LOTv6 English for AUS, NZL, SGP postings

LOTv6 English for AUS, NZL, SGP postings

 

Created Date

Oct 25, 2022

Target PI

 

Created Date

Oct 25, 2022

Target Release

PI#5

Jira Epic

https://economicmodeling.atlassian.net/browse/CE-334
https://economicmodeling.atlassian.net/browse/DT-2896
https://economicmodeling.atlassian.net/browse/DATA-1634
https://economicmodeling.atlassian.net/browse/TX-782

Parent Project Page

https://economicmodeling.atlassian.net/wiki/spaces/DQ/pages/2674229283

Document Status

review

Epic Owner

@Hal Bonella

Stakeholder

@John Pernsteiner @Daniel Leadbeater @Bram Velthuis @Emma Gifford (Deactivated) @Tatiana Harrison

Engineering Team(s) Involved

DOCUMENTS micro C&E

Customer/User Job-to-be-Done or Problem

When looking at demand data/job postings, I want to know what occupations job postings represent so I can better understand the labor market conditions in AUZ, NZL and SGP. This understanding needs to be at a tactical level appropriate to my query and precise enough and broad enough to make decisions on.

Additionally the comparison with other Lightcast data through a common standard standard (Lightcast Occupation Taxonomy) is useful when comparing international markets.

NB - This work is separate from the roll-out of LOTv6 on legacy Emsi global postings, and refers to the logacy BG data in these three primary geographies. It is essentially a “like for like” replacement for clients used to using BGTOccs. The LOTv6 launch will include SpecOccs for the first time in these geographies.

Value to Customers & Users

We are introducing the Lightcast Occupation Taxonomy in our main markets and we want to give customers in these geo’s the ability to use the data in the same way.

Customers will be able to understand global labor market demands better through a unified taxonomy that is both very granular (specialized occupation) but also rolls up to useful levels. This will enable strategic, tactical and operational decision making based on (near) real time demand data information.

Order of magnitude the total value of customers switching from legacy systems to go-forward is around $3 million in conjunction with ANZCO and SSOC.

Value to Lightcast

Migrating customers from legacy systems to systems that use the One Index. Required for feature parity Labor Insights <> Analyst/Spotlight.

Order of magnitude the total value of customers switching from legacy systems to go-forward is around $3 million in conjunction with ANZCO and SSOC.

This project fits in with both the Global Growth and Competitive Advantage initiatives.

Target User Role/Client/Client Category

All lightcast job posting end users for API’s and Snowflake. Will enable delivery via front end tools (Spotlight) at a later stage.

Specifically customers accustomed to working with the Lightcast Occupation Taxonomy.

All APAC customers currently on BG cloud wanting to switch over to go-forward products.

Delivery Mechanism

API’s and Snowflake.

Documentation in knowledge base.

Success Criteria & Metrics

Tagging all English global postings in AUS, NZL, and SGP with an Lightcast Occupation classification

Stretch goal of tagging all English documents in non-core geo’s (Example: Philippines, India, etc)

 Scope of PI#4

Must have:

  • @Nathan Triepke June 21st - Implemented legacy tagger rules to cover gaps in AUS and NZL

  • @Nathan Lambert June 21st - Implement UK fast text model in AUS, NZL, SGP

  • @Oree Wyatt and @Jackson Schuur June 26th - Involvement to pass through fields?

  • @Nathan Craig June 26th (when pipeline is live on new index) - AUS, NZL, SGP 2-digit LOT Career Area change report, per year

    • This is from v3.4 in ANZ and v4.11 in SGP (expected and unexpected)

    • Plus Occupation level if possible

    • We have a change log between v3.4 and v6, but the report creation is still very complex.

  • @Nathan Triepke July 3rd - Quick analysis of change, make strategy call on what needs to be done for QA / burn down and estimate of timeline.

    • Decide start date of burn down: Second half-PI4 or later?

