# The Automation Index

> A sourced database of 50 common business processes (operations, appointment setting, follow-up, sales admin, onboarding, reporting, invoicing and CRM hygiene). For each: a direct answer to "can it be automated?", a task-by-task share automatable today, hours saved per week with assumptions, tools (n8n, Make, Zapier, GoHighLevel, CRMs, AI agents) and a workflow outline. Median process: 69% automatable. Updated 2026-11-09.

Published by CXJ Labs (https://cxjlabs.io), which builds business automations. It is not independent; ownership is disclosed on every page. CXJ Labs is your operating partner. We automate everything else in the business, so you and your team can focus on what makes you money.

Estimates are CXJ Labs' editorial judgements anchored on McKinsey Global Institute activity-level automation potential and O*NET task statements. Data is CC BY 4.0; please cite the entry URL.

## Key pages

- [Methodology and disclosure](https://theautomationindex.com/methodology/): how shares and hours are estimated, limits, who publishes the Index
- [Hours admin is costing you calculator](https://theautomationindex.com/calculator/): team size x admin hours x hourly cost
- [Sources](https://theautomationindex.com/sources/): every cited study with caveats
- [Dataset JSON](https://theautomationindex.com/data/automation-index.json) and [CSV](https://theautomationindex.com/data/automation-index.csv)
- [Free automation game plan from CXJ Labs](https://cxjlabs.io/gameplan)

## Operations

Operations work is the glue between teams: inboxes, requests, handovers, documents and the copying of data from one tool into another. It is the most automatable category in the Index because most of it is data collection and data processing, the two activity types McKinsey rates highest after predictable physical work.

- [Can meeting notes and action items be automated?](https://theautomationindex.com/automate/meeting-notes-and-action-items/): 75% automatable; about 7.5 of 10.5 hours a week back. Mostly. Recording, transcribing, summarising and turning decisions into tasks are largely automatable today with AI note-takers. In our model about 75% of the time goes to software. What stays human is checking that each action has a real owner and date, and chasing the people who said yes and then went quiet.
- [Can shared inbox triage be automated?](https://theautomationindex.com/automate/shared-inbox-triage/): 76% automatable; about 10 of 13.5 hours a week back. Mostly. Sorting, routing, archiving noise and logging enquiries are highly automatable today, and AI now handles routine replies well. In our model about 75% of triage time can go to software. What stays human is spotting the angry or unusual email and answering anything where a wrong reply would cost money or trust.
- [Can internal request intake be automated?](https://theautomationindex.com/automate/internal-request-intake/): 70% automatable; about 6 of 9 hours a week back. Largely. Capturing requests, acknowledging them, sending status updates and answering 'where is my request?' are highly automatable. In our model about 70% of intake time can go to software. What stays human is deciding priority between competing requests and the weekly conversation about what will not get done this week.
- [Can SOP documentation be automated?](https://theautomationindex.com/automate/sop-documentation/): 59% automatable; about 5 of 8.5 hours a week back. Partly. AI can turn a recording of someone doing the job into a usable first draft, add screenshots and answer staff questions from the library. In our model about 60% of SOP time can go to software. What stays human is doing the process well on camera and checking the draft with the person who owns the job.
- [Can shift handover reports be automated?](https://theautomationindex.com/automate/shift-handover-reports/): 67% automatable; about 5.5 of 8 hours a week back. Largely. Pulling what happened from booking, ticketing and phone systems, writing the note and compiling the weekly log are highly automatable. In our model about 65% of handover time can go to software. What stays human is judging which open issues matter most and the incoming lead actually reading and owning them.
- [Can purchase orders be automated?](https://theautomationindex.com/automate/purchase-orders/): 69% automatable; about 7 of 10 hours a week back. Largely. Raising purchase orders, sending them, routing approvals and matching deliveries and invoices to them are highly automatable. In our model about 70% of PO time can go to software. What stays human is approving larger spends, chasing late suppliers by phone and sorting out deliveries that do not match the order.
- [Can data entry between systems be automated?](https://theautomationindex.com/automate/data-entry-between-systems/): 80% automatable; about 8.5 of 10.5 hours a week back. Mostly. Copying records from one tool into another, updating spreadsheets and re-keying data from PDFs are among the most automatable work there is. In our model about 80% of this time can go to software. What stays human is deciding which system is right when records disagree, and fixing the mismatches that rules cannot.

