your product. If your churn is high – 50, 60, 70% or even more then it actually becomes fairly easy to start exhausting that pool of customers. What do you do then? You have to start moving into new niches and extending your product line. That carries a bunch of it’s own risks. Will people like your product or service as much? Will it be as good a market? How many more extensions can you make before you have the same problem again and again?

You see if your retention rates are poor you are a bis like a slash and burn farmer. You are going through a beautiful verdant jungle and chopping it all down for short term profit. What you leave behind is a wasteland of customers who will never return

Cohort Analysis & Retention Rates

What cohort analysis does is to look at which customers come back and how often. The first time I used it I was working on a mobile app. I’d just come onto the job and had had some amazing passionate stuff told to me by the entrepreneur. I like data and so I jumped onto App Annie to get a sense of how good the app was at retaining customers. Despite multiple product improvements 95% of customers never opened the app twice. Close to 100% never opened it after the first day. Oooh! That made my role kind of pointless as there was little point pouring good customers into a bad sewage filled bucket with more holes than plastic in it.

Restaurants and Cohort Analysis

I was recently working with a restaurant and crazily they actually had some useful data that we could do cohort analysis on to look at their retention rates. Why do I say crazily? Well I asked a few restaurant owners and managers

Even the largest chains struggle with this for a number of reasons, starting with the lack of sophistication of the talent they employ.

and

First, they don’t know what and how and why part of the question.

and

I know of no restaurant owners in New Orleans that use anything close to cohort analysis to assist in converting new customers to regulars. I can tell you with certainty we do not do it.

even

Could the industry improve through such analytical methods? I think not.

and finally

most don’t have a way to gather the data and analyze it.

Gulp. Stuff that is standard practice in the tech industry because it helps to deliver consistent scalable results is seen as too difficult and lacking in value. This is in an industry that has an exceptionally high rate of business failure and notoriously low wages and margins.

In this case we used data from a delivery company similar to Deliveroo or FoodPanda that allowed us to see when and how frequently customers ordered. The original data came as a simple excel list. Person X ordered Y at Z date/time. For a more traditional restaurant wanting to analyse walk in customers one approach would be to access the CCTV feed and use facial recognition software to give each customer a unique anonymised identity. So everyone who comes through the door (and says for long than 20 minutes) is given a unique ID. There is no need to know anything else about them – sex, age, name etc – and this makes it easier in many countries to comply with data protection laws.

Creating Cohorts

The first step was to turn this into cohorts. A cohort is a group of people who you can treat similarly for the purposes of analysis. In this case we decided that given the nature of the restaurant and it’s menu we’d treat everyone coming into the business in the same period as a chort. So we split them into months. First of all you convert your order/time data into a date reference. So if someone ordered at 15/06/2013 at 17:30 then we tag them as ordering in June 2013. Then using excel we work out when was the first month that they came in and ordered – their start cohort.

In this case if this was the first time they had ordered we would lump them with all the other people who ordered for the first time in June 2013. From then on the individal disappears into the data. All you are interested in are the June 2013 cohort and how they differ from the July, August 2013 cohorts.

How do you see how they differ? Well the easiest way is to see how many times they come in and visit. You go back to excel and count how many times they visited in June, July, August etc. The you do the same for all the other cohorts.

Simple Tip: This takes ages if you do it by hand. So it’s normally simplest to use pivot table. This have Cohort as the column, Age as the row and count of visits/orders as the value.

Then you can start doing a bit of comparison.

Example Cohort Analysis

This cohort analysis is some random data that I pulled off the internet (so I don’t accidentally breach any NDAs)

If we assume that the column is the cohort, the rows are the number of months since first order and the colours are the retention rates with green being good and red being bad we can see some pretty interesting things with this data.

It is a triangle shape because the first customers have been around for 20 months, the next cohort for 19 months, the next for 18 etc etc

For the first 2 or three months people like the service. Each month you lose a few more people and over time this adds up until no one from that cohort returns any longer. If we look at the top we can see that big red blotch. That’s some sort of service failure, if we are talking about a restaurant. Why did people suddenly stop coming after the fifth month? Was there a change in a customer loyalty programme?

It certainly seems related to how long people had been a customer rather than an event in the restaurant. In one case a customer had had a major flood and as that impacted all customers at once it appeared as a diagonal band of bad retention across the whole chart. In this case because it always hits after a certain period it is far more likley to be the way that customers are being treated.

