As business intelligence consultants, our clients are increasingly asking us: How can we go faster?
Common complaints are a data warehouse team that misses deadlines, business folks grumbling about how long estimates are, and an enhancement backlog more than six months. Often teams report they are cutting corners to get all the work done.
These are symptoms of a team that has problems with execution. If these problems sound familiar, don’t worry, there are concrete steps you can take with an existing team to execute better and become more productive.
Here are the top five causes of poor execution:
1. Poor project management
How many new ETL processes have you written in the last six months? How long does each one take? The typical answer is “we don’t know.” Without measuring productivity, it is hard to improve it. In particular, many teams do not know the full cost of developing a new process, including soft items like status meetings, managers, fixes, and changes.
2. Lack of quality control
Do you have to write the dreaded entire company system-not-available email? Do you spend most of your time fighting fires? Does something break every week? Surprisingly, spending time on quality often reduces total implementation time. Prevent problems by putting more effort into design, testing, and fixing the root cause of problems. Skipping test and reviews is fast in the short term, but slow in the long run.
3. Slow system performance
Do you have to create summary tables for specific reports? Do you test on partial data or only run tests once or twice because processes run a long time? Do developers spend a lot of time clicking refresh waiting for reports and ETL to finish? Slow systems have a big impact on development and test productivity. Tests run longer or get skipped. Unnecessary summary tables and extracts to fix slow queries take effort that could be used on new features. Performance tuning can dramatically improve productivity.
4. Fragmented team design
Do the front end, ETL, database, and modeling folks work on the same team? How many reviews does a change go through before production? Are modest changes reviewed by data model or architecture groups outside your team? Are your DBAs data warehouse specialists? Imagine being able to make changes the same day instead of waiting for a weekly review meting. Having all the right skills working together in one place will dramatically speed your execution.
5. Missing tools and automation
Is migration between environments time consuming? Do you use database tools like Toad and Erwin? How long does it take to refresh data in your test environments? Do you manually download files and compile reports every week? It is easy to let repetitive tasks suck up productive time. In particular, repetitive tasks that have fixed time frames drive out new development. They also demoralize staff. Small investments in tools and automation add up incremental to higher productivity.
Over the next few weeks, we are going to explore each of these causes in more detail, and discuss specific actions you can take to speed up delivery without compromising the quality of your business intelligence program.
by Tom Victory