Reduce Manual Repetitive Work With IT Automation

Free up time for value-adding jobs.

ANALYST PERSPECTIVE

Automation cuts both ways.

Automation can be very, very good, or very, very bad.
Do it right, and you can make your life a whole lot easier.
Do it wrong, and you can suffer some serious pain.
All too often, automation is deployed willy-nilly, without regard to the overall systems or business processes in which it lives.
IT professionals should follow a disciplined and consistent approach to automation to ensure that they maximize its value for their organization.

Derek Shank,
Research Analyst, Infrastructure & Operations
Info-Tech Research Group

Executive summary

Situation

• IT staff are overwhelmed with manual repetitive work.
• You have little time for projects.
• You cannot move as fast as the business wants.

Complication

• Automation is simple to say, but hard to implement.
• Vendors claim automation will solve all your problems.
• You have no process for managing automation.

Resolution

• Begin by automating a few tasks with the highest value to score quick wins.
• Define a process for rolling out automation, leveraging SDLC best practices.
• Determine metrics and continually track the success of the automation program.

Info-Tech Insight

1. Optimize before you automate.The current way isn’t necessarily the best way.
2. Foster an engineering mindset.Your team members may not be process engineers, but they should learn to think like one.
3. Build a process to iterate.Effective automation can't be a one-and-done. Define a lightweight process to manage your program.

Infrastructure & operations teams are overloaded with work

• DevOps and digital transformation initiatives demand increased speed.
• I&O is still tasked with security and compliance and audit.
• I&O is often overloaded and unable to keep up with demand.

Manual repetitive work (MRW) sucks up time

• Manual repetitive work is a fact of life in I&O.
• DevOps circles refer to this type of work simply as “toil.”
• Toil is like treading water: it must be done, but it consumes precious energy and effort just to stay in the same place.
• Some amount of toil is inevitable, but it's important to measure and cap toil, so it does not end up overwhelming your team's whole capacity for engineering work.

Info-Tech Insight

Follow our methodology to focus IT automation on reducing toil.

Manual hand-offs create costly delays

• Every time there is a hand-off, we lose efficiency and productivity.
• In addition to the cost of performing manual work itself, we must also consider the impact of lost productivity caused by the delay of waiting for that work to be performed.

Every queue is a tire fire

Queues create waste and are extremely damaging. Like a tire fire, once you get started, they’re almost impossible to stamp out!

Increase queues if you want

• “More overhead”
• “Lower quality”
• “More variability”
• “Less motivation”
• “Longer cycle time”
• “Increased risk”

(Source: Edwards, citing Donald G. Reinersten: The Principles of Product Development Flow: Second Generation Lean Product Development )

Increasing complexity makes I&O’s job harder

Every additional layer of complexity multiplies points of failure. Beyond a certain level of complexity, troubleshooting can become a nightmare.

Today, Operations is responsible for the outcomes of a full stack of a very complex, software-defined, API-enabled system running on infrastructure they may or may not own.
– Edwards

Growing technical debt means an ever-rising workload

• Enterprises naturally accumulate technical debt.
• All technology requires care and feeding.
• I&O cannot control how much technology it’s expected to support.
• I&O faces a larger and larger workload as technical debt accumulates.

The systems built under each new technology paradigm never fully replace the systems built under the old paradigms. It’s not uncommon for an enterprise to have an accumulation of systems built over 10-15 years and have no budget, risk appetite, or even a viable path to replace them all. With each shift, who bares [SIC] the brunt of the responsibility for making sure the old and the new hang together? Operations, of course. With each new advance, Operations juggles more complexity and more layers of legacy technologies than ever before.
– Edwards

Most IT shops can’t have a dedicated engineering team

• In most organizations, the team that builds things is best equipped to support them.
• Often the knowledge to design systems and the knowledge to run those systems naturally co-exists in the same personnel resources.
• When your I&O team is trying to do engineering work, they can end up frequently interrupted to perform operational tasks.

Personnel resources in most IT organizations overlap heavily between “build” and “run.”

IT operations must become an engineering practice

• Usually you can’t double your staff or double their hours.
• IT professionals must become engineers.
• We do this by automating manual repetitive work and reducing toil.

Build your Sys Admin an Iron Man suit

Some CIOs see a Sys Admin and want to replace them with a Roomba. I see a Sys Admin and want to build them an Iron Man suit.
– Deepak Giridharagopal, CTO, Puppet

Use automation to reduce risk

Consistency

When we automate, we can make sure we do something the same way every time and produce a consistent result.

Auditing and Compliance

We can design an automated execution that will ship logs that provide the context of the action for a detailed audit trail.

Change

• Enterprise environments are continually changing.
• When context changes, so does the procedure.
• You can update your docs all you want, but you can't make people read them before executing a procedure.
• When you update the procedure itself, you can make sure it’s executed properly.