## Appointment setting

Appointment setting covers everything between a lead raising a hand and a call happening: first response, qualification, booking, reminders and rescheduling. The admin is highly automatable. The persuasion is not. The best systems answer in seconds, book without back-and-forth, and hand the conversation to a person the moment judgement is needed.

- [Can appointment reminders be automated?](https://theautomationindex.com/automate/appointment-reminders/): 82% automatable; about 7.5 of 9 hours a week back. Yes. Confirmations, timed reminders and CRM status updates are almost fully automatable today, and together they make up most of the work. In our model about 80% of reminder time can go to software. What stays human is calling the unconfirmed few and answering unusual replies, which is where most saved no-shows actually come from.
- [Can appointment booking be automated?](https://theautomationindex.com/automate/appointment-booking/): 82% automatable; about 7 of 8.5 hours a week back. Mostly. Finding a time, creating the calendar event, sending the invite and logging the booking are rules-based and almost fully automatable today. In our model about 80% of booking time can move to software. What stays human is the minority of requests that are urgent, complicated or need someone to decide who should take them.
- [Can rescheduling and cancellations be automated?](https://theautomationindex.com/automate/rescheduling-and-cancellations/): 72% automatable; about 3.5 of 5 hours a week back. Yes, largely. Offering new times, updating calendars and records, and filling freed slots from a waitlist are rules-based and automate well. In our model about 70% of the work can move to software. What stays human is trying to save a cancellation and applying the cancellation policy fairly when the circumstances are unusual.
- [Can missed-call follow-up be automated?](https://theautomationindex.com/automate/missed-call-text-back/): 63% automatable; about 3.5 of 6 hours a week back. Yes, mostly. Spotting a missed call, texting the caller within a minute, logging the enquiry and sorting replies by urgency are all automatable today. In our model about 60% of missed-call handling can move to software. The callback itself, especially for urgent or high-value jobs, should stay with a person who can quote and reassure.
- [Can lead qualification be automated?](https://theautomationindex.com/automate/lead-qualification/): 60% automatable; about 6.5 of 10.5 hours a week back. Partly. Collecting form answers, researching the company, scoring against clear criteria and logging the outcome automate well today. In our model about 60% of qualification time can move to software. The qualifying conversation for borderline leads stays human, because buying intent, authority and timing are rarely stated plainly on a form.
- [Can speed to lead be automated?](https://theautomationindex.com/automate/speed-to-lead/): 65% automatable; about 7 of 11 hours a week back. Mostly. Spotting new leads, sending a personal first message and routing the lead to an available setter can all happen within seconds, every time, including nights and weekends. In our model about 65% of first-response work is automatable. The first live conversation still matters most, so automation should get a person on the phone faster, not replace them.
- [Can social media DM appointment setting be automated?](https://theautomationindex.com/automate/social-dm-appointment-setting/): 44% automatable; about 9.5 of 21 hours a week back. Partly. First replies, booking links, logging and nudges to quiet conversations automate well. In our model about 45% of DM setting time can move to software. The qualifying conversation and handling objections stay human, because people messaging a personal brand expect a person, and trust built in the chat decides whether they show up.

## Follow-up

Follow-up is where most revenue quietly leaks: leads who went cold, no-shows nobody chased, quotes left unanswered, happy customers never asked for a review. Timing and persistence are easy to automate. What to say to a specific person after a specific conversation is partly automatable now that AI can draft from call notes.

