Odds are you know your business needs business intelligence. Over the past 5 years, big data and BI became more than just buzzwords. Without real time insight into their data, businesses remain reactive, miss strategic growth opportunities, lose their competitive edge, fail to take advantage of cost savings options, don’t ensure customer satisfaction… the list goes on. In response to this increasing need for data analytics, business intelligence software have flooded the market. With the benefits of BI being numerous and the costs of not having good BI growing, it is easy to want to quickly adopt a solution.
Unfortunately this approach could be disastrous. Investing in BI shouldn’t be taken lately. Whether you are starting from scratch, moving past spreadsheets, or looking to migrate to a new BI platform: you need a business intelligence strategy and roadmap in place. We previously discussed big data strategy for small business. Now we are going to take that a step further with the following 11 steps to a better business intelligence strategy. These steps are imperative for businesses, of all sizes, looking to successfully launch and manage their business intelligence.
1. Go into the process with eyes wide open
When you have the right business intelligence solution, it is easy to identify trends, pitfalls and opportunities early on. But implementing the right solution isn’t always easy. Actually, it usually isn’t. We are going to be honest here, even the best BI software needs some initial heavy lifting to maximize its potential. If you go in with the right mindset you will be prepared to address issues like complicated data problems, change management resistance, waning sponsorship, IT reluctance and user adoption challenges. Reminding stakeholders, and yourself, of the pain points that necessitated BI will encourage the process forward. It will be worth it.
2. Determine stakeholder objectives
Odds are everyone at your organization could benefit from increased data access and insights. That doesn’t mean they are all key stakeholders. Right off the bat you must determine who your key stakeholders are. Then find out what they need: visible and vocal executive sponsorship is a must. Gathering and setting executive team expectations early is paramount. Then move past the executive team. They often don’t have the same front line knowledge that other staff do. Collect and prioritize pain points and key performance indicators (KPIs) across the organization. They might not all make it into the initial rollout but it is better to start big and roll back.
3. Choose a sponsor
While a business intelligence strategy should include multiple stakeholders, it is imperative to have a sponsor to spearhead the implementation. It may be tempting to place the Chief Information Officer (CIO) or Chief Technical Officer (CTO). This is usually not the best approach. BI should be sponsored by an executive who has bottom-line responsibility, a broad picture of the organization’s strategy and goals and knows how to translate the company mission into mission focused KPIs.
CFOs and CMOs are good fits. They can govern the implementation with a documented business case and be responsible for changes in scope. Of course whoever the chosen sponsor is, they will need to be in constant communication with the CIO/CTO. Which brings us to the next step…
4. BI is not just a technology initiative
We are going to repeat ourselves a bit here. Because it is that important. To succeed a BI deployment must have the support of key business areas, from the get-go. IT should be involved to ensure governance, knowledge transfer, data integrity, and the actual implementation. But every stakeholder and their respective business areas should also be involved throughout the process.
By involving a range stakeholders you can ensure you cover the three broad classes of business intelligence users: strategic, tactical and operational. These different users types will need customized solutions. Understanding who will use the data and for what purposes can show the type of information needed and its frequency, and help guide your decision making.
The business as a whole must be willing to dedicate the necessary resources: staff, IT resources, costs, etc. BI implementation doesn’t just come out of the IT budget. The best business intelligence strategy lays out these resources in the beginning, with additional wiggle room.
5. Employ a Chief Data Officer (CDO)
Big data and BI guru Bernard Marr recently wrote about The Rise of Chief Data Officers. In the article, he pointed to a pretty fascinating trend: “Experian has predicted that the CDO position will become a standard senior board level role by 2020, bringing the conversation around data gathering, management, optimization, and security to the C-level.” We love that data is moving permanently into the C-Suite. While, like the CIO, the CDO probably shouldn’t be the main sponsor for BI implementation: they (or a similar role) are a great key stakeholder to involve. They will also most likely own the BI after the initial implementation is complete.
6. Assess the current situation
As we have already stated: Usually BI deployment isn’t quick or easy. There is a lot of work to do on the front end. One of the biggest sections on a business intelligence roadmap should be assessing the current situation. Now that you have all the right stakeholders at the table the next step is analyzing the current software stack, and the processes and organizational structures surrounding it (or lack thereof). Find out what is working, as you don’t want to totally scrap an already essential report or process. Find a way to integrate it into the new strategy, or you will have upset employees. On the flip side, document everything that isn’t working. What business questions are you unable to currently answer? Which processes are inefficient or broken?
On top of all this you need to compile which data sources you currently have and how they are being stored. Decide which are necessary to your business intelligence strategy. This should also include creating a plan for data storage. Are the data sources going to remain disparate? Or does building a datawarehouse make sense for your organization?
As with all these steps, both IT and the various business stakeholders should be involved throughout this hefty step.
7. Clean the data
Clean data in, clean analytics out. It’s that simple. Cleaning your data may not be quite as simple. But it will help ensure the success of your BI.
8. Develop a “Data Dictionary”
With Agile development, extensive documentation has become a faux-pas. Large data dictionaries can be cumbersome and hard to keep updated. That said, for BI to succeed there needs to be at least a consensus on data definitions and business calculations. The lack of agreement on definitions is a widespread problem in companies today. For example, finance and sales may define “gross margin” differently, leading to their numbers not matching. To nip this in the bud, get all the SMEs at the same table to hammer the definitions out. Then for knowledge transfer choose the repository, best suited for your organization, to host this information.
9. Identify key performance indicators (KPIs)
While your organization is setting data definitions, the right KPIs should also be determined. KPIs are measurable values that show how effectively a company is achieving their business objectives. They sit at the core of a good BI strategy. KPIs indicate areas businesses are on the right track and where improvements are needed. When implementing a BI strategy it is crucial to consider the company’s individual strategy and align KPIs to company’s objectives. It may be tempting to create KPIs for everything. This can be a runaway train. It is best to start with the most important KPIs; then create standards and governance with the KPIs in mind. You can always expand on these later.
10. Choose the right tool / partner for your business
At step 10 we finally get to choosing a BI software/partner. Yes, you are this far along in your business intelligence roadmap and you don’t even have a tool yet. By preparing properly through steps 1-9 you will be best suited to find the right tool and implement it successfully. During this process you will need to choose between on-premises and cloud-based BI solutions. You also need to make sure to choose a solution that can start small but easily scale as your company and needs grow. Look for flexible solutions that address the needs of all your user. Take advantage of free trials, and don’t rush through this step!
11. Pursue a phased approach
Rome wasn’t built in a day: neither will your BI. A successful BI strategy takes an iterative approach. Think “actionable” and take baby steps. Choose a few KPIs and build a few dashboard examples. Gather feedback. Repeat again with new releases every few weeks. Continuously ask yourself what is working and what stakeholders are benefiting.
A good BI roadmap doesn’t have an end date. Your organization should be invested in it for the long term. You should be continually measuring and refining your processes, data and reports. Don’t let your BI become stagnate: continually raise the bar.
Use These Steps
With these 11 steps your business intelligence roadmap may look a bit daunting. Without them you will end up an even bigger headache. When done right BI is the gift that keeps giving. You just need to stick to your business intelligence strategy to get there.