If you’re not analyzing your customer speech and text, you’re getting an incomplete picture which could be misleading your decisions.
Furthermore, discovering the top topics from unstructured data is certainly important, but has anyone noticed that top topics alone are not actionable?
It takes strong human intervention to align text topic categories to your business decision areas. Additionally, a great text analysis should be living/breathing–ensuring it stays relevant as your business and markets change.
In this presentation, we’ll help you understand what steps you need to take to make your text analytics truly actionable according to your own stakeholders.
This presentation forms part of the online Festival of NewMR webinar series. Presented by Alison Bushell and Jason Bryce from Confirmit.
“Tried ordering on your website it wouldn’t accept the offer
code which was frustrating enough, then I called and spoke to
your agent Julie who was very pleasant and professional but
could not fix the problem either. Not impressed.”
Order Web Site Staff Sentiment
your website your agent
§ Categorizing your data can be a
§ Organizations are turning to AI
powered technologies to speed
up the process
§ AI isn’t a solution on its own
Getting to Grips with Your Data
AI or Human?
Approaches Advantages Disadvantages
A.I. (automated) § Saves time
§ May identify previously
§ Low Precision
§ Not actionable/not business-aligned
§ Not easy to explain nor
Rule-based (human) § High Precision
§ Easy to explain and
§ Requires human expertise
§ Not as quick
§ Risk of Overfitting
Maximizes advantages and minimizes disadvantages
Getting Emotional – Categorization Model
Awareness Purchase Service Claim Renew
7 steps for you to take away
and think about for your text
Your Checklist for Success
§ Hard lessons are expensive
§ Competencies are available
to help you succeed
1. Don't Reinvent the Wheel
¨ Read up on it (thoroughly)
¨ Talk to experts
¨ Get help
2. Do Your Due Diligence
¨ Meet with stakeholders and understand
¨ Get buy-in
¨ Plan forward. Manage backward.
¨ Understand your data and where it's
¨ Understand what A.I. can and cannot do
What to Understand:
§ Journey maps
§ Org charts
§ Strategic and operational plans
§ KPIs and company metrics
§ Pain points
§ Corporate goals/mission/vision
§ What executives are on the hook
3. Draft a Framework Examples
Operational Strategic Customer Journey
Year 2+ benefits
4. Make the Most of AI Topic Extraction
Guide Refine Solo
Tell it what specific
topics or concepts
to build around
Use Boolean to
add, remove or
5. Audit & Test
¨ Scrutinize uncategorized bucket
¨ Meet with stakeholders again after results are available
¨ Are the categories working?
¨ Are results too detailed, not detailed enough?