By Ishan Rana, Founder · Updated July 2026
The AI Opportunity Checklist: 12 Places to Look Before You Spend a Dollar
The best AI opportunities hide in your most repetitive, rules-based work: data re-entry, first-reply lag, quote and invoice prep, scheduling, reporting, and lookups. Work this 12-point checklist, self-score each item 0-3, and total your score. Anything you score 2 or 3 is a candidate worth costing before you spend a dollar on tools.
- Score each of the 12 items 0-3. Anything scoring 2 or 3 is a real AI or automation candidate.
- The best first target is high-frequency, rules-based, and low-judgement work, not your hardest problem.
- Copy-paste between apps, first-reply lag, and quote/report prep are the three that pay back fastest for most SMBs.
- Skip anything rare, one-off, or heavy on human judgement. AI there costs more than it saves.
- Total your score at the end to see whether you have one obvious win or a full backlog.
The AI opportunity checklist (before you buy anything)
Most AI spend gets wasted because owners buy a tool, then go looking for a job it can do. Do it backwards. Find the job first. The best AI and automation wins hide in your most repetitive, rules-based, low-judgement work, not your hardest problem.
Work through these 12 checks. For each one, tick the box and give yourself a self-score, 0 to 3:
- 0 = doesn’t happen here
- 1 = happens, but rare or one-off
- 2 = happens weekly, eats real time
- 3 = happens daily, painful, and always the same shape
Add up your scores at the end. Anything you score 2 or 3 is a genuine candidate worth costing before you spend a dollar.
The 12 checks
1. ☐ Copy-paste between two apps. Check: does anyone re-type or re-paste the same data from one system into another (orders into accounting, leads into a CRM, form replies into a sheet)? Why: this is the single most common, highest-payback automation. It’s pure rules, zero judgement. Score 0-3: ____
2. ☐ First-reply lag. Check: how long before a new lead or customer message gets any reply? Measure it honestly over a week. Why: speed-to-first-reply drives close rate. An instant AI acknowledgement or triage often lifts conversion more than a better tool ever will. Score 0-3: ____
3. ☐ The same questions, all day. Check: do you and your team answer the same handful of questions (hours, pricing, availability, status) over and over? Why: a repeatable question set is a clean candidate for an AI assistant or a saved-reply flow. If the answers are stable, a machine can give them. Score 0-3: ____
4. ☐ Quote, estimate, or proposal prep. Check: how long does it take to turn an enquiry into a sent quote, and how much of that is assembling the same blocks? Why: templated quoting is fast to automate and directly speeds cash in. If 80% of every quote is boilerplate, that 80% is the opportunity. Score 0-3: ____
5. ☐ Invoicing and chasing payment. Check: are invoices raised by hand, and does someone manually chase overdue ones? Why: reminders, follow-ups, and status nudges are rules-based and money-linked. Automating the chase alone often pays for the whole project. Score 0-3: ____
6. ☐ Scheduling and booking back-and-forth. Check: how many messages does it take to book a call, job, or appointment? Why: if the answer is more than one, a booking flow removes the ping-pong and the no-shows. Low effort, visible daily win. Score 0-3: ____
7. ☐ Reporting you rebuild every week or month. Check: does anyone hand-build the same report (sales, ops, stock, KPIs) on a schedule by pulling numbers into a sheet? Why: recurring reports are the definition of automatable. Same inputs, same shape, same day. Only the numbers change. Score 0-3: ____
8. ☐ Reading and sorting inbound documents. Check: does someone open PDFs, emails, or forms to pull out a few fields (invoices, receipts, applications, resumes) and route them? Why: extract-and-route is exactly what AI is now good at. If a human is a glorified copy machine here, that’s a strong signal. Score 0-3: ____
9. ☐ Lookups mid-conversation. Check: do staff put customers on hold to go find an answer in a manual, spreadsheet, or old thread? Why: an internal search or assistant over your own docs kills hold time and inconsistent answers. The knowledge exists, it’s just hard to reach. Score 0-3: ____
10. ☐ Content you rewrite from a template. Check: are product descriptions, listings, replies, or social posts written from the same skeleton each time? Why: templated writing is where AI genuinely saves hours, drafting the 80% you then edit. Note: draft, not publish-unread. Score 0-3: ____
11. ☐ The bottleneck person. Check: is there one person every task waits on, whose absence stops the business? Why: whatever sits in that queue is your real constraint. Automating around it, or capturing what only they know, is worth more than any generic tool. Score 0-3: ____
12. ☐ The task you dread. Check: what’s the recurring job you or your team actively avoid because it’s mind-numbing? Why: dread usually means high-frequency, low-reward, and rules-based, which is the exact profile of a good automation. Your gut already knows. Score 0-3: ____
Where NOT to bother
Not everything wants AI. Skip anything that is:
| Signal | Why to skip it |
|---|---|
| Rare or one-off | Setup costs more than you’ll ever save |
| Heavy on human judgement | Trust, negotiation, and taste don’t automate well yet |
| Different every single time | No pattern means no rules means no reliable automation |
| Legally or safety-critical with no check | Never let a machine be the last word on something that can hurt you |
| Cheap and already fast | If it takes 30 seconds a month, leave it alone |
Chasing these is how AI budgets get burned. Be honest and cross them off.
Score yourself
Add up your 12 scores. Max is 36.
- 0-8: No urgent AI case yet. Fix process first. Come back in a quarter.
- 9-18: You have one or two clear wins. Cost those, ignore the rest.
- 19-28: A real automation backlog. Sequence it by frequency times minutes times error-cost, and start at the top.
- 29-36: You’re leaking hours daily. Prioritise hard, or you’ll try to boil the ocean.
Your first build should be the highest-scoring item that also needs the least human judgement. Not your hardest problem. Your most repetitive one.
Want this done for you?
This checklist is the manual version of what we do in an AI Game Plan: a $500 audit where we sit with your actual workflows, find your highest-ROI automation, scope it, and hand you a build plan. The $500 is credited toward the build if you move forward. See how it fits the wider ladder on pricing, or go deeper with how to integrate AI into your business.
Our guarantee is simple: we’ll find you a day a week to save, or the Game Plan’s free.
Book a 15-minute intro call: calendly.com/ishanranawork/15-minute-intro-call.
FAQ
How do I find where AI would actually help my business?
Look for work that is frequent, rules-based, and low-judgement: data entry, first replies, quote and report prep, scheduling, lookups. Score each on this 12-point checklist. The high-scorers are your candidates. Ignore rare or judgement-heavy tasks.
What should I automate first?
The task with the highest frequency times minutes-per-run times error-cost, that also needs the least human judgement. That is usually copy-paste between two apps, chasing invoices, or answering the same handful of questions all day.
Do I need to buy AI tools to run this checklist?
No. That is the point. Run the whole checklist first, self-score, and only then price a fix for your highest-scoring items. Most owners find one or two obvious wins and a pile of noise.