Chapter 8: The Challenge of Pivoting or Persevering

At any given point in time, every entrepreneur faces the common challenge of when to pivot and when to persevere. There is a time when we entrepreneurs would question if we are making enough progress to believe that our initial strategic hypothesis is correct or if we need to make a significant change. That change is called a pivot. Unfortunately, due to the scientific methodology that guides the Lean Startup, there is always a misunderstanding that it offers a strict clinical formula for making pivot or persevere decisions.

8.0 Innovation accounting leads to faster pivots

To understand how innovation accounting leads to faster pivots, we will consider the example of David Binetti, CEO of Votizen. In the early 1990s, he helped build USA.gov, which is the first portal for the federal government.

David wanted to solve the problem of civic participation in political procedures. His initial product concept was a social network of verified voters which was a place where people who were passionate about civic causes could communicate their ideas and recruit their friends. David built his first MVP for more than 1200 USD in 3 months and launched it. He was building something everyone wanted, and from its early days, Votizen was attracting early adopters who loved the main concept. David had to refine his product and business model like all entrepreneurs. His initial concept included 4 giant leaps of faith:

1. Customers would be interested in the social network to sign up (registration).
2. Votizen would be able to verify those customers as registered voters (activation).
3. The registered voters would eventually engage with the site’s activism tools (retention).
4. Engaged customers would tell friends about the network and recruit them for civic causes (referral).

In the initial cohorts, 5% signed up for the service, and 17% were verified as registered voters. The numbers were low, and there wasn’t sufficient data to tell what engagement or referral would occur. It was thus time to iterate. David spent 5000 USD in split testing new product features, messaging and improving the product’s design to make it convenient for customers’ use. Those tests showed improvements from a 5% registration rate to a 17% registration rate and from a 17% activation rate to more than 90%. This is the power of split testing. This optimization gave David a significant number of customers to measure the next two leaps of faith. However, those numbers were disappointing as he achieved a referral rate of only 4% and a retention rate of 5%.

David knew he had to do more development and testing. For months he continued to optimize, split test and refine his pitch. He spoke to customers, had focus groups and did many A/B experiments. Chapter 7 showed that different product versions are offered to various customers in a split test. Observing behavioural changes between the 2 groups allows one to guess the impact of the different variations. The referral rate went up to 6%, and the retention rate went up to 8%. David had spent 8 months and 20,000 USD on building a product that wasn’t consistent with the expected growth model. David was facing the difficulty of deciding whether to pivot or persevere. This is one of the most complicated challenges that entrepreneurs face. The aim of creating learning milestones isn’t to make easy decisions but to ensure that relevant data is available when it’s time to decide.

David had spoken to many customers and had plenty of learning that he could use to justify the failure he was experiencing with the current product. This is precisely what most entrepreneurs do. David had 2 benefits of helping him get out of his mess:

1. Although he was committed to a vital vision, he did his best to launch the product early and iterate. Therefore, he was in a pivot or persevered situation in just 8 months. The more money, time and creative energy that had been drilled into an idea, it’s harder to pivot, but David had done well to escape that problem.

2. David had discovered his leap of faith questions and had made quantitative predictions about each question. It wouldn’t have been hard for him to declare success retrospectively from the first venture. Ultimately, some of his metrics, like activation, were doing well. Regarding gross metrics like total usage, the company had positive growth. It’s only because David had focused on actionable metrics for each of his leap of faith questions that he was able to acknowledge that his company was failing. Moreover, since David had not invested in PR, he could escape public embarrassment or distraction.

Failure is a precondition for learning. David believed that although his optimization was improving the metrics, they were not building towards a model that would fully sustain the business. However, he didn’t give up too soon like all promising entrepreneurs. He decided to pivot and test a new hypothesis. A pivot requires keeping one foot fixed in what you’ve learned so far while making a significant change in strategy to find greater validated learning. In this situation, David’s contact with customers proved to be helpful. He received 3 frequent feedback, and based on that feedback, he wanted to refocus the product on what previously had been considered one feature of a larger whole. Based on the customer feedback, they liked the concept like the voter registration technology, but they weren’t getting any value out of the social networking side of the product.

He decided to change Votizen into @2gov, a social lobbying platform. Instead of integrating customers into a civic social network, @2gov allows them to contact their elected representatives immediately and easily through Twitter. The customers communicate digitally, but @2gov translates digital communication into paper. As a result, Members of Congress received printed letters and petitions. David’s leap of faith question in @2gov was that passionate activists were ready to pay to have @2gov facilitate contacts on behalf of voters who cared for their issues. David’s new MVP took 4 months and 30,000 USD. He’d spent a total of 50,000 USD and worked for a year, but the results from his subsequent testing were fair – registration rate was 42%, activation rate was 83%, the retention rate was 21%, and referral was 54%. However, the number of activists ready to pay was less than 1%.