- [Can lead follow-up be automated?](https://theautomationindex.com/automate/lead-follow-up/): 61% automatable; about 7 of 12 hours a week back. Mostly. Deciding who to chase, sending timed emails and texts, and stopping the sequence when someone replies are all automatable today. In our model about 60% of follow-up time can move to software. Calls and replies to real questions stay human, because a lead who answers wants a person, not the next message in a sequence.
- [Can no-show follow-up be automated?](https://theautomationindex.com/automate/no-show-follow-up/): 69% automatable; about 3 of 4.5 hours a week back. Yes, largely. Detecting the no-show, sending a 'sorry we missed you' message with a rebook link, and running a short rebooking sequence are all automatable. In our model about 70% of no-show work can move to software. The quick phone call straight after the missed slot stays human, and it often rebooks more people than the messages do.
- [Can post-call follow-up be automated?](https://theautomationindex.com/automate/post-call-follow-up/): 68% automatable; about 6 of 9 hours a week back. Mostly. Recap emails drafted from call transcripts, sending promised resources, updating the CRM and booking the next meeting can all be automated or AI-drafted. In our model about 65% of post-call time can move to software. Reviewing what the recap promises and deciding the deal's next move stay with the rep who was on the call.
- [Can quote follow-up be automated?](https://theautomationindex.com/automate/quote-follow-up/): 51% automatable; about 3.5 of 7 hours a week back. Partly. Tracking open quotes, sending timed reminders and alerting the team when a quote is opened are automatable today. In our model about 50% of quote follow-up time can move to software. The call on larger quotes and the negotiation that follows stay human, and they decide most of the outcome.
- [Can review requests be automated?](https://theautomationindex.com/automate/review-requests/): 72% automatable; about 4.5 of 6 hours a week back. Yes, mostly. Spotting finished jobs, sending the request with a direct link, one reminder and tracking review counts are all automatable. In our model about 70% of the work can move to software. Replying to reviews, especially negative ones, and dealing with unhappy customers stay human because they are public conversations about your reputation.
- [Can lead reactivation be automated?](https://theautomationindex.com/automate/lead-reactivation/): 63% automatable; about 5 of 8.5 hours a week back. Mostly. Segmenting old leads, cleaning the list, sending campaigns in safe batches and reporting results are automatable today. In our model about 60% of reactivation work can move to software. Writing the offer and handling the replies stay largely human, because reactivation works only when the message gives an old lead a genuinely new reason to talk.

## Sales admin

Sales admin is the non-selling work that, by Salesforce's own survey, takes about 70% of a sales rep's week: logging calls, updating stages, writing proposals, routing leads, chasing signatures and working out commission. Most of it is structured data work, so it automates well, provided the CRM is set up to receive it.

- [Can proposal writing be automated?](https://theautomationindex.com/automate/proposal-generation/): 71% automatable; about 7 of 10 hours a week back. Mostly. Pulling deal details, assembling scope and pricing, sending and tracking can all be handed to software, and AI can draft a first version of the tailored summary from call notes. In our model about 70% of proposal time is automatable. The rep still owns the diagnosis and any pricing that breaks the rules.
- [Can contracts and e-signatures be automated?](https://theautomationindex.com/automate/contract-generation-and-e-signature/): 82% automatable; about 5.5 of 7 hours a week back. Mostly. Filling a contract from deal data, sending it for signature, chasing, filing and triggering the next step are close to fully automatable today. In our model about 80% of contract admin can go to software. What stays human is reviewing non-standard terms a client asks for, because that is a legal and commercial judgement.
- [Can call notes and CRM logging be automated?](https://theautomationindex.com/automate/call-notes-and-crm-logging/): 81% automatable; about 14 of 17.5 hours a week back. Yes. AI transcription and summaries now write usable call notes, and the same output can fill CRM fields, log the outcome and create follow-up tasks. In our model about 80% of logging time is automatable. A rep still needs to skim each summary, and judging whether a deal is really at risk stays mostly human.
- [Can commission tracking be automated?](https://theautomationindex.com/automate/commission-tracking/): 74% automatable; about 5 of 7 hours a week back. Mostly. Pulling closed deals and payments, applying commission rules and producing statements are well suited to software. In our model about 75% of commission work is automatable. Matching payments to the right rep is harder than it looks, and disputes stay human, because commission is pay and people need someone accountable.
- [Can sales pipeline updates be automated?](https://theautomationindex.com/automate/pipeline-stage-updates/): 66% automatable; about 5.5 of 8.5 hours a week back. Largely. Moving deals when something objective happens, such as a booked meeting, a sent proposal or a signature, is close to fully automatable. In our model about 65% of pipeline upkeep can go to software. Whether a deal is really progressing, and its realistic value and close date, still needs the rep's honest judgement.
- [Can the handoff from sales to delivery be automated?](https://theautomationindex.com/automate/sales-to-delivery-handoff/): 58% automatable; about 4 of 7 hours a week back. Largely. Compiling the client brief, creating the project from a template and notifying the delivery team can all be automated from the closed deal. In our model about 60% of handoff time is automatable. The conversation between the closer and the delivery lead about what was promised should stay human, because that is where expectations get lost.
- [Can lead routing be automated?](https://theautomationindex.com/automate/lead-routing/): 88% automatable; about 12 of 13.5 hours a week back. Yes. Assigning leads by territory, product, availability or round robin, notifying the rep and reassigning untouched leads are almost fully automatable. In our model about 90% of routing work can go to software. What stays human is the handful of edge cases, such as existing customers or strategic accounts, that need someone who knows the history.

