Yesterday at an AI conference, I saw an unusual number of companies offering Insights-as-a-Service. One after the other, brilliant entrepreneurs got on stage to pitch a very similar story: ‘we can help your organization uncover business insights hidden in your big data, using our unique machine / deep learning platform.’
Since almost every company today is a big-data company, it comes as no surprise that so many startups are drawn to offer Insights-as-a-Service. Which raises the million dollar question: what does it take to succeed in an Insights-as-a-Service model? What do these startups need to do to prevail?
If you’re one of those entrepreneurs you probably think it’s the data. You quote the economist’s famous saying that data is the new oil. You augment that with the ‘data race’ assuming that the company that will reach the critical data-mass first will dominate the market using the data network-effect.
That may be true, but also extremely challenging to fund. Especially in light of the huge amount of data needed to train AI models today. So if you want to win the data-network effect, you’d need to find a solid business model that will get you there. A data generating business model that can fund your growth, or at least support it.
Here are 8 business models you may want to consider. Feel free to tweak or mix n’ match to fit for your specific company:
Professional services
Perhaps the most common model for market penetration. You offer your platform for a subscription fee, but you also offer professional services to help first time users. These are provided by consultants, advisors, vertical experts or AI specialists. They can be part of your team or via partnerships. Your customers will write them off as NRE’s, so they won’t count against your subscription fees. While this model is great to generate some success stories and references customers, it is not a scalable model. You’ll need to evolve this model into one of the following models in order to grow to customers who do not need such scaffolding.
Educational Expert
The Professional Services model works well when augmented with the Educational Expert model: conferences, course, publications, etc. These help establish credibility and position your company as a market expert. It also helps attract customers to other subscription models. When an industry expert recommends a certain tool or when customers know this expert is using your tools to come up with great insights, they’ll want to have access to the same tool. You can partner for these educational services, and offer free or discounted version to such KOL experts. Unlike Professional Services model, Educational Expert model has a much longer effect and as such is a good seed to plant going forward.
Freemium
Free has a strong appeal. But it’s only good if you can convert free users to your paid packages. The most common levers for conversion are: volume (time, content, interactions…), duration, features, service level and customization. In B2B, free users are often companies that trial your service before making a purchase decision, or small companies (or teams) that get enough value off your free service. Don’t discard the latter as taking advantage of your Freemium model. Consider them as your community. They can help generate data which feeds into your paid products. And they are often your biggest supporters giving your marketing efforts an authentic voice.
Tiered Pricing
Similar to Freemium, tiered pricing can also work in the same dimensions of volume, duration, features, service level and customization. But it’s uniqueness stems from your ability to design tiers that match the right price for the right value of each specific segment. Take Salesforce Premiere support as an example of fast turnaround on support tickets. This tiered price plan is targeting organizations where real time is critical. Others offer tiers that allow (and even encourage) customers to grow their usage and upgrade over time. Whatever you do, make sure to keep the gap between price and value to a minimum to avoid churn and shaming.
Exclusive Membership
Just like some B2C models offer an elite product or service to a small, yet exclusive group, so do some businesses. Often these private clubs set a high price as ‘self selection’ threshold. Just imagine what might be the price that professional investment funds would pay to get financial trading insights 5 minutes ahead of everyone else, especially if they knew that only a handful of players can get this service. This model can fit well with Tiered Pricing. It can also be used as an initial model during market penetration, first establish a high-end positioning before creating less expensive versions of the service.
Analytics
Leverage your customers’ inherent data generation ‘engine’ to build long term engagements. One of the key risks of Insights-as-a-Service is that customers may not need insights on a continuous basis. Some industries still evolve slowly (e.g. insurance). And some players may take a while to implement your last insights before they can consume new ones. By augmenting your insights with an analytics services (home grown or partnered) you not only help your customers implement and measure the impact of your insights, but you also keep a stronghold to know when its time for up-selling another insight. Provided you follow your own advice and apply AI over your analytics data to trigger your own actions…
Content Access
As you grow, try to build a body of knowledge, an accumulation of insights in a specific vertical. Offer access to paid members. Your body of knowledge has to be big and interesting enough. This model works well in verticals where you can generate frequent insights that your customers will value. You can augment the Content Access model and apply various access levels based Tiered Pricing. You can also use this model in conjunction with the Freemium one, offering a free insight as a means to on-board prospects to your funnel.
Scoring Model
We are all in a tsunami of data. Most people (and organizations) try to keep afloat using the benchmark strategy, aka wisdom of crowd. We often rely on established entities and KOL’s and follow their lead. Insights-as-a-Service providers are well positioned to offer, or power an offering that generates the industry standards by scoring the players or products, and determining who are those opinion leaders. In markets where these scores change frequently, this could be a nice addition. Just imagine adding your own scores to the Content Access model.
Summary
Designing the right business model for your Insights-as-a-Service offering can be challenging. It requires trial and error, but you can learn from others. Go over the above models, mark the ones that might be relevant for your company. Then try to bundle 2–3 of them together (e.g. Tiered Pricing with Freemium). Use temporal planning to strategize how you evolve over time (e.g. penetrate the market as Educational Expert or Exclusive Membership and grow into Content Access and Scoring Model). And tweak the model continuously as you learn.
Remember, unicorns are made up of great tech served in the right business model.