Before considering David’s next pivot, we need to consider how he was able to show validated learning. With this new product, he hoped to improve his leap of faith metrics fairly. He did this not by working harder but by working smarter by taking his product development resources and applying them to a new and different product. Compared to the previous months of optimization, the further 4 months of pivoting had led to a fairly higher return on investment. However, although his metrics and product improved, it wasn’t improving fast enough.

David pivoted again. This time instead of relying on activists, he went to large organizations, professional fundraisers and big companies with a professional or business interest in political campaigns. The companies were extremely eager to use and pay for David’s service, and he quickly signed the letters of intent to build the functionality they required. In this pivot, David did a ‘customer segment pivot’ by keeping the product’s functionality the same but changing the audience’s focus. He focused on who pays by going from a B2C company to a B2B company. He changed his growth model to one where he would be able to fund growth out of the profits generated through each B2B sale.

3 months later, David had built the functionality he promised. However, there were problems when he went back to collect his checks from the companies. Each company delayed and declined the opportunity. Although they had been excited to sign the letter of intent, closing an actual sale was complicated. It turned out that those companies weren’t early adopters.

Based on the letters of intent, David had increased his headcount by adopting extra sales staff and engineers, expecting to service higher margin B2B accounts. When the sales didn’t happen, the whole team worked harder to find revenue somewhere. But regardless of the amount of sales calls and optimization they made to the product, the model wasn’t working. Coming back to his leap of faith questions, David concluded that the results disapproved of his B2B hypothesis. He decided to pivot again and reduce his staff by trying the ‘platform pivot’. Rather than selling an application to a single customer at a time, David pictured a new growth model inspired by Google’s AdWords Platform. He made a self-serve sales platform where anyone could become a customer by paying through a credit card. No matter what causes customers were passionate about, they could visit @2gov’s website, and @2gov would help them find new people to get involved with.

The new product took a month to build and had immediate results where 51% was the sign-up rate, 92% was the activation rate, 28% was the retention rate, and 64% was the referral rate. Most importantly, 11% of these customers were ready to pay 20 cents a message. This was the start of an actual growth model that could succeed. Although receiving 20 cents a message didn’t sound a lot, the high referral rate meant that @2gov could grow its traffic without spending much marketing money. One of the most important lessons to note in Votizen’s story is the acceleration of MVPs. The first MVP took 8 months, the next one took 4 months, 3 and then 1. Each time he could validate or refute his following hypothesis quicker than before.

How can one understand this acceleration? Most parts of the product had to be thrown away between pivots. Worse scenario – the remaining product was a legacy product that was no longer suitable to achieve the company’s goals. The effort needed to change the legacy product took more work. Preventing these forces were the challenging lessons David had learned in each milestone. Thus Votizen accelerated its MVP process as it was learning important things about its customers, market and strategy.

Today, Votizen is a successful company whose system can currently process voter identity at once for 47 states representing 94% of the US population and has delivered millions of messages to Congress.

8.1 A startup’s runway is the number of pivots it can yet make

Experienced entrepreneurs always speak about the runway their startup has left behind – the time left in which a startup should achieve lift-off or fail. This is commonly defined as the remaining cash in bank divided by the monthly burn rate or net drain on that bank balance. For example, a startup with a million dollars in the bank spending 100,000 dollars per month has an estimated runway of 10 months.

When startups start running low on cash, they can extend the runway in two ways:

• By cutting costs.
• By raising additional funds.

However, when entrepreneurs start to cut costs randomly, they are bound to cut costs that allow the company to get through its Build-Measure-Learn feedback loop as they are to cut waste. If the cuts lead to the feedback loop to slow down, all they have achieved is to help the startup go bankrupt.

The accurate measure of a runway is how many pivots a startup has left – the number of opportunities it has to make an essential change to its business strategy. The startup will have to find ways to achieve the same number of validated learning at a lower cost or in lesser time.