Follow Info-Tech’s approach: Start small and snowball

• It’s difficult for I&O to get the staffing resources it needs for engineering work.
• Rather than trying to get buy-in for resources using a “top down” approach, Info-Tech recommends that I&O score some quick wins to build momentum.
• Show success while giving your team the opportunity to build their engineering chops.

Because the C-suite relies on upwards communication — often filtered and sanitized by the time it reaches them — executives don’t see the bottlenecks and broken processes that are stalling progress.
– Andi Mann

Info-Tech’s methodology employs a targeted approach

• You aren’t going to automate IT operations end-to-end overnight.
• In fact, such a large undertaking might be more effort than it’s worth.
• Info-Tech’s methodology employs a targeted approach to identify which candidates will score some quick wins.
• We’ll demonstrate success, gain momentum, and then iterate for continual improvement.

Invest in automation to reap long-term rewards

• All too often people think of automation like a vacuum cleaner you can buy once and then forget.
• The reality is you need to perform care and feeding for automation like for any other process or program.
• To reap the greatest rewards you must continually invest in automation – and invest wisely.

To get the full ROI on your automation, you need to treat it like an employee. When you hire an employee, you invest in that person. You spend time and resources training and nurturing new employees so they can reach their full potential. The investment in a new employee is no different than your investment in automation.– Edwards

Measure the success of your automation program

1 15 FTEs x 80k/yr.; 20% of time on MRW, reduced by 20%
2 15 FTEs x 80k/yr.; 30% project capacity, increased by 30%
3 25k/hr. of downtime.; 5 hours per year of downtime caused by errors

Automating failover for disaster recovery

CASE STUDY

Industry Financial Services
Source Interview

Challenge

An IT infrastructure manager had established DR failover procedures, but these required a lot of manual work to execute. His team lacked the expertise to build automation for the failover.

Solution

The manager hired consultants to build scripts that would execute portions of the failover and pause at certain points to report on outcomes and ask the human operator whether to proceed with the next step.

Results

The infrastructure team reduced their achievable RTOs as follows:
Tier 1: 2.5h → 0.5h
Tier 2: 4h → 1.5h
Tier 3: 8h → 2.5h
And now, anyone on the team could execute the entire failover!

Info-Tech offers various levels of support to best suit your needs

DIY Toolkit

“Our team has already made this critical project a priority, and we have the time and capability, but some guidance along the way would be helpful.”

Guided Implementation

“Our team knows that we need to fix a process, but we need assistance to determine where to focus. Some check-ins along the way would help keep us on track.”

Workshop

“We need to hit the ground running and get this project kicked off immediately. Our team has the ability to take this over once we get a framework and strategy in place.”

Consulting

“Our team does not have the time or the knowledge to take this project on. We need assistance through the entirety of this project.”

Diagnostics and consistent frameworks used throughout all four options

Reduce Manual Repetitive Work With IT Automation – project overview

Reduce Manual Repetitive Work With IT Automation – Workshop overview

Contact your account representative or email Workshops@InfoTech.com for more information.

PHASE 1

Identify Automation Candidates

Reduce Manual Repetitive Work With IT Automation

Info-Tech’s approach to automation combines quick wins with a long-term process to manage your program

Structure automation around business value

• Just because a task can be automated, doesn’t mean it should be.
• Automation can require a large investment.
• Use a defensible, consistent method to demonstrate the value of the automation you intend to implement.

Direct automation toward your pain points

• What problems can we solve with automation?
• Look at where you’re currently experiencing pain and dig in to identify manual repetitive work that can be automated.

Consider the most common waste-creating scenarios

• Frequent interruptions
• Requires another person for access
• Unnecessary hand-offs
• Performing the task is very time-consuming
• Introduction of errors or mistakes in performing the task requires more time to fix or troubleshoot
• Performing the task correctly requires a time-sucking study of documentation

(Source: Edwards)

Identify pain points involving manual repetitive work (MRW)

1.1 30 minutes

1. Distribute sticky notes to each participant.
2. Have the participants brainstorm pain points for the IT group and write them down.
3. Post the sticky notes on a board and combine any duplicates together.
4. Group the pain points into two groups, each on one side of the board: those that involve manual repetitive work and those that don’t.
5. List the pain points involving manual repetitive work in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• Brainstorming

OUTPUT

• List of pain points involving MRW

Materials

• Markers
• Sticky notes
• Whiteboard

Participants

• IT staff

Example of pain points

(Image source: Info-Tech)

Automating reboots for a sketchy server – part 1

CASE STUDY

Industry: Oil & Gas
Source: Interview

Challenge

A webserver that collected data from hydrogen field stations kept locking up. The version of IIS would not clear its own cache, so it ran out of memory and had to be rebooted.

