Moveworks Total Cost of Ownership (TCO) vs. Competition

V.1.0

​

November 2022

© 2021 Moveworks, Inc.

TCO - a primer

• What is TCO? “Total cost of ownership” is an estimate of all the direct and indirect costs involved in acquiring, deploying and operating a product or system over its lifetime.
• What costs are typically considered? Typical costs for technology include: 1) Planning, 2) Development / acquisition, 3) Integration, 4) Maintenance / upgrades, 5) Retirement
• Why is TCO important? Most all Procurement departments are required to measure TCO to inform investment decisions. However, while TCO is important, it is not the only consideration – other data points include, net present value, internal rate of return, payback period, etc.
• What does our TCO model specifically measure? The current version of our TCO model looks at Moveworks costs in comparison to those of a bot toolkit (e.g., ServiceNow VA, Amelia, Microsoft PVA, etc.). The costs considered include the following: 1) Bot Development / Implementation, 2) Translation Services / Multilingual Support, 3) Resource Development, 4) Software Licenses, 5) Hardware & Infrastructure, and 6) Ongoing Operations, Maintenance, & Enhancements. See next page for more detail.
• How does Movework TCO compare to that of our competitors? While our software license may be more expensive than alternative solutions in the market, our TCO when factoring in all costs over a multi-year period will be lower. See next page for more detail.
• In what sales scenario should I use this TCO model? When a prospect tells you TCO is an important piece of criteria, ask if we can help model this out, as we have experience doing this hundreds of times across our customer base. Note when you do measure TCO, you should always measure it in conjunction with a value opportunity analysis.
• When in the sales cycle should I use it? It can be used starting in Scoping (where you start to collect some cost-related inputs from your prospect), through Business Justification and Negotiation phases (where you are comparing / contrasting Moveworks TCO to the alternative solution(s) being evaluated). For more detailed, customized analysis, please engage your BVS rep.

If you have any questions, please reach out to your BVS rep!

© 2022 Moveworks, Inc.

TCO Cost Assumptions

© 2022 Moveworks, Inc.

Sample slides from a Moveworks vs. ServiceNow VA comparison

© 2021 Moveworks, Inc.

5

Development & Maintenance

Fully managed as a service with a dedicated implementation and customer success team committed to your success

OOTB solution with true AI, multiple capabilities, and thousands of intents that can be leveraged on Day 1 with low effort conversation design

Data-driven optimization with comprehensive analytics and actionable recommendations on how to improve overall bot performance and employee experience

$5.2M

Costs Saved Over SNOW VA over 3 years

Continuous innovation allows speed to deployment & value by leveraging platform extensibility for new use cases & continuous R&D & product innovation

​

​

© 2022 Moveworks, Inc.

TCO Considerations – Moveworks vs. SNOW VA (1 of 2)

6

Bot Development & Implementation

Multilingual Support

Resource Development

Faster Time to Launch & Value

​

8 - 12 weeks average implementation including all skills (form finding, knowledge search, concierge, etc). No model training (intents), conversational flow design / build, or integration work owned by <Company> team.

​

Project Team:

​

• 1 Project Manager to coordinate with process / tech owners, support management, ARB, etc.
• QA Testers to evaluate performance of bot prior to launch
• Tech stack systems owner support to create service accounts, participate in process workshops

�Deployment process:

​

1) Provide Moveworks access to core systems

2) Use case prioritization

3) Testing

4) Bot promotion and launch

Slower Time To Launch & Value

​

9 - 12 months at minimum for <Company> to convert current UNA capabilities to VA. Assuming enterprise-grade conversational AI program setup (Bot that supports all <Company> languages across a minimum of 80 topics).

​

Project Team (at minimum):

​

1) 1 Conversational AI specialist

2) 1 Conversation experience designer

3) 1 Technical developer/engineer

4) 1 QA tester

5) 1 Project/program manager

​

Deployment process:

​

Planning

1) Assemble team

2) Build out governance

model

3) Define use case(s)

​

6) 1 Data scientist

7) 1 Solution architect

8) 1 Implementation engineer

9) ML team

10) Tech stack systems owners

Implementation / Improvements

1) Bot promotion

2) Performance Measurement

3) Topic Identification

​

Development

1) Train NLU models

2) Build dialog flows

3) Connect to automations & KBs

4) Connect to chat platforms

5) Repeat for each use case/language

​

Included in subscription

​

Optimized for over 100+ languages out of the box which includes dynamic translation of all support experiences (e.g., dynamic flows, knowledge articles, forms, employee communications, notifications, etc.). NLP is localized to all intents in regional language preferences.