8.2 Pivots need courage

Most entrepreneurs wish they had pivoted sooner for three reasons:

1. First, vanity metrics make entrepreneurs form wrong conclusions and live in their own world. This damages the decision to pivot as it robs the team of believing it’s essential to change.
2. When an entrepreneur has a vague hypothesis, it’s nearly impossible to experience complete failure. But, without failure, there is no momentum to embark on the radical change that a pivot needs.
3. Many entrepreneurs are scared. Most entrepreneurs are not scared that their vision is wrong but that their vision might be deemed wrong without being given a real chance to prove itself. This fear drives most of the resistance to the MVP, split testing and other techniques to test the hypothesis. Interestingly, this fear drives up the risk as testing doesn’t happen until the vision is entirely represented. However, it’s always too late to pivot by that time as funding is finishing. To avoid this issue, entrepreneurs must handle their fears and be ready to fail publicly. Unfortunately, entrepreneurs who are famous or part of a renowned brand face a terrible version of this problem.

8.3 The decision of pivoting or persevering

You need sound judgment and a proper mindset to decide if you want to pivot. The decision to pivot is an emotional situation for a startup to be in, and it has to be addressed in a structured way. One way to reduce this challenge is to schedule the ‘pivot or persevere’ meeting in advance. Eric Ries believes that every startup needs a regular pivot or persevere meeting. Every pivot or persevere meeting requires the presence of the product development and business leadership teams. The product development team should bring a complete report of the results of its product optimization efforts over time and a comparison report of how the results pile against expectations. The business leadership team should bring detailed accounts of their conversations with present and prospective customers. In this manner, startups can make a sound decision on whether to pivot or persevere.

8.4 Failing to pivot

Deciding to pivot is very challenging that most companies fail to make it. A few years after IMVU was founded, the company had tremendous success. The business had reached 1 million dollars monthly revenue, creating 20 million avatars for their customers. However, without their knowledge, they had ignored the principles behind the success of their early efforts. Therefore, they missed the chance to pivot even when the signs were in front of them. They had created MVPs to test new ideas and run experiments to tune the engine of growth. However, before they began to enjoy success, many people had warned them about their low-quality MVP and experimental approach, urging them to slow down.

They wanted IMVU to do things correctly and focus on quality instead of speed. After their approach was vindicated, the advice they received changed. People advised them to continue with the process, but it was wrong. It’s important to remember that the reason behind building low-quality MVPs is that developing any features more than what early adopters need is a waste. Once you’re successful with early adopters, you want to sell your product to mainstream customers. Mainstream customers have various requirements and are much more challenging than early adopters. Therefore, the type of pivot IMVU needed was a customer segment pivot. In this pivot, the company understands that the product it’s building solves a real problem for real customers but that they are not customers it originally planned to target. This means that the product hypothesis is partially confirmed.

A customer segment is a complicated pivot to execute because the actions that made IMVU successful with early adopters were entirely against the actions they had to adopt to be successful with mainstream customers. They lacked a clear understanding of how their engine of growth operated. They had trusted their vanity metrics and stopped using their learning milestones to hold themselves responsible. Instead, they focused on exciting stuff such as breaking new records in signing up paying customers and active users, monitoring their customer retention rate, etc. While all this was happening, they missed a lot of clear signs that it was time for them to pivot. For example, they had spent months trying to improve their product’s activation rate (the rate where new customers become active customers of the product), which remained low. They did many experiments such as usability improvements, incentive programs, customer quests and other game-related features. These new features and marketing tools were independently successful. They measured these experiments carefully using A/B experimentation. However, for many months they saw careless changes in the whole drivers of their engine of growth.

They had ignored the signs as the company was still growing and delivering good results. But they were quickly exhausting their early adopter market. It was getting harder for them to find customers they could gain at the prices they were used to paying. As they were pushing their marketing team to find more customers, they were forced to approach more mainstream customers, but mainstream customers were disappointed with early products. New customers’ activation and monetization rates started to reduce, driving up the cost of searching for new customers. Suddenly, their growth was dying, and their engine of growth stopped. It took them too long to make the necessary changes to fix the situation. As with all pivots, they had to go back to square one and start the innovation accounting process again. They had to reintroduce themselves to their new mainstream customers. Their interaction designers guided them by developing a clear customer archetype that depended on extensive in-person convos and observation. Next, they had to invest heavily in a significant product investigation designed to make the product convenient to use. Due to their excess attention on fine-tuning, they had stopped making more substantial investments like these and preferred investing in lower risk and lower yield testing experiments.

However, investing in quality, design, and larger projects didn’t mean they should abandon their experimental roots. Once IMVU realized their mistake and made the pivot, those skills came in handy. As they built the product, they continued to test their new design toe to toe against the old one. At first, the new design performed worse than the old one as it lacked the old design’s features and functionality and had many new mistakes. But the team kept improving the design until it started performing better after a few months. The new design thus laid the foundation for their future growth. The foundation had generously paid off. By 2009, their revenue had more than doubled to more than 25 million dollars annually. However, they would have enjoyed this success earlier if they had pivoted sooner.