Solution

A member of the infrastructure team scheduled a planned outage at the same time each month and wrote a script to reboot the server at that time.

Results

IT operations no longer had to deal with incidents from this server locking up or perform any manual reboots.

Drill down MRW pain points into tasks

1.2 1 hour

1. For each MRW pain point you’ve identified, break that pain point down into particular tasks – the pieces of work that need to be accomplished.
2. Write the tasks on the whiteboard underneath their corresponding pain points.
3. Note whether any pain points share certain tasks (if indeed the tasks are the same), and if they do, combine these tasks.
4. Document the tasks in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• Pain points involving MRW

OUTPUT

• Tasks involving MRW

Materials

• Sticky notes
• Whiteboard
• Whiteboard Markers

Participants

• IT staff

Drill down into tasks - example

(Image Source: Info-Tech)

Estimate the MRW involved in each task

1.3 1 hour

1. If necessary, cull the tasks down to a more manageable sub-set for the first wave of evaluation (you can always go back to the others later), e.g. 10-20 tasks.
2. For each task, estimate the amount of MRW involved:
• Number of times task is performed
• Personnel time required to perform the task
3. If relevant, estimate the average amount of queue time and wait time required for the task.
4. If relevant, estimate the error rate for the task and the impact of the average error (e.g. downtime and/or rework).
5. Document the current state in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• MRW tasks

OUTPUT

• Estimated MRW for each task

Materials

• Sticky notes
• Whiteboard
• Whiteboard Markers

Participants

• IT staff

Estimate MRW – example

(Image source: Info-Tech)

Automate with a sniper rifle, not a shotgun

Some manual tasks may have a good reason for being done manually. Automation is very expensive to implement, to support, and to maintain. If you're not being strategic, the value may not outweigh the cost. Take a keen look. Automation can actually make things worse.
– Andrew Kum-Seun
Info-Tech Senior Research Analyst

Focus automation on cost, speed, and consistency

Benefits of Automation

Cost

Do it with less

Speed

Do it faster

Consistency

Do it the same way every time

Consider which benefits are best for each task

The importance and impact of the automation benefit will vary depending on the particular task. Identify which matters the most for each task.

Cost

It requires fewer resources to perform an action.

Speed

The action is accomplished more quickly (i.e. we must wait less time for the action to be complete).

Consistency

No more errors or inconsistencies.

Info-Tech Insight

Note the distinction between cost and speed. When you finish something using less manual work, it will naturally be completed more quickly. The difference between cost and speed is that cost is about how much resources are required, whereas speed is about how long we’re waiting for the task to be completed. A task might take a lot of manual repetitive work, but if it is normally scheduled in advance with enough lead time, there won’t be a speed benefit to automating it.

Consider the cost side of the automation equation

• Some automations might be of great benefit but could be infeasible for a number of reasons.
• Think about feasibility in three dimensions:

1. Effort/Expenditure

How much cash must we spend?

How much work must our team do?

2. Time

How long until we see value?

3. Maintainability

How much care and feeding will be required?

How much overhead will it add to our workload?

Approach databases with caution

Databases are different: Automating actions relating to release, configuration, and provisioning of databases introduces challenges not faced with similar actions on applications and infrastructure components.

The DBMS matters: Complexity of the changes required, the ability to rollback changes, and the integrity of the commands themselves will vary depending on the particular database management system (DBMS).

Synchronization: Database automation must account for keeping test databases synchronized and available for application program testing.

Failure hurts: It can be challenging to recover from database changes that have been improperly specified or to back off a migration and roll back the database.

Tools matter: In most cases, specialized database change management tools will be required.

Rank the tasks based on value and ease of automation

1.4 1 hour

1. Draw a 2x2 grid on a whiteboard. Use the y axis as value for automating the task and the x axis as ease of automating the task.
2. Have participants plot the tasks on the 2x2 grid based on their judgement of the value and the ease.
• Value
• Speed
• Cost
• Consistency
• Ease
• Effort/Expenditure
• Time
• Maintainability
3. Discuss each task in relation to the criteria.
4. Document the outcome in the IT Automation Presentation.
   Give numerical ratings in the IT Automation Worksheet.