    • Workload risk due to UKSOC 2020 project

  • Burn down kick off - Unknown date

    • Train several teammates on editorial team to create rules for AU,NZ,SG

Must have:

  • @Nathan Triepke June 21st - Implemented legacy tagger rules to cover gaps in AUS and NZL

  • @Nathan Lambert June 21st - Implement UK fast text model in AUS, NZL, SGP

  • @Oree Wyatt and @Jackson Schuur June 26th - Involvement to pass through fields?

  • @Nathan Craig June 26th (when pipeline is live on new index) - AUS, NZL, SGP 2-digit LOT Career Area change report, per year

    • This is from v3.4 in ANZ and v4.11 in SGP (expected and unexpected)

    • Plus Occupation level if possible

    • We have a change log between v3.4 and v6, but the report creation is still very complex.

  • @Nathan Triepke July 3rd - Quick analysis of change, make strategy call on what needs to be done for QA / burn down and estimate of timeline.

    • Decide start date of burn down: Second half-PI4 or later?

    • Workload risk due to UKSOC 2020 project

  • Burn down kick off - Unknown date

    • Train several teammates on editorial team to create rules for AU,NZ,SG

Nice to have:

  • July 5th - Quick evaluation of Singapore? Release ready within PI4?

Not in scope:

  • Accuracy sampling by DS team (to take into PI5)

 Scope of PI#5

Must have:

  • Burn down complete @Nathan Triepke

  • Pipeline run finishes on September 22nd (Past end of PI#5)

  • Data available to clients on that date.

  • Data Delivery - Replace the data that is already in the dataset from the global classifier with this “local” classifier. Need to confirm exactly how that will work. @Matt McNair ?

  • Client coms? Joint effort @Hal Bonella , @Nathan Triepke and @Daniel Leadbeater

Must have:

  • Burn down complete @Nathan Triepke

  • Pipeline run finishes on September 22nd (Past end of PI#5)

  • Data available to clients on that date.

  • Data Delivery - Replace the data that is already in the dataset from the global classifier with this “local” classifier. Need to confirm exactly how that will work. @Matt McNair ?

  • Client coms? Joint effort @Hal Bonella , @Nathan Triepke and @Daniel Leadbeater

Nice to have:

  • Accuracy sampling by Data Solutions team (To be done in PI#6) @Tatiana Harrison

Not in scope:

  •  

Success Criteria & Metrics (multiple PI’s)

Aspects that are out of scope (of this phase)

Out of scope; Front end work, other classifications. BG Cloud

Solution Description

Dependencies

LOT v6 for English (UK?) postings

Legal and Ethical Considerations

Just answer yes or no.

Have you thought through these considerations (e.g. data privacy) and raised any potential concerns with the Legal team?

High-Level Rollout Strategies

Roll out on API and Snowflake as initial delivery scope. Additionally will inform Legacy Burning Glass customers in Australia and New Zealand specifically.

After evaluation and feedback, new development and rollout based on specific feedback.

Risks

Bad crosswalk/rules resulting in poor tagging

Open Questions

What are you still looking to resolve?

  • Todo: Data quality assurances

  • Will the data be ready in time for Micro to deliver their part within the timeline of this PI?

Useful links:
Summary of data needs:
https://economicmodeling.atlassian.net/wiki/spaces/DQ/pages/2674229283


Complete with Engineering Teams

 

Effort Size Estimate

Estimated Costs

Direct Financial Costs

Are there direct costs that this feature entails? Dataset acquisition, server purchasing, software licenses, etc.?

 

Team Effort

Each team involved should give a general t-shirt size estimate of their work involved. As the epic proceeds, they can add a link to the Jira epic/issue associated with their portion of this work.

Team

Effort Estimate (T-shirt sizes)

Jira Link

Team

Effort Estimate (T-shirt sizes)

Jira Link

Taxonomy

Medium

 

Documents

Small

 https://economicmodeling.atlassian.net/browse/DT-3629

Micro

Small

https://economicmodeling.atlassian.net/browse/MIC-1598