And this is the strength of cohort analysis. By comparing the behaviour of similar customers, or prospects or leads over time you have the ability to get feedback on what you are doing right or wrong. Designing a new menu may result in increased sales – but if it is too narrow and does not have enough variety to keep customers coming back then this will be obvious in the cohort analysis far earlier than it will be in the sales.

This is just starting to scratch the service as all we are looking at is frequency of customer visits at this stage. If we start mashing it with sales or marketing data or using cohorts segmented by persona we can then start asking interesting questions about the impact of different changes on customer spend, the impact on long term customers and how easy it is to convert someone to a long term customer

Conclusion

As a final point. As a business process consultant data is absolutely critical to helping me make informed analysis. Without it I am waving my hands in the air and who knows how valuable that is. I spoke to one CEO about this and he said ‘We can’t do it. We’re not sophisticated enough’. He gave me access to the POS data and within just a few hours we had a fully functioning cohort analysis for his business that he was then able to use to make powerful decisions that had a significant impact on his success

Food Delivery Rider Modelling with Monte Carlo Analysis — Image via spoonly.in

In this post I will talk about Food Delivery Rider Modelling – working out how to ensure that you have enough riders to meet demand whilst not having over supply

By some accounts the food delivery company FoodPanda has more than 20% of its sales in a two hour time window on a Sunday night. This is great for its customers who are looking for something to make themselves feel good before going back to work on Monday. It is also great for the restaurants as it allows them to turn a slow Sunday evening into something almost as good as Friday and Saturday. This leaves them with only four slow days a week and thus has a good impact on sales and thus profitability.

However it is pretty terrible for FoodPanda. 20% of the orders in 1% of the time. If we break it down by actual working hours it works out at about 20% of the orders in about 3% of operation hours. Practically that means if you average out demand over the other working hours they need six times as many food delivery riders on Sunday night.

Variability in Food Delivery

One of the things about food delivery, like many jobs, is that most people want reasonably consistent work on a regular basis so that they have reliable cash flow. In fact most businesses are the same. Pension funds are one extreme. They invest in the most stable risk free cash flow they can find – government bonds – so that they can reliably deliver on their pension liabilities.

Factories with highly variable demand are highly inefficient. Utilisation rates drop right down as does break-even time on machinery and generally the factory climbs the experience curve far more slowly as experience is gained in fits and starts rather than at reliable and consistent pace.

So a company like FoodPanda with highly variable demand is destined to be inefficient. Systems have to be built to scale for that massive peak in Sunday demand and for the rest of the week they are woefully underutilised. Any business that has demand peaks has a similar problem. one of the best examples of this is Amazon which has the twin peaks of Thanksgiving and Christmas in Q4 each year. The infrastructure that it requires to meet this demand is massively underutilised during the rest of the year.

Managing Variability

It adopted two solutions firstly it started providing cloud services, AWS, to utilise all the spare supply during the rest of the year. Secondly it sought to moderate demand by such promotions as Amazon prime. Whilst Prime doesn’t shift the needle much in the peak periods, Prime contributes more than 20% of sales overall and that has a significant impact on infrastructure utilisation rates during the rest of the year.

Many food delivery startups find that they have a midweek peak with many more people ordering on Tuesdays and Wednesdays than on Mondays and Fridays. This has a similar impact. This can be as much as 30% of production. The impact of this is that as the delivery part of food delivery is non-scalable the wage costs on the weaker days are far higher than would normally be ideal.

Leading, Lagging and Matching Strategies

Essentially this is because you cannot react instantly to demand. It takes time to recruit and train new riders for a surge in demand. Riders who are underutilised churn very rapidly. So you either have to adopt a leading or a lagging strategy. A leading strategy means having more riders and thus ensuring that you can always meet customer demand. On the other hand it also means that you have excess supply which can really hammer the break-even point.

A lagging strategy means reducing the cost base and never having excess riders, but takes the hit in customer satisfaction when you don’t have the correct number of riders to deliver on time, or indeed at all. A matching strategy – which is basically the principle of load balancing is really only possible if there is a significant excess supply of food delivery riders and as a result people will turn up for work and you can choose enough people for work and give them nothing is you don’t have enough. This passes the risk onto the rider and obviously is not popular.