INPUT

• MRW tasks

OUTPUT

• Graph ranking tasks for automation

Materials

• Sticky notes
• Whiteboard
• Whiteboard Markers

Participants

• IT staff

Rank the tasks – example

Select top candidates and define metrics to measure success

1.5 1.5 hours

1. Based on the available resources of your team, select the top candidates for automation (in most cases, Info-Tech recommends starting with one to three tasks and rolling those out before re-evaluating and iterating).
2. For each task you’ve selected, define metrics to measure success for automation. Pick the metrics most relevant to how automating the task provides value (cost, speed, consistency).
3. Estimate the current state metrics for each task (if possible) or devise a way to measure that later. Define a target state after implementing automation.
4. Document the top candidates and the delta from current state to future state in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• Graph of MRW tasks

OUTPUT

• Shortlist of tasks to automate
• Metrics for success

Materials

• Sticky notes
• Whiteboard
• Whiteboard Markers

Participants

• IT staff

Candidates and success metrics – example

(Image source: Info-Tech)

Leverage a lightweight project charter to communicate

Draft a lightweight project charter for each task automation

1.6 1 hour

1. Using Info-Tech’s template in the IT Automation Presentation, write a one-pager project charter for each automation initiative.
2. The project charter should include the following elements (if relevant):
• Abstract
• Value to the business
• Business alignment
• Cost
• Opportunity risk
• Operational risk
• Enabled capabilities
3. Peer review the project charter to refine it and correct any errors or omissions.
4. Document the one-pager for each task automation in the IT Automation Presentation.

INPUT

• Task automation shortlist
• Automation success metrics
• One-pager template

OUTPUT

• Lightweight project charter for each task automation

Materials

• Desktop or laptop workstation

Participants

• IT staff

If you want additional support, have our analysts guide you through this phase as part of an Info-Tech workshop

Book a workshop with our Info-Tech analysts:

• To accelerate this project, engage your IT team in an Info-Tech workshop with an Info-Tech analyst team.
• Info-Tech analysts will join you and your team onsite at your location or welcome you to Info-Tech’s historic Toronto office to participate in an innovative onsite workshop.
• Contact your account manager (www.infotech.com/account), or email Workshops@InfoTech.com for more information.

The following are sample activities that will be conducted by Info-Tech analysts with your team:

1.4		Rank the tasks based on value and ease of automation Use a rating system to find out which benefit provides the most value for the required cost.
1.5		Select top candidates and define metrics to measure success Pick the top automation candidates and plan how you’ll measure the success of those initiatives.

PHASE 2

Map & Optimize Process Flows

Reduce Manual Repetitive Work With IT Automation

Info-Tech’s approach to automation combines quick wins with a long-term process to manage your program

Document the current process before automating

• Before automating, we need to understand exactly what it is that we want to automate.
• When we perform a procedure manually, we naturally “chunk” our understanding of that procedure in a way that bundles together discreet steps and/or omits certain decision points.
• Mapping out the current process visually allows us to clearly define and understand the current state of the task we are considering to automate.

Map process flows for the top tasks

2.1 2 hours

1. For each of the top task candidates, map out the process flow of the actions that must be taken to accomplish that task.
2. Use sticky notes for each action so you can re-arrange them on the wall as you modify the process flow.
3. Focus on capturing the current-state process flow that reflects the way your team members complete that process now (we will review the flows in the next activity).

INPUT

• Task automation shortlist

OUTPUT

• Process flow(s)

Materials

• Sticky notes
• Markers

Participants

• IT staff

Map process flow – example

Building Test Server Workflow

Image source: Info-Tech

Optimize the process before automating

• Just because we’re doing something a certain way now, does not mean that that’s the best way.
• We need to take a good hard look at what the execution of the task currently looks like, to make that that’s actually the best way to build out the requirements for execution.
• If we automate a process before optimizing it, we can enshrine waste or inefficiencies – for example, unnecessary hand-offs or approvals that interrupt the workflow.

Review and optimize process flows

2.2 2 hours

1. Review the process flows with the goal of optimizing the process to reduce waste.
2. The following are examples of waste:
3. Unnecessary hand-offs
• Multiple back-and-forths seeking more information or requirements
• Queues and delays
• Approvals that interrupt the work
4. Where possible, modify the process flow to optimize it (e.g. through changes such as moving any approvals or requirements up front).
5. After optimizing, review the flow to ensure that it still achieves the proper end result.

INPUT

• Draft process flow(s)

OUTPUT

• Optimized flow(s)

Materials

• Sticky notes
• Whiteboard
• Whiteboard markers

Participants

• IT staff

Include end-to-end business logic to make automation a success

• The optimized process might capture how to perform the task, but does it have all the proper logical traps and exit points that are necessary for something we’re to automate completely?
• To automate the task, we need the full end-to-end business logic to be included, which means any traps or exit points necessary must be built into the flow.
• At the beginning of the flow, we should verify the prerequisites, and at the end of the flow, we should validate the expected results.

The most important things – and the things that people miss – are prerequisites and expected results. People jump out and build scripts, then the scripts go into the ditch, and they end up debugging in production.
– Darin Stahl, Info-Tech, Research Director, Infrastructure & Operations

Automation must change with the times – part 2

CASE STUDY

Industry: Oil & Gas
Source: Interview

Challenge

The automated reboots of the web server were based on the time of the month. When the production cycle changed, the memory would be exhausted at a different time.