Additional costs

ServiceNow has support for 17 languages with varying levels of capabilities. For instance, today, only 5 languages support search, NLU performance, intent discovery, conflict detection, and system entities. Flows need to be translated in each language. More details here.

Lower cost

​

Integrations maintained for <Company>, i.e., AD and Workday. Reduced resource development & maintenance costs, e.g., knowledge management (KBA suggestions / External Knowledge) and catalog items (Form recommendations).

Higher cost

​

Knowledge management & maintenance costs will remain (currently out of Brazil / India, language hubs, etc.) due to lack of ML-driven resource improvements. Integrations into ServiceNow maintained by <Company>. New integrations built by <Company> or outsourced to third party.

© 2022 Moveworks, Inc.

TCO Considerations – Moveworks vs. SNOW VA (2 of 2)

7

Ongoing Operations, Maintenance & Enhancements

12 Month Achievements

Included in subscription

​

Model Maintenance: ML model maintenance optimized by Moveworks including model training, new model selection, data selection, additional data annotation, etc. Online / Offline tests performed to monitor model performance where AutoML will introduce improvements without human intervention. Human annotation team supplies supervised data to improve models, such as new entity identification. Collective learning provides high quality / high volume of data to maintain best-in-class prediction performance across all use cases and languages.

​

Conversational Design: Enterprise cache automatically indexes changes from back end-systems, e.g., if a service owner updates a support KBA, the article is auto-snippetized, and if a form is published it is automatically transformed to a chat native experience. The dynamic flow engine (action-bid) results in real-time ranking of support answers. Moveworks therefore recommends the best answer at all times without manual intervention, such as if a transactional use case is introduced to automate a process where the previous best answer was informational. Net-net, no maintenance of flows.

​

Integration Maintenance: Complex integrations such as real-time changes / caching of support resources in non SNOW systems such as AD, Workday, and Coupa maintained by Moveworks.

​

Dedicated Success Team: Expert team including engineering, data science, and engagement manager, focused solely on driving adoption and performance of Una.

​

20 - 50% of dedicated time from FTEs

​

Model Maintenance: As more topics and use cases are added, ML specialists will need to carefully spot potential keyword/phrase triggers that may indicate multiple intents. Model management and achieving high performance across a large number of intents is challenging due to data constraints, requires manual input of new utterances and rework when conflicts occur. ��Conversational Design: As new topics are defined, existing support material such as KBAs must be translated into conversation flows to achieve best experience. New subtopics require rework of existing flows creating a bottleneck on Una team. Changes to applications and processes require rework to conversation flows, which at scale will result in delay as <Company> has over 1000 application that will require some form of support experience. ��Integration Maintenance: As all integrations are maintained by <Company>, the team should expect additional testing during upgrades. Net-new integrations must be developed and can result in significant delay of new functionality. ��Dedicated Success Team: Additional paid packages may not be experienced in VA rollout at scale. Often focused on assisting with the above three categories + new topic creation rather than transformational efforts.

Service desk efficiencies & employee productivity achieved through autonomous resolution and acceleration of service desk tickets:

​

• > 1000 out of the box intents with > 50,000 mapped entities (collective learning) leveraging all existing support experiences without rework.
• True self-learning AI that improves without <Company> intervention.
• Take back ownership of employee experience from MSP + reduce support costs by lowering reliance on human support resources (internally and outsourced).
• Faster deployment and lower 3 year total investment of resources and effort for planning, development, and implementation.
• Modeled <Company> performance expects 252k auto-resolved inquiries in Y1.

Limited service desk efficiencies & employee productivity gains achieved:

​

• Launch with a subset of necessary topics, likely certain languages prioritized (based on current SNOW NLU functionality), resulting in lower than desired adoption.
• In the process of launching new use cases without ML-driven prioritization.
• Manual work to maintain existing support experiences (not true AI).
• Larger reliance on live chat experience due to support gaps, promoting behavior of leveraging Una as a live chat service (vs. AI support) resulting in higher human support costs long term.
• Continued MSP reliance, adding to additional total cost of ownership for Una
• Main performance metrics for bot program becomes “Deflection Rate” (over-indexing success on chats not resulting in case creation, which may perpetuate “watermelon effect”).

​

© 2022 Moveworks, Inc.

TCO Considerations – Annual Comparison

8

© 2022 Moveworks, Inc.