The short version
- Hire a freelancer for focused work with the senior person actually doing it. Hire an agency for parallel specialisms and guaranteed cover.
- Ignore published day rates. The pricing model and the scope determine what you pay.
- The best filter question: “What would you refuse to automate, and why?”
- Agree the baseline numbers before work starts, or you will never settle whether it worked.
- Own your accounts. Always. This is the mistake that costs the most to undo.
Do you actually need one?
Some honesty first, because it costs me work to write it. You probably do not need to hire anyone yet if:
- You get fewer than a handful of enquiries a month. Your problem is demand, not efficiency — and consulting on automation will not create demand.
- Nobody can tell you where your current leads come from. Fix that yourself first; it is free and it changes what you should buy.
- You have not tried the obvious things: replying faster, asking for reviews, making the enquiry path clearer.
- You want someone to “do AI” without a specific outcome attached. That engagement ends badly regardless of who you hire.
You probably do benefit from hiring someone when:
- You have real lead flow and a bottleneck you can name.
- You are spending on ads and cannot tell what is working.
- Your team is doing several hours a week of repetitive marketing work.
- You tried building automation yourself, it half-works, and maintaining it is now a job.
- You need someone accountable for a number rather than an activity.
If you cannot name the number you want to move, you are not ready to hire — you are ready to measure.
Freelancer, agency or hire
| Freelancer | Agency | In-house hire | |
|---|---|---|---|
| Who does the work | The person you spoke to | Often someone more junior than the person you spoke to | Your employee |
| Best for | Focused scope, fast decisions, senior attention | Several specialisms at once, formal process | Ongoing work with deep product knowledge |
| Speed to start | Days | Weeks | Months |
| Main risk | Single point of failure — holidays, illness, capacity | Paying for account management rather than the work | Expensive if the role is not yet full-time work |
| Knowledge when they leave | Leaves with them unless documented | Leaves with them unless documented | Stays, mostly |
Note the last row applies to two of the three. Whoever you hire, make written documentation a deliverable rather than a courtesy — the brand voice file, the account structure, how each automation works, what to do when it breaks. That document is the asset you are actually buying, and it is the one most people forget to ask for.
What the work actually involves
A real engagement is less exciting than the pitch. If someone describes only the third item on this list, they are selling the demo.
- Finding out what is true. Where leads come from, what converts, what is broken. Usually the first week, and it frequently changes the plan.
- Writing down how your business works. Offers, audience, tone, qualification criteria. Dull, and it determines the quality of everything after.
- Building the thing. The automation, the campaign, the pages. The visible part, and often the smallest.
- Making it survive contact with reality. Testing, fixing, adjusting when it fails at 2am on a Saturday.
- Proving it worked. Baseline versus after, in numbers you agreed on beforehand.
- Handing it over. Documentation, access, training, so you are not permanently dependent.
Steps 1, 2, 5 and 6 are where engagements succeed or fail. Step 3 is what everyone talks about.
Pricing models and what drives the number
I am not going to publish a day rate, and you should be wary of articles that do — rates vary enormously between markets, scopes and levels of experience, and a single number quoted out of context does more harm than good. What is genuinely useful is understanding the models and what moves the price.
The four models
| Model | Works well when | Goes wrong when |
|---|---|---|
| Hourly | Scope is genuinely unknown, or you want advice rather than delivery | Nobody caps it, and slow work is rewarded |
| Fixed project | Scope is written down and both sides understand it | Scope was vague, so everything becomes a change request |
| Monthly retainer | Ongoing work with a defined set of deliverables per month | It quietly becomes a subscription nobody reviews |
| Performance-linked | Attribution is clean and both sides trust the numbers | Attribution is messy, which for most small businesses it is |
For a first engagement, a fixed-scope project is usually the right structure. Both sides find out how the other works, the cost is knowable, and neither of you is locked in. Move to a retainer afterwards if there is genuinely ongoing work — and put a review date on it.
What actually drives your quote
- Scope clarity. A vague brief gets a padded quote, because the person quoting is pricing their uncertainty. A precise brief gets a precise number.
- Access. If it takes three weeks to get analytics access, you pay for that.
- Decision speed. One decision-maker who replies quickly costs meaningfully less than a committee.
- State of your data. Broken tracking must be fixed before anything else can be measured. It usually is broken.
- One-off build vs ongoing. Building is a project. Running is a retainer. Do not blur them.
- Market. Rates differ substantially between India and the US for the same work. Neither is automatically better value — judge on outcome per rupee or dollar, not on rate alone.
India and US buyers
Indian buyers tend to be quoted lower rates and should push hardest on scope clarity and documentation — that is where value is most often lost, not on price.
US buyers hiring internationally get strong value, and should be explicit about time zones, response windows and who is on the call. Agree those in writing before starting; they cause more friction than rates do.
Freelance enquiry · India & US
Want a straight answer on whether this is worth hiring for?
Email me what you are trying to fix and I will tell you honestly whether it needs a consultant, a one-week project, or nothing at all. I have talked people out of hiring me more than once — usually when the real problem was demand, not efficiency.
