Analytics — 5 mistakes that companies make

A SQL database would surely do the job here, for half the price.

The price tag may be the single issue that kills your project.

Putting the BI/Analytics department under IT/Marketing/FinanceTo achieve true Data Democracy, your Analytics department cannot be under a functional area.

It needs to be side by side with them.

Otherwise, it will be strongly biased towards that area and will pull the company to it.

If it is under IT, it will be technically focused and will not be able to provide new products and revenue streams for the company.

If it is under Finance, it will be too focused on cost/return and may undermine innovations, experimental projects and non-profit projects (think community well-being and such).

If it is under Marketing, it will be a dream factory focused on fancy words, and may not bring in the desired results.

A balance of technically strong, creative people with an eye for efficiency is necessary for a good data analytics operation.

Separating Data Science and Data Engineering teams, or blending data scientists and engineers into a single personaSome managers like to mix these two totally different skill sets into a single persona.

Although they almost sound the same, they should be two separate teams.

Data Engineers are IT specialists working alongside functional areas.

Data Scientists/Analysts are visionaries, more prone to mathematics and statistics.

You might have a single person doing both, but it doesn’t scale.

A tiny operation might work for a while, but those guys will have to specialize sooner or later as your projects grow.

Data Science and Data Engineering teams need to be working together as much as possible.

Put them under the same manager, on the same room, sitting across from each other, on the same email groups, etc.

There is not one without the other, and a middle man will just get in the way.

Data engineers need to see their work being used, and they might identify minor improvements that a data scientist will thank the Gods for.

Data scientists need to see their data making its way to them, and they might save the data engineering a lot of time in stuff they might not need.

Wasting perfectly good dataLots of wasted potential all around.

Companies that are not data-driven are missing out on a lot of opportunities.

Management is so busy with operations that they constantly overlook the value of their data.

There should be a permanent watch for analytics potential within the company.

Useful data is not hard to identify, see a few easy things to keep a lookout for:a.

Customer behavior data Anything that may help segment, profile or cluster customers is good data.

It can be used to leverage marketing campaigns/recurring sales/aggregate sales and even targeting new leads.

b.

Supplier behavior dataSame idea.

Although I see the bigger waste here.

Companies don’t usually analyze their suppliers as much as their customers, and they are missing out.

A good relationship with key suppliers is priceless.

c.

Telemetry/sensor data/industrial dataThe same data that operations use for daily monitoring and tasks may be used for predictive maintenance.

And predictive maintenance is the shining star of advanced analytics.

If you are not using it, this is the time to pull out the ML algorithms and see those maintenance costs being cut to the bone.

Someone will get promoted here, I guarantee it!d.

Sales performanceROI needs to be in everyone’s mind.

Campaign tracking is crucial.

To know, with precision where each sale came from is to hold the power for bigger sales.

A visual tool is a must here.

e.

Warehouse/inventory statusAny data about inventory movement can identify waste, potential for optimization, future demand and improvements on loss prevention.

This is my honest opinion, I’d love to discuss this topic further.

As my first post, this is exciting! Please let me know what you think :).

. More details

Leave a Reply