Solution

After performing further root-cause analysis, the operations team wrote a PowerShell script to monitor the cache size and take action to correct memory issues.

Results

Now the server reboots would stay in sync with the strain on memory issues. Operators would no longer suffer outages while waiting for Microsoft to fix the IIS product.

Clarify process flow business logic, and finalize the future-state process flows

2.3 2 hours

1. Review each process flow to clarify the business logic.
2. Be sure to include any necessary traps or exits to the process flow (watch out for any unspoken or implicit decision points normally taken care of by a human operator).
3. This flow will be the requirements for what you’ll actually automate; make sure that these are precisely the actions you want taken every time that workflow is kicked off.

INPUT

• Optimized process flow(s)

OUTPUT

• Optimized flow(s) with complete business logic

Materials

• Sticky notes
• Markers

Participants

• IT staff

Clarify process flow business logic – example

If you want additional support, have our analysts guide you through this phase as part of an Info-Tech workshop

Book a workshop with our Info-Tech analysts:

• To accelerate this project, engage your IT team in an Info-Tech workshop with an Info-Tech analyst team.
• Info-Tech analysts will join you and your team onsite at your location or welcome you to Info-Tech’s historic Toronto office to participate in an innovative onsite workshop.
• Contact your account manager (www.infotech.com/account), or email Workshops@InfoTech.com for more information.

The following are sample activities that will be conducted by Info-Tech analysts with your team:

2.1		Map process flows for the top tasks Visually document the process flow for performing the top automation candidates.
2.2		Review & optimize process flows Pick the top automation candidates, and plan how you’ll measure the success of those initiatives.

PHASE 3

Build a Process for Managing Automation

Reduce Manual Repetitive Work With IT Automation

Info-Tech’s approach to automation combines quick wins with a long-term process to manage your program

Adopt an engineering mindset

• With automation, the focus changes from doing the work to building a system that does the work.
• This shift requires a change in mindset.

[I]f automation is an afterthought, IT is left with even more complexity and no solution.
– ActiveBatch, Shift Left! A Look at Automation As Prologue, Not Postscript.

"This significantly changes the role of operations from order-takers to owners and operators of a service-providing organization. Instead of treating each service request as a one-off, operations teams develop and offer a menu of standardized services aligned with business objectives."
– Mann

Leverage SDLC best practices

• With widespread automation, I&O professionals have become developers whether they like it or not.
• We don’t need to build our own process from scratch – we can adopt best practices from the development world for our own purposes.

"Since everything above the hardware is software, why not treat it as such and use all of the established software management best practices that we can. These include using versioned source control, having an automated “build” process, using a [well-defined] promotion process to move code from one environment to the next (hopefully as immutable artifacts), and regular code reviews."
– Edwards

Plan for testing up front

• Before we begin to develop each automation, we should already have a testing plan in mind.
• We don’t have to automate the testing, but we do have to make sure that the automations work as intended before we roll them out.

"Many people think infrastructure testing should be fully automated, relying on continuous integration and an infrastructure-as-code approach. While automation is more reliable and generally faster, keep in mind that it’s the validating that matters, and that you can test infrastructure changes manually.
Why are we pointing this out? Because infrastructure changes can vary widely, and while some lend themselves to automation with a reasonable amount of effort, other changes are just too infrequent or expensive to validate in an automated fashion. So don’t get too locked into the method — just make sure that you validate infrastructure changes prior to a production deployment." – Mann

Kick off your test plan for each automation

3.1 1 hour

1. Draw a chart on a whiteboard with categories of items that could affect the proper functioning of the automation (focus on only a single task automation at a time).
2. List each category at the top of a column.
3. Include the following categories in the chart:
• Inputs/data
• Activity/behavior
• Dependencies
• Attacks
4. Under each category, list potential things that could go wrong that might break the automation. Think about the expected ideal case and then how real-world use cases may differ (e.g. unexpected data range)
5. Document this chart in the IT Automation Worksheet and IT Automation Presentation.
6. Add an item to your roadmap to leverage this chart in building a detailed test plan for each task automation.

INPUT

• Finalized task process flows

OUTPUT

• List of potential issues to inform test plan

Materials

• Whiteboard
• Whiteboard markers

Participants

• IT staff

Test plan – example

Build Test Server

Image source: Info-Tech

Define a lightweight SDLC to make automation a success

• It may be tempting to handle automation on an ad hoc basis, but defining a lightweight SDLC will lead to better results.
• Your automation SDLC will ensure that you follow a consistent process for determining which automations to build, how to test those automations, and how to deploy and maintain them.
• The SDLC will aim to support higher quality and ensure that the work takes place within an automation program that delivers consistent business value in line with the metrics you have defined.