## Onboarding

Onboarding covers new clients and new staff: collecting forms and documents, creating accounts, sending welcome material and scheduling the first sessions. The checklist is very automatable. The relationship is not. Done well, automation removes the dead time between a signature and a first real conversation, which is where buyer's remorse grows.

- [Can client onboarding be automated?](https://theautomationindex.com/automate/client-onboarding/): 55% automatable; about 5.5 of 10.5 hours a week back. Partly, and the automatable part is the part clients notice most. Welcome messages, intake forms, folder and project setup, scheduling and chasing are largely automatable. In our model about 55% of onboarding time can go to software. The kickoff call, which is most of the remaining time, should stay human because it sets the relationship.
- [Can employee onboarding be automated?](https://theautomationindex.com/automate/new-hire-onboarding/): 54% automatable; about 4.5 of 8 hours a week back. Partly. Paperwork, form collection, account setup, first-week scheduling and manager reminders are largely automatable. In our model about 55% of onboarding time can go to software. Welcoming the person, explaining how the team really works and checking in during the first weeks should stay human. Gallup's research suggests that human part is where most companies fall short.
- [Can collecting onboarding documents be automated?](https://theautomationindex.com/automate/onboarding-document-collection/): 74% automatable; about 5.5 of 7.5 hours a week back. Mostly. Sending the request list, chasing, renaming and filing, and pulling key fields into your systems are well suited to software. AI can now check whether a document is the right type and readable. In our model about 75% of the work is automatable. Spotting real problems in the documents still needs a trained eye.
- [Can setting up staff accounts and access be automated?](https://theautomationindex.com/automate/account-provisioning/): 61% automatable; about 4 of 6 hours a week back. Largely. Creating accounts, assigning licences by role and recording who has access are well suited to automation for tools with APIs. In our model about 60% of provisioning time is automatable. Approving elevated access and setting up physical equipment still need a person, though device management can handle much of the laptop setup.
- [Can customer welcome sequences be automated?](https://theautomationindex.com/automate/customer-welcome-sequences/): 69% automatable; about 5.5 of 8 hours a week back. Largely. Welcome messages, timed check-ins and spotting customers who have not started are well suited to automation. In our model about 70% of welcome work can go to software. Answering real questions and personally reaching out to customers who are stuck should stay human, because that contact decides whether they become active customers.
- [Can employee offboarding be automated?](https://theautomationindex.com/automate/employee-offboarding/): 56% automatable; about 3 of 5.5 hours a week back. Partly. Revoking access, updating records, preparing final pay inputs and starting file transfers are largely automatable. In our model about 55% of offboarding time can go to software. Recovering equipment, handing over client relationships and the exit conversation need people. Access removal is the step where automation matters most, because a missed account is a security risk.

## Reporting

Reporting is collecting numbers from several tools, reconciling them and writing down what they mean. The collection and arithmetic are almost fully automatable today. The interpretation is partly automatable with AI summaries, but only if the underlying definitions are agreed first. Most failed reporting automations are definition problems, not tool problems.

- [Can weekly KPI reporting be automated?](https://theautomationindex.com/automate/weekly-kpi-reporting/): 76% automatable; about 4.5 of 6 hours a week back. Mostly. Pulling numbers from each tool, pasting them into a sheet and working out week-on-week changes is almost fully automatable today. In our model about 75% of the reporting time can go to software. What stays human is agreeing what each metric means and explaining why a number moved, which is the part leaders actually read.
- [Can sales pipeline reporting be automated?](https://theautomationindex.com/automate/sales-pipeline-reporting/): 61% automatable; about 3.5 of 5.5 hours a week back. Largely. Pulling deal lists, weighting the forecast and flagging stuck deals are mostly automatable today. In our model about 60% of the weekly pipeline reporting time can go to software. What stays human is judging whether a deal is real and coaching the rep who owns it, which is what the pipeline review is for.
- [Can marketing attribution reporting be automated?](https://theautomationindex.com/automate/marketing-attribution-reporting/): 69% automatable; about 4.5 of 6.5 hours a week back. Largely, but only after lead sources are captured properly. Exporting spend, matching leads to campaigns and calculating cost per booked call are mostly automatable today. In our model about 70% of the weekly time can go to software. What stays human is explaining gaps between ad platforms and the CRM, and deciding where the budget goes next.
- [Can end-of-day reports be automated?](https://theautomationindex.com/automate/end-of-day-reports/): 68% automatable; about 6 of 9 hours a week back. Mostly. Counting dials, messages and bookings, compiling everyone's numbers and checking them is almost fully automatable, because the phone system and CRM already hold the data. In our model about 70% of end-of-day reporting time can go to software. What stays human is the short note on blockers and the manager's follow-up with each person.
- [Can cash-flow reporting be automated?](https://theautomationindex.com/automate/cash-flow-reporting/): 67% automatable; about 3 of 4.5 hours a week back. Largely. Pulling bank balances, categorising transactions and rolling a cash forecast forward are mostly automatable today. In our model about 65% of weekly cash reporting time can go to software. What stays human is adjusting for one-offs the books do not know about yet, such as a big deal or tax bill, and deciding what to do.
- [Can client reporting be automated?](https://theautomationindex.com/automate/client-reporting/): 66% automatable; about 6 of 9.5 hours a week back. Largely. Pulling each client's numbers, building charts and sending the report are almost fully automatable today. In our model about 65% of client reporting time can go to software. What stays human is the commentary on what the results mean and the review call, which is where clients decide whether to renew.

