Are You Wondering What Caused Your Dashboard to Break? It Could be Unexpected Changes in the Data. What to Look for When Things Break Down.
5 Ways Data Analysts Can Win the Trust of Business Users
The key to maintaining the relationship between technology & business is trust. Here are 5 ways data professionals can generate trust across the business.
Data and analytics has reached the big leagues of corporate strategy — and it’s now a team sport. Up-to-date analytics drive every aspect of modern businesses, which means data professionals are now in the service industry.
In many ways, the evolving role of data teams resembles the rise of IT in the ‘90s, when desktop productivity software turned every office worker into a PC user. When the whole workplace relies on the infrastructure you provide, uptime and other drivers of customer satisfaction become critical.
Whether it’s the IT team hunting down a VPN error that’s blocking an enterprise’s distributed workforce or the data team resolving an algorithmic glitch that’s hobbling Q4 forecasts, the key to the relationship between technology and business is trust.
Here are five steps to generate trust across your enterprise.
1. Treat business users as your customers.
In what ways does your company earn the trust of customers? What sort of commitment to reliability and responsiveness does it make via marketing, sales, and success and support groups? Use that model to inform the way you approach the internal customers who consume your reports.
The specific language of your customer satisfaction campaign will vary with your company’s external practices. In any case, you can’t go wrong by:
- Interviewing your internal customers to understand their needs and get feedback on the reports you prepare for them.
- Developing personas for the reports you generate. An exec needs different levels of detail than a marketing manager who just ran an SEM campaign and wants to understand its performance.
- Responding to their support questions very promptly. Consider using a ticketing system.
Roll out the red carpet to your internal customers, and they will trust your commitment to their needs.
2. Align your data goals with corporate strategy.
In most companies, the most trusted BI person isn’t the one who’s most technically advanced — it’s the one who best understands the company’s business.
Take time to understand the strategic goals of your company as a whole and of the departments you support. This deeper perspective will help with:
- Creating reports that better meet business users’ needs.
- Sniffing out potential inconsistencies in the data.
You never want to hear a business stakeholder tell you, “This doesn’t make sense. I think that you may be misunderstanding how our agreements with our customers are structured...”
3. Remember: Less is more.
As a data wonk, you’re probably eager to convey the sheer wealth of information available. All those metrics, cut across every dimension, are impressive and exciting.
Unfortunately, the average business user needs 10% of the data you may be throwing at them, and they don’t have the time or context to identify the relevant bits. Keep your dashboard clean and simple, focusing on the key, top-level metrics.
Remember the “personas” I mentioned above? If a user truly needs more detail, you can always create more detailed reports.
4. Release dashboards or reports only when they’re fully vetted.
You have a brand-new, exciting report, and you can’t wait to see how it’s received. Then you release it to your business users, and disaster strikes: They find a flaw that you hadn’t noticed, and all the trust you’ve painstakingly developed over the past few months goes up in smoke. You have to unpublish the report until you get it fixed.
Business users have a very different context from yours. They are able to reality-check your data with factors outside your purview. For instance, you might report no sales in a given week, when they know full well that they personally closed two deals on Tuesday.
You can’t expect to find all the issues yourself, even if you triple-check your data and logic. As with any software, you need to get these reports into the hands of a few customers who can help you debug it. Release the report as a beta (and label it as such) to a few business users whom you trust and respect, and have them test it. Get their feedback, iterate a few times, and release it only once you’re confident that it is accurate and useful.
5. Maintain consistency from one dashboard to another.
Let’s say you have two dashboards for two different products, and you are using the same KPIs to measure the performance of these products. You need to ensure the two dashboards have the same structure and use the same metrics.
Each time users see a report with a different structure, they need to invest brain cycles to adjust their interpretation. This sort of context switching comes at a cost to trust. For example, if one report compares actuals to budget by using “% of budget” (e.g., “107% of budget”), make sure that the other report doesn’t use a “variance to budget” (e.g., “+7%”). Ensure consistency. The business user will absorb the report a lot faster.
Think of cars. Whenever you get into a new car, there are things you know instinctively: the location of the brake and gas pedals, how the steering wheel works, even how to activate the blinkers. However, the cruise control and sound system are usually different from other car models, and therefore they take time to understand. Because it takes time to figure them out, you’re likely to skip using them if you’re in a rush.
Now think of your reports: How many different cruise controls do you have in them? Are your customers likely to skip them because they take too much time to figure out?
6. Indicate the freshness of your reporting.
If they’re supposed to trust your data, business users shouldn’t have to wonder if it’s current.
Spell out when the data was last updated, how frequently it’s updated, and whether there’s a lag in reporting. For instance, you may report weekly on a given metric, but you might be receiving the data with a two-week lag. At the beginning of Week 10, you are updating the report with data through the end of Week 7. That’s fine, but make sure your business users understand the lag. Otherwise, they’ll assume your report is simply out of date and therefore not to be trusted.
The Bottom Line
Any good analyst will be excited about the richness and flexibility of the data sets they generate. And that’s a good thing! You want to convey that passion to your peers on the business side.
But remember: Your data is only useful to the extent your colleagues trust it and act on it to achieve organizational goals. It falls to you to understand their priorities and provide data in a way they can assimilate and ultimately act on.
IN PRIVATE BETA
Simple Dashboard Tools to Build Trust with the Business
Mighty Canary is building tools that will allow you to be more proactive when dashboards and data change. Spend less time answering the same questions and more time building meaningful dashboards that drive your organization forward.
Sign up to try the beta and start building more trust with your stakeholders.