How to design your Insights-as-a-Service business model
8 business models for Insights-as-a-Service
Yesterday at an AI conference, I saw an unusual number of companies offering Insights-as-a-Service. One after the other, brilliant entrepreneurs got on stage to pitch a very similar story: ‘we can help your organization uncover business insights hidden in your big data, using our unique machine / deep learning platform.’
Since almost every company today is a big-data company, it comes as no surprise that so many startups are drawn to offer Insights-as-a-Service. Which raises the million dollar question: what does it take to succeed in an Insights-as-a-Service model? What do these startups need to do to prevail?
If you’re one of those entrepreneurs you probably think it’s the data. You quote the economist’s famous saying that data is the new oil. You augment that with the ‘data race’ assuming that the company that will reach the critical data-mass first will dominate the market using the data network-effect.
That may be true, but also extremely challenging to fund. Especially in light of the huge amount of data needed to train AI models today. So if you want to win the data-network effect, you’d need to find a solid business model that will get you there. A data generating business model that can fund your growth, or at least support it.
Here are 8 business models you may want to consider. Feel free to tweak or mix n’ match to fit for your specific company:
Professional services
Perhaps the most common model for market penetration. You offer your platform for a subscription fee, but you also offer professional services to help first time users. These are provided by consultants, advisors, vertical experts or AI specialists. They can be part of your team or via partnerships. Your customers will write them off as NRE’s, so they won’t count against your subscription fees. While this model is great to generate some success stories and references customers, it is not a scalable model. You’ll need to evolve this model into one of the following models in order to grow to customers who do not need such scaffolding.
Educational Expert
The Professional Services model works well when augmented with the Educational Expert model: conferences, course, publications, etc. These help establish credibility and position your company as a market expert. It also helps attract customers to other subscription models. When an industry expert recommends a certain tool or when customers know this expert is using your tools to come up with great insights, they’ll want to have access to the same tool. You can partner for these educational services, and offer free or discounted version to such KOL experts. Unlike Professional Services model, Educational Expert model has a much longer effect and as such is a good seed to plant going forward.
Freemium
Free has a strong appeal. But it’s only good if you can convert free users to your paid packages. The most common levers for conversion are: volume (time, content, interactions…), duration, features, service level and customization. In B2B, free users are often companies that trial your service before making a purchase decision, or small companies (or teams) that get enough value off your free service. Don’t discard the latter as taking advantage of your Freemium model. Consider them as your community. They can help generate data which feeds into your paid products. And they are often your biggest supporters giving your marketing efforts an authentic voice.
Tiered Pricing
Similar to Freemium, tiered pricing can also work in the same dimensions of volume, duration, features, service level and customization. But it’s uniqueness stems from your ability to design tiers that match the right price for the right value of each specific segment. Take Salesforce Premiere support as an example of fast turnaround on support tickets. This tiered price plan is targeting organizations where real time is critical. Others offer tiers that allow (and even encourage) customers to grow their usage and upgrade over time. Whatever you do, make sure to keep the gap between price and value to a minimum to avoid churn and shaming.
Exclusive Membership
Just like some B2C models offer an elite product or service to a small, yet exclusive group, so do some businesses. Often these private clubs set a high price as ‘self selection’ threshold. Just imagine what might be the price that professional investment funds would pay to get financial trading insights 5 minutes ahead of everyone else, especially if they knew that only a handful of players can get this service. This model can fit well with Tiered Pricing. It can also be used as an initial model during market penetration, first establish a high-end positioning before creating less expensive versions of the service.
Analytics
Leverage your customers’ inherent data generation ‘engine’ to build long term engagements. One of the key risks of Insights-as-a-Service is that customers may not need insights on a continuous basis. Some industries still evolve slowly (e.g. insurance). And some players may take a while to implement your last insights before they can consume new ones. By augmenting your insights with an analytics services (home grown or partnered) you not only help your customers implement and measure the impact of your insights, but you also keep a stronghold to know when its time for up-selling another insight. Provided you follow your own advice and apply AI over your analytics data to trigger your own actions…
Content Access
As you grow, try to build a body of knowledge, an accumulation of insights in a specific vertical. Offer access to paid members. Your body of knowledge has to be big and interesting enough. This model works well in verticals where you can generate frequent insights that your customers will value. You can augment the Content Access model and apply various access levels based Tiered Pricing. You can also use this model in conjunction with the Freemium one, offering a free insight as a means to on-board prospects to your funnel.
Scoring Model
We are all in a tsunami of data. Most people (and organizations) try to keep afloat using the benchmark strategy, aka wisdom of crowd. We often rely on established entities and KOL’s and follow their lead. Insights-as-a-Service providers are well positioned to offer, or power an offering that generates the industry standards by scoring the players or products, and determining who are those opinion leaders. In markets where these scores change frequently, this could be a nice addition. Just imagine adding your own scores to the Content Access model.
Summary
Designing the right business model for your Insights-as-a-Service offering can be challenging. It requires trial and error, but you can learn from others. Go over the above models, mark the ones that might be relevant for your company. Then try to bundle 2–3 of them together (e.g. Tiered Pricing with Freemium). Use temporal planning to strategize how you evolve over time (e.g. penetrate the market as Educational Expert or Exclusive Membership and grow into Content Access and Scoring Model). And tweak the model continuously as you learn.
Remember, unicorns are made up of great tech served in the right business model.
orensteinberg
You might also like
How Machine Learning Transforms Insurance
7 Unexpected Ways Your PM and BD Can Collaborate to Improve Your Business
Next Article7 Unexpected Ways Your PM and BD Can Collaborate to Improve Your Business