## Invoicing and payments

Invoicing and payments covers creating and sending invoices, chasing overdue ones, recovering failed card payments, matching payments to invoices and preparing payroll and expenses. It is rules-based and well served by accounting software APIs, so a high share is automatable. The exceptions, disputes and approvals still need a person who owns the money.

- [Can invoice creation and sending be automated?](https://theautomationindex.com/automate/invoice-creation-and-sending/): 83% automatable; about 5 of 6 hours a week back. Mostly. Creating the invoice, applying tax and terms, sending it to the right contact and recording its status are almost fully automatable once the deal or job holds the right data. In our model about 85% of invoicing time can go to software. What stays human is billing queries, disputes and unusual work that does not fit a rule.
- [Can overdue payment reminders be automated?](https://theautomationindex.com/automate/overdue-payment-reminders/): 63% automatable; about 3 of 5 hours a week back. Largely. Building the overdue list, sending polite reminders on a schedule and logging promises to pay are almost fully automatable today. In our model about 65% of chasing time can go to software. What stays human is phoning long-overdue accounts and deciding when to pause work or escalate, because those are decisions about a relationship.
- [Can failed payment recovery be automated?](https://theautomationindex.com/automate/failed-payment-recovery/): 64% automatable; about 3.5 of 5 hours a week back. Largely. Spotting failed charges, retrying them at sensible times, asking customers to update their card and pausing access are almost fully automatable today. In our model about 65% of recovery time can go to software. What stays human is calling high-value customers after repeated failures, where the real reason is often not the card.
- [Can payment reconciliation be automated?](https://theautomationindex.com/automate/payment-reconciliation/): 71% automatable; about 4 of 5.5 hours a week back. Largely. Importing bank feeds, matching exact payments to invoices and splitting card-processor payouts into sales, fees and refunds are mostly automatable today. In our model about 70% of reconciliation time can go to software. What stays human is investigating partial and unidentified payments, and signing off the period before the numbers are reported.
- [Can expense receipt capture be automated?](https://theautomationindex.com/automate/expense-receipt-capture/): 77% automatable; about 6.5 of 8.5 hours a week back. Mostly. Capturing receipts, matching them to card transactions, coding them and chasing missing ones are largely automatable today. In our model about 75% of expense receipt time can go to software. What stays human is approving out-of-policy spending and fixing the odd receipt that cannot be read or matched.
- [Can payroll preparation be automated?](https://theautomationindex.com/automate/payroll-preparation/): 68% automatable; about 3.5 of 5 hours a week back. Largely. Collecting timesheets, checking hours against schedules, adding commissions and entering starters and leavers are mostly automatable today. In our model about 70% of payroll preparation time can go to software. What stays human is answering pay queries and the final review before money moves, because payroll mistakes reach people's bank accounts.

## CRM hygiene

CRM hygiene is keeping customer records accurate: merging duplicates, filling missing fields, tagging lead sources, closing dead deals and capturing form data cleanly. It is unglamorous and almost entirely automatable, and it decides whether every other automation in this Index works. Bad data in a CRM is automated into bad messages at scale.