12 questions to ask
Ask all twelve. The answers to 3, 7 and 11 tell you the most.
- What would you measure in the first 30 days, and what is our baseline today?
- What would you do first, and why that rather than something else?
- What would you refuse to automate, and why?
- Who actually does the work, and who will I speak to week to week?
- What access do you need, and what is the minimum you can work with?
- What happens to our data, and does any of it go into third-party tools?
- Tell me about something you built that did not work. What did you do?
- What do we own at the end — accounts, documentation, the automations themselves?
- What happens when something breaks after the engagement ends?
- How much of our team’s time do you need each week? Be specific.
- What would make you tell us to stop and not spend this money?
- What does the handover look like, in writing?
Question 3 separates people who understand the risk from people selling a demo. Question 7 tells you whether they have shipped anything real. Question 11 tells you whether they are optimising for your outcome or their invoice. A confident, specific answer to all three is worth more than a polished deck.
7 red flags
- Guaranteed results. Nobody can guarantee rankings, leads or placement in AI answers. A guarantee is either ignorance or a sales tactic.
- They want to own your accounts. Your ad account, analytics, domain and CMS are yours. Grant access, never ownership. This is the most expensive mistake to undo.
- No questions about your business. If they quote without asking who you sell to, they are quoting a template.
- Tool-first pitching. Leading with a stack rather than your problem means the solution was decided before they heard the problem.
- Vagueness about who does the work. Ask directly. Ask again if the answer moves.
- No documentation in the deliverables. Undocumented automation is a dependency, not an asset.
- They cannot explain it simply. If someone cannot explain what they will build in language you follow, either they do not understand it or they are hiding the scope.
The one that costs the most
Account ownership. Businesses lose years of ad history, analytics data and domain control this way, and recovering it is slow, sometimes impossible, and always more expensive than the engagement was. Create the accounts yourself, in your own name and billing, and add people as users. No exceptions, including for me.
The brief that gets an accurate quote
The single highest-leverage thing you can do as a buyer. Send this to three people and you get three comparable quotes instead of three different guesses. Ten minutes to write.
SUBJECT: [What you want] — [your business]
About us
- What we do, who we sell to, which markets
- Website URL
- Rough size: team, monthly revenue band or ad spend
The problem
- What is happening now that we want changed
- What we think is causing it (even if we are unsure)
- What we have already tried
Success looks like
- The number we want to move: [cost per lead / enquiries / hours]
- Where it is today: [your baseline — say "unknown" if unknown]
- By when
Practical
- Budget range we are working with
- Who decides, and how fast we can decide
- Access we can give: analytics, ads, CMS, CRM
- Anything off-limits
What we want back
- Fixed-scope quote, deliverables listed
- What you need from us, weekly
- What we own at the endThe two lines people leave out are budget range and baseline. Both feel like a disadvantage to share. Neither is. A budget range means you get a scoped proposal rather than a fishing expedition, and a stated baseline means you get someone who tells you honestly whether the goal is realistic. Saying “we do not know our baseline” is itself useful information — it tells them where to start.
Judging the first 30 days
Agree these checkpoints before anything begins, in writing.
- Week 1: they should be able to tell you something about your own business you did not know. If not, they did not look properly.
- Week 2: a written baseline — current numbers, what is broken, what they will fix first and why.
- Week 3: the first thing working end to end, however small. Momentum matters more than scale at this stage.
- Week 4: measurement in place, and an honest statement of what has and has not moved.
If week 4 arrives with impressive activity and no numbers, raise it immediately rather than at month three. The projects that go wrong usually showed the same warning sign early: lots of work, no agreed measure of whether it mattered.
If you want to know what a good measurement setup looks like before you start, it is in How to Track AI Marketing Automation So It Shows Up in Revenue — and the list of what is worth building first is in The 10 Marketing Automations Every Small Business Should Set Up First.
Frequently asked questions
Freelancer or agency?
Freelancer when the work is focused, you want the senior person actually doing it, and you can decide quickly. Agency when you need several specialisms in parallel, guaranteed cover, and formal process. For most small businesses running one or two channels, a freelancer gives more senior attention for the money.
How much does it cost?
Rates vary enormously by market, scope and seniority, so treat any single published number with suspicion. Focus on the pricing model and what drives it: scope clarity, access, decision speed, the state of your data, and whether it is a one-off build or ongoing work. Ask for a fixed-scope quote against a written brief.
What is the most important question to ask?
“What would you refuse to automate, and why?” It separates people who understand the risks from people selling a demo, and the answer tells you how they will behave when something goes wrong.
How long should a first engagement be?
Long enough to deliver one complete thing and measure it — commonly four to eight weeks for a focused project. Be cautious about a long retainer before either side has seen how the other works.
What if we are not happy?
Agree the exit up front: notice period, what gets handed over, and that access reverts to you. A professional will have no problem putting that in writing. Reluctance at this stage is the clearest signal you will get.
Thejas Acharya
Digital growth consultant based in Bangalore, working with founders and marketing teams across India and the US. I build websites, landing pages, Meta Ads campaigns and AI automation that are judged on one thing: whether they generate leads. You own every account and every document I touch.