Define process for automation rollout

3.2 45 minutes

1. Review Info-Tech’s automation rollout process workflow (see the following slide).
2. Make any modifications you deem necessary for your organization (or simply use our workflow out of the box).
3. On a whiteboard, write down each stage of the rollout process at the head of a column.
4. For each piece of the workflow, define who will carry out the work (those responsible) and who is accountable for ensuring appropriate outcomes of each step. List the accountable person underneath that workflow and the responsible roles or people who will do the work beneath it (note that these could be the same for each automation process or could differ depending on the process).
5. Document the process flow and the personnel responsibilities in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• Automation process rollout template

OUTPUT

• Automation rollout process
• Associated roles & responsibilities

Materials

• Whiteboard
• Whiteboard Markers

Participants

• IT staff

Lightweight automation SDLC

Identify Process →Document Process→Determine Metrics →Optimize Process
Define Requirements →Document Dependencies →Develop Testing Plan →Build
Test →Change Management →Update Documentation →Deploy
Evaluate and Measure →Update Documentation →Maintain →Retire

Automation SDLC RACI – example

Define how to manage your automation program

• Automation isn’t a one-and-done. You should continually improve automation as you iterate.
• We’ve defined the SDLC for automation, but we also must define how we’re going to manage the automation program as a whole.
• We need to put a process in place to ensure that our investments in automation are delivering their expected return and the program as a whole continues to align with and drive business value over the long term.

Define process to manage your automation program

3.3 30 minutes

1. Define the process by which your automation program will be managed.
2. Using Info-Tech’s template, complete a RACI for the process that will manage the automation program.
3. Complete the template that will determine what will be involved in reviewing the program and how frequently this review will take place.
4. Document the RACI and program review process in the IT Automation Worksheet and IT Automation Presentation Template.

INPUT

• RACI template
• Program review template

OUTPUT

• Automation program governance process

Materials

• Desktop or laptop workstation

Participants

• IT staff

Structure your program around key metrics

Consider more metrics based on group

Define metrics to measure success of your automation program

3.4 1 hour

1. Use Info-Tech’s template to populate the chart.
2. Define metrics to measure the success of the automation program. Describe the high-level metric, the specific number or figure that will be measured, the source for the data, the report on which the metric will be reported, and the review cycle.
3. Not all metrics are equal. Do a forced ranking of the metrics to determine which are the most important.
4. Document the results in the IT Automation Worksheet and IT Automation Presentation Template.

INPUT

• Metrics library
• Metrics template

OUTPUT

• KPIs for automation program

Materials

• Desktop or laptop workstation

Participants

• IT staff

Automation program metrics – example

Automation program metrics – example 2

Image source: Info-Tech

If you want additional support, have our analysts guide you through this phase as part of an Info-Tech workshop

Book a workshop with our Info-Tech analysts:

• To accelerate this project, engage your IT team in an Info-Tech workshop with an Info-Tech analyst team.
• Info-Tech analysts will join you and your team onsite at your location or welcome you to Info-Tech’s historic Toronto office to participate in an innovative onsite workshop.
• Contact your account manager (www.infotech.com/account), or email Workshops@InfoTech.com for more information.

The following are sample activities that will be conducted by Info-Tech analysts with your team:

3.2		Define process for automation rollout Leverage a lightweight SDLC to make automation a success.
1.5		Define process to manage your automation program Build program management into your way of working to drive long-term value through automation.

PHASE 4

Build Automation Roadmap

Reduce Manual Repetitive Work With IT Automation

Info-Tech’s approach to automation combines quick wins with a long-term process to manage your program

Automation is the gateway to self-service

Consider end-to-end self-service

The quick wins that automation has enabled for your team are extremely valuable and are an important step in your journey – but they’re not the destination!

[I]t’s absolutely possible to deliver real value incrementally before you reach the point of a comprehensive self-service catalog. Teams should build self-service systems for themselves and then their adjacent teams, next expanding outwards through the organization. This is exactly what the data shows successful teams do.– Mann

Self-service is harder than it looks

• Automating our own tasks is the first step toward a move to enabling self-service for other groups.
• Simply saying “self-service” won’t make it so.
• Self-service will ideally be slick and simple for the end user, but that’s just the tip of the iceberg.
• Users have no visibility into all the work that goes into the back end that makes self-service possible.

Consider the three elements for each piece of automation

In enabling self-service for automated procedures, we need to consider the following elements.

Define

“Who creates the definition of the automated procedure and how do they do it?”

Execute

“Who can execute the automated procedure and how are they provided the ability to do it?”

Govern

“Who has governance over the procedure and how do they manage security, management oversight, and compliance?”