Food Delivery Rider Modelling

What can you do? Depending on the nature of your food delivery business you can look at your historical records of demand. You can use actual orders or some other proxy that correlates to order, for example web traffic. You can then use this for your period of analysis. With very high demand variability during the week this is often the day, the shift or even the hour. (hourly analysis needs a much larger data set than weekly analysis). Once you know the average for the period of analysis and the standard deviation you can estimate the number of riders that your require.

The problem is that this masks many peaks and troughs in demand. As a result you will find yourself quickly facing over capacity some days and under capacity on others. And indeed you will generally find that there is a tradeoff that you have to make depending on your appetite for risk, the importance of customer satisfaction on growth and the amount of cash you have available for growth.

Monte Carlo Analysis

Taking this a step further you can do Monte Carlo analysis, which is a form of scenario modelling, and this allows you to see what is likley to happen in the future, if the same demand patterns are maintained (it gets a bit more complicated if you have, say, 15% growth a month). You are then able to get fairly precise estimates such that you can estimate the mean time between failure (MTBF) of how frequently you will not have enough riders and the min max for that and thus the cost of foregone orders and customer dissatisfaction. Equally you are able to see the likely wastage cost each day and build that into your financial models.

The way that I normally do this is to:

Gather the historical data sets in excel and clean them
Identify the period of measure – this normally means adding a new column and using a formula to create the values. For example =weekday(date) in Excel will give you a number for each day of the week. Sunday =1, Monday =2 etc that rapidly allows you to segment by day of the week
The use a filter, pivot table or histogram to gather the relevant order data for the period
Once you have this you can generate the mean and the standard deviation
Then you can use =rand() with norm.inv to generate random values based on your distribution of orders (assuming that the set is normally distributed)
Then I tend to use a data table to run a 1000 events for each production value that I am testing
You can then clearly see the breakeven points for under capacity and over capacity and the transition zone between them where you will have to make your tradeoff

If this sort of analysis sounds useful to you fill in the form below and we can have a chat

← Back

Thank you for your response. ✨

This is a quick post on why code and the ability to programme is not important to you – initially – when you are looking at building your MVP. No code in an MVP!

I’m working with a couple of startups at the moment and we ended up having the same conversation in both at the moment. The first is a sports based startup in Los Angeles. I was talking with the CEO when the message came through on slack from the CTO

Backend for the MVP will take about 3 1/2 months
No Programmers does not mean no MVP

There was a long pause and a mutual feeling of – “we really need to start rethinking a lot of things right now”

That was the first.

The second was with Leander when we we were having our weekly meeting working on Lean Card. The actual detail is for a later conversation but basically we were working through ideas for the product road map and what the MVP should look like. Then Leander said

If our unique value proposition is that we offer “awesomely easy business cards” then what we are offering can be described as project management. I don’t need a computer to test that.

So in both cases we ended up sitting down and thinking how the business team could sit down and test all the key hypotheses without a single line of code.

This drive came from a few reasons

We can do it now
If it doesn’t work we can change easily
We don’t waste Dev time or money

MVP for Sales

For the sports startup the key testable hypothesis was whether suppliers would make a commitment which would validate part of the model. The LA sales team already has verbal commitment but we felt it was really important to actually get suppliers signed up to provide traction and real validation.

The idea is then to build some ultra simple sales sites on Facebook – provisionally with Wufoo and stripe and test several different customer segments. That will be done in the next few days and will start giving us some real validation – and just as importantly real life experience of which segments are best and how we need to design the app.

MVP for Operational Processes

For Lean Card the idea was exactly the same. Instead of using a cobbled together tech solution the MVP here was to sample do the process of project managing business card design with family, then friends, then acquaintances, then strangers, so that we could experience what worked and what didn’t. We have a strong idea of what the roadmap should be, but the getting out of the building, totally non scalable process of doing everything manually ourselves is going to make it very clear what is really needed in the first electronic versions and what is not – and maybe even that we need to pivot.

In both cases the net result is to derisk the development process as much as possible. A few months ago I was working with a startup who was looking to persuade people to wear their iphones to stream TV style adverts on public transport. We stuck some iphones to people with sticky tape and started playing Coca Cola dvert using YouTube. Then we send the ‘volunteer’ walking through a busy shopping centre to see what happened. The rest of the team tried to count how many people looked at the screen and how long they looked for. Total failure and an early pivot.

The learning point for this is to forget about the code and brainstorm simple and easy ways that you can start looking for validation of key hypotheses, start testing workflows that are critical to operations and see how customers react.

As someone said to me yesterday

It’s scary going and talking to strangers

Very true, but get out there and do it!