8.5 Types of pivots

A pivot is a particular type of change made to test a new fundamental hypothesis about the product, business model and engine of growth. Following are the types of pivots:

1. Zoom-in pivot

In this scenario, a previous single feature in a product becomes a whole product. For example, this is the type of pivot Votizen adopted when it pivoted from an entire social network to a simple voter contact product.

2. Zoom-out pivot

In this scenario, a single feature cannot support a complete product. In this pivot, what was a whole product becomes a single feature of a much larger product.

3. Customer segment pivot

The company understands that the product it’s building solves a real problem for real customers but that they are not the customers it was actually planning to target. This means that the product hypothesis is partially correct, solving the right problem but for a different customer segment than initially expected.

4. Customer need pivot

By getting to know customers very well, it sometimes becomes clear that the problem we’re trying to solve for them is unimportant. However, due to this customer intimacy, we mainly discover other relevant problems to solve for these customers. In most cases, these relevant problems may need little more than repositioning the current product. But, sometimes, it might need a completely new product. So, again, this is a situation where the product hypothesis is partially correct, that is, the target customer has a problem that is worth solving but not the one that was initially expected.

5. Platform pivot

This pivot implies a change from an application to a platform or vice versa. Usually, startups that dream of creating a new platform begin to sell a single application, which is a killer app for their platform. Eventually, this platform emerges as a vehicle for third parties to leverage to make their related products. However, this is a complicated pivot to get right, and some companies must execute this pivot many times.

6. Business architecture pivot

This pivot takes a concept from Geoffrey A. Moore, who noticed that companies usually follow 1 out of 2 major business architectures – high margin, low volume (complex systems model) or low margin, high volume (volume operations model). The complex systems model is usually connected to B2B, or enterprise sales cycles and the volume operations model is generally connected to consumer products.

7. Value capture pivot

There are many ways to get the value a company creates. These methods are usually referred to as monetization or revenue models. Monetization means it’s a different feature of a product that can be added or removed if necessary. In reality, capturing value is an integral part of the product hypothesis. Changes in how a company captures value can have life-changing consequences for the rest of the business, product and marketing strategies.

8. The engine of growth pivot

There are 3 essential engines of growth that power startups – the viral, sticky and paid growth models. In this pivot, a company changes its growth strategy to find faster or more profitable growth. Sometimes, the engine of growth also needs a change in how value is captured.

9. Channel pivot

A sales or distribution channel determines the product’s price, features and competitive landscape. A channel pivot recognizes that the exact requirements can be made through a different channel with better efficiency. A channel pivot occurs whenever a company lets go of a previously complicated sales process to “sell direct” to its customers.

10. Technology pivot

Technology pivots are common in established businesses. Established companies are successful in this pivot as nothing much is changing. The customer segment remains the same, the value capture model is the same, and the channel partners don’t change. However, the only doubt is if the new technology can provide a better price and or performance than the existing technology.

8.6 A pivot is a strategic hypothesis

Pivoting is not a substitute for good strategic thinking. The issue with providing famous pivot examples is that most people are familiar with the successful end strategies of well-known companies. However, they don’t know the pivots adopted to discover the successful end strategies that made them successful businesses today. Companies have a solid reason to connect their PR stories around the founder and make it look like their success was undoubtedly the result of a good idea. Therefore, although startups mostly pivot into a strategy that seems similar to that of a successful company, it’s important not to put too much stock in these similarities. It’s tough to know if the similarities have been made correctly. Have we made the same essential features or just superficial features? Will, what worked in that industry work in ours? Will, what worked previously work now? A pivot is better understood as a new strategic hypothesis needing a new MVP to test.

Pivots are a permanent certainty for any growing startup. Therefore, even after a company achieves early success, it should continue to pivot. Those who are familiar with the technology life cycle concepts of Geoffrey A. Moore know certain next phase pivots such as the chasm, the tornado and the bowling alley. Managers should match these pivots to their current situation to consider the right advice at the right time.

Conclusion

A pivot isn’t a pressure to change. You should never forget that it’s a type of structured change made to test a new fundamental hypothesis of the product, business model and engine of growth. It’s integral for the Lean Startup method. It makes companies that follow the Lean Startup model strong despite their mistakes. That is, if you make a mistake, you have the tools you need to understand it and the cleverness to discover another solution.