Source: Edwards

Put system configurations in version control

• Version control is the practice of archiving previous versions of an artifact and tracking the changes from one version to the next.
• Each version of a given artifact has a unique version number.
• Version control allows the ability to revert back to an earlier version in the event that the latest version is broken.

Storing system configurations in version control is a vast improvement over scripts living on people’s workstations, providing a number of advantages:

• You can see changes over time, see how they evolved, and know who made them.
• Anyone with access to the version control system can audit changes.
• You get automatic backups of key configuration files.

Leverage version control for infrastructure

• Each piece of automation we’re building becomes part of a control layer for our infrastructure.
• We should not only use version control for the automations themselves, but also for the infrastructure changes we’re making.
• Version control for infrastructure will allow us to test changes before moving to production and to roll back if things break.

The use of version control for all production artifacts is highly correlated with IT performance. It’s the first step to continuous delivery of your infrastructure code. Use of version control makes it easy to recreate environments for testing and troubleshooting, boosting throughput for both Dev and Ops. It also reduces the time to recover if an error is identified in production. You can quickly either redeploy the last good state, or fix the problem and roll forward, all with history and auditing capabilities. – Mann

Consider automation tools as a next step

• Tools can be great enablers of your automation program.
• Now that you understand some of your use cases and priorities, you’ll be in a better position to make an informed decision about automation tools.
• Add an item to your roadmap to look into automation tools and solutions. When selecting a tool, consider elements such as the following:
• Usability
• Technical requirements
• Integrations
• Vendor stability
• Availability of support
• Support for best practices such as SDLC and version control

Consider standardizing your technology stack

• Building out a program for automation and self-service is much easier if you standardize your technology stack, for example, by moving toward the following:
• Deploy on a single OS (or OS family)
• Use a single set of key value stores
• Standardization of the technology stack allows you to re-use automations from one area in another area.
• Put an item on your roadmap to evaluate the potential of standardizing your technology stack and reducing variability.

Consider a portal that integrates with other tools and services

• The holy grail is to have a single portal, or "self-service operations hub," from which users can access the services that they need.
• When enabling self-service in this way, keep in mind that users may not have the technical expertise necessary to be comfortable working off multiple screens at once to gain the context they require.
• Building as much integration as possible into the hub helps give users the visibility they need to perform self-service effectively.
• Add an item to your roadmap to plan for a program focused on enabling self-service for other groups.

The more integrations you can create, the more context your users will have without leaving the Self-Service Operations hub.– Damon Edwards

Build a roadmap for next steps

4.1 45 minutes

1. Use Info-Tech’s IT Automation Roadmap.
2. Assign an owner (person accountable) and timeline for beginning each initiative on the roadmap.
3. Add or remove initiatives as desired.
4. Document the highest-impact future planned initiatives, along with the initiative owner, in the IT Automation Worksheet and IT Automation Presentation.

INPUT

• Metrics library
• Metrics template

OUTPUT

• KPIs for automation program

Materials

• Desktop or laptop workstation

Participants

• IT staff

Follow program management process for continual improvement

• Execute on your automation roadmap to enhance your capabilities – but realize that this isn’t the end.
• Build the automation program management process into your way of working and continue to iterate to improve.

If you want additional support, have our analysts guide you through this phase as part of an Info-Tech workshop

Book a workshop with our Info-Tech analysts:

• To accelerate this project, engage your IT team in an Info-Tech workshop with an Info-Tech analyst team.
• Info-Tech analysts will join you and your team onsite at your location or welcome you to Info-Tech’s historic Toronto office to participate in an innovative onsite workshop.
• Contact your account manager (www.infotech.com/account), or email Workshops@InfoTech.com for more information.

The following are sample activities that will be conducted by Info-Tech analysts with your team:

4.1		Rank the tasks based on value and ease of automation Use a rating system to find out which benefit provides the most value for the required cost.
1.5		Build a roadmap for next steps Lay out a timeline of high-impact initiatives that will enhance your automation capabilities.

Insight breakdown

Optimize before you automate.

Just because you currently do something in a given way, doesn't mean it’s the best way. Map the process end-to-end and take a good hard look before you start building out automation.

Foster an engineering mindset.

Your team members may not be process engineers, but they should learn to think like one. Designing efficient systems to deliver value should always be the focus of automation, and this means directing your efforts to where they will have the greatest impact.

Build a process to iterate.

Effective automation can't be a one-and-done. Define a lightweight process to manage your program. You should curate and manage a living, breathing set of solutions and refine things over time.

