PART 1

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

As an institutional researcher, I need to know if the demand for skills is peaking or declining so that I can make an informed decision if we should build curriculum around those skills or not.

As a recruiter I want to know not only what skills are in demand today but how demand for those skills will change in the future so I can better plan for future demand.

Value to Customers & Users:

Some of the most valuable data Lightcast provides to our users is our skills data. While it is very helpful to understand what demand looks like today and in the past, being able to understand how demand for those skills will change in the near future is immensely valuable.

Value to Lightcast

One of our biggest differentiators is our skills data and adding the ability to project how demand for skills will change strengthens that position. We also know from the Legacy BG side that skill projections were very popular among users and difficult to compete against. Adding this to our product will help with both new sales opportunities as well as renewals.

FINANCIAL IMPACT NOTES:

• We do not expect any ARR impact in 2023 given how late PI 7 is in the year, the time to close new sales deals, and the likelihood that Success has already worked ahead on these renewals
• ARR Calculations only include Edu Analyst impacts -- benefits to other BUs and API/Snowflake sales & retention would be in addition to these estimates

Assuming Edu Analyst has:
• ARR = 17M/year

• Sales = 3.5M/year

• Revenue retention = 90%

Assume this Initiative yields:

• 1% improvement in ARR retention (incl. price increases) = 1% * $17M = $170k / year

• 3% improvement in sales = 3% * $3.5M = $105k / year

• Total = $275k / year

Calculations:
• 2023 ARR = $0 (per note above)
• 2024 ARR = $275k (per calculation above)
• 2025 ARR = (90% x $275k) + 105k = $353k

Target User Role/Client/Client Category

All Analyst platform users (e.g. Skillabi and Alumni Pathways) and Skills API customers.

Delivery Mechanism

Skills Projection API and Analyst platform.

Success Criteria & Metrics

Analyst will be able to consume and use projection categories for the majority of our skills and Analyst will be able to show a projection for a skill within the context of an occupation.

Qualitative feedback from Beta users across the different BU’s indicating that they are confident in our methodology and implementation of Skill Projections. This means they trust skill projections to give signals about whether skills are stable, growing, or decreasing.

Aspects that are out of scope (of this phase)

Any workflows or implementations of skill projections beyond the original implementation which was adding skill projections to our standard skills visualizations.

PART 2

Solution Description

Early UX (wireframes or mockups)

Open

Dependencies

No current dependencies for this PI. There will be future dependencies and those are defined below:

• Open questions on dependencies:

  • Overall ownership: Where does the models team work end and the micro team's work begin? Does models create the estimates only? Or also the buckets?

  • Final data: how do we take the above responses and additional validation and get to a point where we say "these are our skill projections, we're done"?

1. Models team creates/updates the model and makes available to Micro

2. Micro team creates/updates the API endpoints that make the model consumable by Analyst and API

3. An Analyst Front-End team creates/updates the API calls, filters, and/or packets that display this data in Analyst.

Legal and Ethical Considerations

No.

High-Level Rollout Strategies

• Initial rollout to internal Lightcast employees.

  • As part of this internal rollout we need our Go-to-Market (GTM) documents ready for internal employees' review.

    • We will need a methodology document describing how skill projections work as well as something guiding users on the types of signals they should be taking from these projections.

• Initial external release will be to specific beta customers across the different BU’s

  • We have had numerous customers ask frequently about skill projections so we should target these customers for our beta.

• Once we feel confident based on feedback from our Beta users we will release broadly.

  • Qualitative feedback from Beta users across the different BU’s indicating that they are confident in our methodology and implementation of Skill Projections.

    • Users largely agree or trust the results for our skill projections so while they may not agree with all our projections they feel confident enough in the methodology to still report results they may not agree with.

    • Users feel the documentation we have is not only enough to answer their questions but also to answer or get buy in from their stakeholders.

Risks

We need to make sure that our methodology is very defensible and that our documentation is very solid (Analyst PM’s will be responsible for creating and maintaining the customer-facing KB articles). We can utilize the documentation already created by Rohit to give Lightcast employees a better understanding of how the model works. Since we released and then had to retract skill projections we want to be sure that we are confident in the model and are able to instill confidence in our users.

Open Questions

• On Dependencies

  • Overall ownership: Where does the models team work end and the micro team's work begin? Does models create the estimates only? Or also the buckets?

  • Models will own the buckets as well.

  • Final data: how do we take the above responses and additional validation and get to a point where we say "these are our skill projections, we're done"?

• Technical: What issues do we need raised for API-dev, Analyst, TT, and any other to get these out and sequence appropriately

  • Everett and Jeff are creating issues for their respective teams.

  • We feel we are ready to hand off to Micro.

• GTM: What collateral do we need, and who is going to build that, in order to get this pushed through?

  • For the orignal Lightcast skills projections release Rohit created this document which we can use as a starting point: https://docs.google.com/document/d/1CUHGhfYywwxmI7cFTC9TL73bgWou-dCH/edit

Complete with Engineering Teams

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.