- [Can duplicate contact merging be automated?](https://theautomationindex.com/automate/duplicate-contact-merging/): 74% automatable; about 4 of 5.5 hours a week back. Largely. Finding duplicates, merging exact matches and blocking new duplicates at the point of entry are highly automatable today. In our model about 75% of the work can go to software. What stays human is reviewing fuzzy matches, such as a shared office phone number, and settling arguments over who owns a merged contact.
- [Can contact data enrichment be automated?](https://theautomationindex.com/automate/contact-data-enrichment/): 78% automatable; about 12 of 15.5 hours a week back. Mostly. Looking up company size, industry, role, LinkedIn profile and verified contact details is almost entirely automatable with enrichment tools. In our model about 80% of research time can go to software. What stays human is checking the personalisation line before it is sent and judging fit for leads that do not match any rule cleanly.
- [Can lead source tagging be automated?](https://theautomationindex.com/automate/lead-source-tagging/): 67% automatable; about 4 of 6 hours a week back. Largely. Capturing ad and campaign data on forms, tagging calls by tracking number and filling sources from ad platforms are highly automatable. In our model about 65% of tagging time can go to software. What stays human is tagging referrals and DM leads that leave no digital trail, and deciding attribution rules when sources conflict.
- [Can stale deal clean-up be automated?](https://theautomationindex.com/automate/stale-deal-cleanup/): 63% automatable; about 3.5 of 6 hours a week back. Partly. Finding deals with no activity, nudging reps, sending a final check-in and moving lost deals into nurture are automatable today. In our model about 60% of clean-up time can go to software. What stays human is the rep's honest call on whether a deal is alive and the manager's review of what it does to the forecast.
- [Can form-to-CRM data capture be automated?](https://theautomationindex.com/automate/form-to-crm-data-capture/): 88% automatable; about 8 of 9.5 hours a week back. Mostly. Copying web form submissions into a CRM, formatting fields, matching existing contacts and alerting the right person are almost fully automatable. In our model about 90% of the work can go to software. What stays human is reading unusual free-text answers and re-keying the odd handwritten form that AI cannot read.

## Roles

- [Business owner](https://theautomationindex.com/role/owner-operator/): Chief Executives (O*NET 11-1011.00)
- [Operations manager](https://theautomationindex.com/role/operations-manager/): General and Operations Managers (O*NET 11-1021.00)
- [Executive assistant](https://theautomationindex.com/role/executive-assistant/): Executive Secretaries and Executive Administrative Assistants (O*NET 43-6011.00)
- [Office administrator](https://theautomationindex.com/role/office-administrator/): Secretaries and Administrative Assistants, Except Legal, Medical, and Executive (O*NET 43-6014.00); Office Clerks, General (O*NET 43-9061.00); Data Entry Keyers (O*NET 43-9021.00)
- [Receptionist](https://theautomationindex.com/role/receptionist/): Receptionists and Information Clerks (O*NET 43-4171.00)
- [Appointment setter](https://theautomationindex.com/role/appointment-setter/): Telemarketers (O*NET 41-9041.00)
- [Sales rep or closer](https://theautomationindex.com/role/sales-rep/): Sales Representatives of Services, Except Advertising, Insurance, Financial Services, and Travel (O*NET 41-3091.00)
- [Sales manager](https://theautomationindex.com/role/sales-manager/): Sales Managers (O*NET 11-2022.00)
- [Customer success or account manager](https://theautomationindex.com/role/customer-success/): Customer Service Representatives (O*NET 43-4051.00)
- [Bookkeeper](https://theautomationindex.com/role/bookkeeper/): Bookkeeping, Accounting, and Auditing Clerks (O*NET 43-3031.00); Billing and Posting Clerks (O*NET 43-3021.00); Payroll and Timekeeping Clerks (O*NET 43-3051.00)
- [HR coordinator](https://theautomationindex.com/role/hr-coordinator/): Human Resources Assistants, Except Payroll and Timekeeping (O*NET 43-4161.00)
- [Marketing coordinator](https://theautomationindex.com/role/marketing-coordinator/): Market Research Analysts and Marketing Specialists (O*NET 13-1161.00)

## Categories

- [Operations](https://theautomationindex.com/category/operations/)
- [Appointment setting](https://theautomationindex.com/category/appointment-setting/)
- [Follow-up](https://theautomationindex.com/category/follow-up/)
- [Sales admin](https://theautomationindex.com/category/sales-admin/)
- [Onboarding](https://theautomationindex.com/category/onboarding/)
- [Reporting](https://theautomationindex.com/category/reporting/)
- [Invoicing and payments](https://theautomationindex.com/category/invoicing/)
- [CRM hygiene](https://theautomationindex.com/category/crm-hygiene/)

## Optional

- [Full text of every entry](https://theautomationindex.com/llms-full.txt)