Summary of accomplishment

• How to prioritize tasks for automation
• How to define metrics to measure the success of automation projects
• How to define a program to measure the success of IT automation on an ongoing basis

Processes Optimized

• IT automation intake and prioritization
• IT automation SDLC
• IT automation program management

Deliverables Completed

• IT Automation Worksheet
• IT Automation Presentation
• IT Automation Roadmap

Summary of steps for IT automation

1. Identify MRW tasks.
2. Rank automation candidates.
3. Pick top candidates.
4. Define metrics for success.
5. Build end-to-end process flows.
6. Define automation SDLC.
7. Build process for managing automation program.
8. Define program success metrics.
9. Build automation roadmap.

Info-Tech Insight

This project has the ability to fit the following formats:

• Onsite workshop by Info-Tech Research Group consulting analysts.
• Do-it-yourself with your team.
• Remote delivery (Info-Tech Guided Implementation).

Research contributors and experts

Steven Berkovitz, Chief Platform Officer, Tecsys Inc.		Steven is a proven technology leader with hands-on experience building great software companies. He co-founded OrderDynamics and successfully scaled from 2 to 500 employees across nine countries and saw the company through acquisition, subsequent funding, reorganization, and other M&A activities.
Deepak Giridharagopal, CTO, Puppet		Deepak has delivered large-scale solutions using a diverse set of techniques and tools. His responsibilities have ranged from small, self-contained projects to large, company-wide initiatives and roadmaps. He thinks that building stuff is good, but building the right stuff is more interesting and challenging.
Peter Madison, Founder, Xodiac Inc.		Peter creates and implements strategies to reach business objectives and make DevOps and Agile practices a reality within complex organizations. His focus is always on improving how businesses operate by enhancing their processes and applying modern practices and technology.
Nabeel Yousif, Executive Consultant		Nabeel is a strategic thinker and an award-winning authentic leader. He couples his passion for what he does with rigor to ensure that his commitment to drive value contributes to the bottom line.
Sergio Zanardo, VP Information Technology		Sergio is an IT executive with proven leadership abilities in technologically enabling explosive growth endeavors and transformations, including acquisitions due diligence and integration, turnarounds, and startups.

Related Info-Tech research

Develop an IT Infrastructure Services Playbook

Automation, SDI, and DevOps – build a cheat sheet to manage a changing Infrastructure & Operations environment.

Map Technical Skills for a Changing Infrastructure & Operations Organization

Be practical and proactive – identify needed technical skills for your future state environment and the most efficient way to acquire them.

Bibliography

Automation 101. ConnectWise. Web.
Cagen, Dan, et al. DevOps and Infrastructure Have a Symbiotic Relationship. TechTarget, Aug. 2017. Web.
Caum, Carl. “What is Infrastructure as Code?” Puppet, 10 Feb. 2017. Web.
Clavell, James. Shogun. Dell, 1975. Print.
Continuous Automation for the Continuous Enterprise. Chef, Jan. 2018. Web.
“DevOps: A Survival Guide for Infrastructure Teams.” WWT Blog. Web.
Edwards, Damon, et al. Self-Service Operations: A Guide to Fewer Interruptions, Less Waiting, and Getting More Done. Ver. 2.0. Rundeck. Web.
Gorelik, Stacy. “Nuts and Bolts of Building a Platform Team.” Datadog. YouTube. Web.
Hering, Mirco. “How to Define Your DevOps Roadmap.” DZone, 25 Jan. 2016. Web.
“How to Build a DevOps Roadmap to Kickstart Your Digital Transformation Journey.”
Kishore, Nanda R. Getting IT Infrastructure Right by Implementing a DevOps Framework. Tata Consultancy Services. Web.
Laird, Cameron. “DevOps Nuts and Bolts.” ITPro Today, 9 Dec. 2016. Web.
Laliberte, Bob. Five Steps to Data Center Network Agility. Enterprise Strategy Group, July 2018. Web.
Lederman, Sol. Containers 101. Linux Journal, 2019. Web.
Modi, Ritesh. “What is DevOps?” TheNewStack, 17 May 2017. Web.
Mann, Andi, et al. 2018 State of DevOps Report. Puppet. Web.
Markx, Gino. “Assessing Maturity of a Technology Department.” Xodiac, 4 Apr. 2019. Web.
Mathon, John. “Building a Better DevOps-enabled Cloud.” Jaxenter, 19 Mar. 2018. Web.
Null, Christopher. “Infrastructure As Code: The Engine at the Heart of DevOps.” TechBeacon. Web.
Reddy, Shravanthi. “Microservices and Containers Together Enable DevOps.” APIfriends, 31 July 2018. Web.
Shift Left! A Look at Automation As Prologue, Not Postscript. ActiveBatch. Web.
Vitus, Andy. “Get to Know the New DevOps Stack.” DevOps.com, 6 Oct. 2016. Web.
Workload Automation: How IT Can Meet Evolving Business Needs. ActiveBatch. Web.