How do I improve inconsistent and incomplete data documentation? What has worked for your team?
CIO in Education, 2 - 10 employees
Incomplete data documentation has caused a lot of issues for analytics and insights. Consolidating data silos and sources, integration of CMS platforms worked for our team to improve inconsistent and incomplete data documentation.QA Engineer in Software, 10,001+ employees
I think it would be best for the documentation to go through a review process, and even a double-check before it is posted (the final version).CIO in Consumer Goods, 10,001+ employees
This is quite a broad question, so a few considerations are necessary to determine actions.1. Is the missing/inconsistent data identifiable? Can it be intercepted by rules? Could these rules be applied to fix/normalise the data?
2. If it cannot be identified or addressed in 1. above, can the impact be quantified? Will the analysis be flawed in the absence of correction, or can it be excluded/ignored/minimised in terms of the impact on the outcome.
3. If after 2. it is still an issue, then do not use the output/outcome, because the decisions will be based on flawed inputs
Chief Technology Officer in Manufacturing, 10,001+ employees
A year back, we took three actions to improve this. 1. Hired an outside agency to train resources on how the documentation consistency, completeness and quality can be improved and what are the parameters that need to be kept in mind
2. A peer review mechanism started to check quality, consistency and completeness of documentation. For this a new step in the workflow was added and AI was used to select right peer to review. This is helpful as the creator can not collaborate with the peer reviewer in advance (as each qualified and first time right document enters into a raffle)
3. A competition where the best work has to be nominated by the creator and winners get a reward and recognition. Also, the best work was earmarked in our collaboration app so that those can be used as benchmarks for future documents
So far working well and now we have up the ante so looking to take this next level
Chief Information Officer in Healthcare and Biotech, 5,001 - 10,000 employees
Keep the documentation in a simple format so that it’s not viewed as an overhead. Build the review process in quick cycles so that the time gap is eliminated and there by procrastination of documentation
Carrot vs Stick for compliance is unavoidable
Chief Technology Officer (CTO) in Software, 2 - 10 employees
Having the documentation process as part of the development cycle. The issue is not complete until the documentation has been updated and verified. It also needs to become part of the company culture as a whole such that it will stand out when it is not completed. Fractional CIO in Software, 2 - 10 employees
Documentation is commonly left off the project plan. The rush to get the product into production and the on to the next project in many cases causes project documentation to be rushed and data documentation is incomplete. One important factor is to make sure that the documentation tasks are part of the project plan.CTO in Software, 11 - 50 employees
The first thing I would do is check with the data source to see whether they have a suitable data verification process in place along with tools that can effectively identify duplicate records in the data. The other thing I would focus on is establishing precise guidelines for data governance.Founder and CTO in Software, 51 - 200 employees
Improving the documentation coverage and consistency through peer reviews and steps in the process has helped our teams to some extent, although we still find ourselves in situations where the documentation is stale and perhaps has become outdated over time. Continuous efforts and regular reminders to consolidate and review documentation need to be in place to complement the aforementioned approaches to keep it consistent.
Chief Information Technology Officer in IT Services, 201 - 500 employees
We use our ISTM tools and knowledge base but we have also developed small video and recording our documentation which is more easy for our staff.Content you might like
Employee recommendations40%
Past experience62%
Peer reviews62%
Analyst reviews35%
Individual research25%
Brand recognition17%
204 PARTICIPANTS
Director of Data Science & Analytics in Healthcare and Biotech, 10,001+ employees
Below is a few measure to think about...1) Data Existence (Measure completeness of records and missing-ness by domain)
2) Accessibility (Measure number of reports pulled per month, totals users, number of new ...read more
CTO in Software, 201 - 500 employees
Without a doubt - Technical Debt! It's a ball and chain that creates an ever increasing drag on any organization, stifles innovation, and prevents transformation.