Web Based Dash Boards

Hi Crystal. Having spent 7 years in the management and implementation of financial systems I found the first obstacle is getting organisational agreement on the KPI's to be included. Everyone wants everything and in the end you have control the amount of information to be included. The number of external systems to integrate is another deciding factor. If you are implementing a new finance solution talk to your vendor and make sure they can seamlessly integrate into your other systems. Watch out for any solution involving CSV files as this is rarely seamless. Check the output files from other systems can hold your chart of accounts. Manipulating data add further complications. If you have to manipulate then some systems provide an ETL (Extract Transform & Load) solution. Finally don't overload your dashboard with too many charts, dials and tables. I hope this helps.

Crystal, I'd recommend you work with your team to define what you want in the resulting dashboards as a first step. Create mockups in Excel and get buy-in from your entire team for the specific mockups you create. When you’ve done this, consider it release 1 and speak about it that way. All new requests/changes should be slated for release 2 of the system. Once you have release 1 defined, you know a lot more about the features you actually need in the resulting system and won’t overbuy features. You’ll also know the key data source for the system, and the ancillary data sources you’ll need to include. Ensure the key data source, most likely your ERP system, syncs seamlessly with the new BI system. And ensure there are simple ways to sync with the ancillary data sources.

In my experience, dashboarding is an iterative process. If your company needs to do it, getting started is the hardest part. Once your team has version 1 up-and-running, there will be many new ideas for visibility into your data. Do somewhat controlled releases. Expectations will be set correctly and your team will gain ever more powerful insights over time. And you’ll maintain your sanity.

I'd just state to start small, do test runs, get comments and then add to the KPI's using the same methodology.

Make sure the user-group (per KPI) agrees on the definition of the KPI and the data sets included in the KPI. Too many times even simple terms like "sales" and "orders" are misunderstood.

Nothing is worse then having KPI's that report wrong data. Independently verify the results until you are sure that it gibes. Also, have explanations and a reconciliation where managerial data doesn't mesh exactly with financial data.

Dashboards are most useful when designed just like in your car. Real time info, pulled directly from the source, and mostly operational in nature (not finance). Think idiot light, too. Focus on exceptions, like over 90 days receivables rather than a profile of the whole range. A gauge showing cash, to alert the team to variance vs. plan. Overdue sales pipeline more important than total. Limit the number of things to look at, it is a KPI dashboard. Focus attention on what matters and what can be changed/fixed-what needs attention, not stuff like historic trend lines. Get buy in from sales and operations, the financial stuff alone is minimal. Be wary of trying to report too much, and picking the wrong KPI's are worse than none. There is a temptation from many to cherry pick stuff that shows the best, and bias towards reporting the easy stuff, stick to what matters.
I have used Domo, and iDashboards, both useful tools and recommend both.