In previous articles, we have covered some of the practical steps that IT teams can take around deploying analytics and getting people to adopt these tools. However, one area that we have not touched on is how to do this when your users aren’t your employees.
Embedding analytics into your products can be a lucrative new revenue stream. According to research commissioned by Birst, about 74 per cent of companies are considering how to offer their customers new analytics or data products, while 71 per cent found that including analytics in their products provides a perceived competitive advantage. This represents a great opportunity for companies to increase market share and customer spend, but it does also represent a risk.
In the same survey, about 50 per cent of respondents stated that they had failed in deploying analytics using in-house tools. Additionally, 100 per cent of those surveyed admitted that deploying analytics without a concrete plan was doomed to fail. This begs a question: how can companies create a successful path for developing analytic products and applications?
Strategies for creating successful analytic products
The first step to take is to look at embedded analytics projects as meeting different goals compared to internal analytics roll-outs. While internal projects have a defined set of users with knowledge levels that can be understood from the start, embedded analytics roll-outs often have a much larger group of users with a far greater spread of awareness.
To make a business case for creating analytic products, it’s important to check with existing customers and prospects on what their expectations are, both now and in the future. Would analytics be a valuable offer for them, and therefore an additional revenue opportunity? Or would it be a loss leader strategy, to keep customers tied into core products that now embed analytics in their interface?
Another thing to consider is the customers’ current knowledge of the potential for analytics. Are customers data-savvy and aware of what they can do with new sources of information, or are they satisfied with standard reports and visualisations to start off with? Based on these answers, companies can rationalise analytic capabilities that they offer to their customers.
In thinking about pricing and packaging, it’s worth looking at whether to offer one product to everyone, or tier services depending on their level of spending. Trying to take a ‘one size fits all’ approach can make it simpler to sell and deploy, but it can risk and devalue advanced functionality that only some customers need and would be willing to pay for.
For example, simple reporting and dashboards can be applicable to every customer, but some may want to go further in “mashing up” different data sets to generate richer reports and more insight. The value can be substantial for a subset of customers, while the work required in maintaining these services is also considerable. In this case, the company may decide to offer data mashups as an advanced service to only some clients who are willing to pay a premium for it.
Supporting embedded BI projects
It’s important to recognise that once you start offering standard reporting and analytics, your customers’ demand for data increases. Rather than being satisfied with simple reports, customers will start asking for more detail or customization around the analytics they get. This increase in the scale of analytics should be factored in from the start. It is a good idea to set boundaries in the level of service you deliver. For example, you may include some report customization in your standard product, but require a premium for advanced cases.
Similarly, from day one, it’s worth looking at the costs involved in supporting embedded BI projects. Investing too much can stall the business case and make it impossible to deliver a return on investment. Conversely, not spending enough can lead to unhappiness in the longer term, as those customers on the new services get frustrated. Look for BI and analytic platforms that minimise staffing, headcount, and resource needs to support your embedded analytic products.
It is also worth looking at the different customer support options. This can involve creating new service options, some of which only include training videos, community forums, and education materials, while others offer time with the support and developer teams.
Planning ahead for analytics
Embedding analytics into a new service can be an attractive option for information-rich enterprises such as those in financial services, healthcare, telecoms, retail, and high tech. However, deploying BI and analytics is not something to be taken lightly. Rather than looking at this as a small one-off project, analytics has to be planned thoroughly to manage expectations both internally and externally.
The role of analytics can help build up stronger relationships with customers, and it can provide new revenue opportunities for the future. However, without a full picture of what customers do desire from data and insight, it’s possible to miss out on these opportunities.
Whether you aim to create large-scale adoption across thousands of customers, or define detailed services for a chosen few, it’s important that your embedded BI strategy follows your short and long term goals. Using a multi-tenanted cloud BI platform can help, as it lowers your Total Cost of Ownership compared with implementing traditional BI or individual instances of Cloud BI for each customer. You can scale out to many customers without the cost overhead of each, or you can scale down to a selected few that have deeper pockets, but require lots of iterations and customization. Ultimately, the benefit from cloud multi-tenancy is to eliminate setting up separate BI environments for every new development.
Companies looking at analytics as a new product or service need to examine their customer expectations against their development and support costs. Success lies exactly between customer value and positive product margins. A good product leader knows the right go-to-market strategy to make a successful analytic product.
The preceding article was originally published in It Pro Portal on June 13, 2016.