The Best Case You Have Is the Question You Answer Forty Times a Day
Where is my order. Can I change the delivery address. Will this fit. When does it arrive. High volume, almost no judgement, and one correct answer sitting in a system you already have. That combination is genuinely rare and e-commerce is full of it — which is why this is the vertical where automation most reliably earns its cost, and also why the opposite case has to be named on the same page.
No credit card required · 30-minute call · No obligation to sign up
Six questions, and three of them belong to a person
The dividing line is not how common the question is. It is whether there is one correct answer available without a judgement call.
- Where is my order?Automate · the courier knows
- Can I change the address?Automate, up to a cut-off point
- Will this fit me?Automate · if your sizing is real
- It arrived damagedPerson · every time
- I want to return it, latePerson · it is a judgement
- This is the third timePerson, and quickly
The last row is the one that gets missed and it is the most expensive. A customer whose second complaint is being handled by a bot has already decided about you; the third one just confirms it. Repeat contact about the same order should escalate automatically, and that rule is two lines of logic that almost nobody builds.
It can only answer what your systems actually know
Every good automation case here depends on a fact being retrievable. “Where is my order” is answerable because a courier reference exists and a status can be fetched. “Will this fit” is answerable only if your size data is complete and honest — and on a lot of catalogues it is neither, which turns a good automation case into a confident wrong answer at scale. The build starts by checking what is actually knowable rather than what would be nice to answer.
That means the first deliverable is frequently unglamorous data work. Filling the gaps in the fields that customers ask about, or accepting that a question cannot be automated until they are filled. It is less exciting than an assistant and it is the part that decides whether the assistant is useful or embarrassing.
The second thing to build is the escalation logic, and it is smaller than people expect. Anything about damage, anything about a late return, anything where the same customer has come back about the same order — those go to a person without the bot attempting a first answer. Attempting one and then handing over is worse than handing over immediately, because the customer has now been delayed and mishandled.
And the honest limit: automation reduces the number of ordinary contacts and it does not reduce the difficult ones. If anything the difficult ones become a larger share of what your team handles, which is a good outcome and a different job. Plan for a support team doing harder work rather than less of it, because that is what actually happens.
- ✓Only what is retrievable — A courier reference exists. Honest sizing data often does not.
- ✓Data work comes first — Unglamorous, and it decides whether the assistant embarrasses you.
- ✓Escalate before attempting — A failed first answer then a handover is worse than a handover.
- ✓Repeat contact escalates — Two lines of logic. Almost nobody builds them.
- ✓The team does harder work, not less — The difficult cases become a bigger share. Plan for it.
A customer whose second complaint is handled by a bot has already decided about you. The third just confirms it.
What we build for an online retailer
Six things, and the first is checking what your systems can actually answer.
What Is Actually Knowable
Which customer questions have a retrievable answer today and which need data work first. It decides the scope, and it is where good projects separate from confident wrong ones.
Order Status, Answered Properly
Pulled live from the courier rather than from a stale record, with an honest answer when the tracking itself has not updated for two days.
Escalation Before Attempting
Damage, late returns and anything with a judgement in it go straight to a person. A failed first answer followed by a handover is worse than the handover alone.
Repeat Contact Detection
Same customer, same order, second time — escalate immediately. Two lines of logic, rarely built, and it is where a recoverable situation becomes a lost customer.
Sizing and Fit, If Your Data Allows
Answerable only where measurements are complete and honest. Where they are not, we say so rather than letting a system guess confidently at scale.
What It Could Not Answer
Reported monthly, because that list is your next piece of data work — and frequently a product page that needs writing rather than anything to do with automation.
Product Copy at Catalogue Scale
Bulk product descriptions drafted from the attributes you already hold, in your own tone, for catalogues too large to write by hand. They stay drafts — and where the attribute data is thin, the honest output is a gap report rather than invented detail.
Stock-Aware Content
Inventory-based generation so a category page is not promoting something that sold out on Tuesday. Unglamorous, and it prevents the specific annoyance of a customer clicking through to nothing.
After the Order, Not Just Before It
Post-purchase sequences that are genuinely useful — care instructions, what to expect, when to reorder — rather than another discount two days later. Abandoned-cart recovery is the same machinery pointed earlier in the journey.
What clients say
Real clients, quoted in their own words — published with their permission.
“We’re very thankful to Ghalib and his team for building a fully-functioning website for our business.”
“It was a pleasure working with Ghalib and his team.”
“Ghalib’s clear guidance has helped improve our website’s SEO performance.”
More of them, in full, on our reviews page.
— Our Proprietary Methodology —
The Visibility Framework™, applied in E-commerce
The method doesn’t change by market. What it’s pointed at does.
Audit The Hours
Where time actually goes, task by task, scored on volume, repetition and the cost of getting it wrong. Ends in a ranked blueprint with estimated hours saved — yours to keep either way.
Design The Guardrails
Before any building: what the agent may touch, where a human must approve, what happens when it is unsure, and which data is never allowed near a third-party model.
Build & Evaluate
One workflow at a time, in your accounts, scored against real examples from your business before it touches live work. Shipped early so it meets reality while it is still cheap to change.
Run & Improve
Monitored for cost, failures and quality drift. Models change, your business changes, and an automation nobody tends becomes a liability rather than an asset.
Honest, No-Nonsense Commitment
If the audit concludes that a task is not worth automating, we will tell you and refund the difference rather than build it anyway. And if a workflow we built does not hit the outcome we agreed in the blueprint, we keep working on it at no extra cost until it does or we take it out.
AI Automation pricing for E-commerce, in GBP
Quoted in pounds after we have established what your systems can actually answer. That check is small and it regularly shortens the scope — usually by finding that two of the four questions you wanted handled need data work before anything else.
Order Questions
The high-volume, low-judgement set.
- Live order status with honest gaps
- Escalation before attempting, on the hard cases
- Repeat-contact detection built in
Full Enquiry Handling
The usual one. Site, email and WhatsApp.
- Everything in Order Questions, across channels
- Fit and sizing where your data supports it
- Monthly report of what it could not answer
- Optional: bundle in SEO or Website Development
Catalogue Scale
Large ranges or several brands.
- Data work to make more questions answerable
- Per-brand rules and escalation routing
- Evaluated before each expansion goes live
- Optional: combine all 4 services for full-funnel growth
Every plan is scoped around your market — start with a free first look and we’ll recommend what fits, priced in GBP.
What you’re actually committing to
Most agencies keep this in a contract you only see after the sales call. We would rather you knew now, because it is the question everyone asks second — right after the price.
- The audit is credited, not sunkPay for the audit, and the full amount comes off the build if you proceed. If you don’t, the blueprint is still yours to hand to anyone else.
- A fixed build price after the auditQuoted once we know what we are building. If it takes longer than we estimated, that is our risk — the price only moves if you change the scope.
- You own everythingAccounts, API keys, workflows, prompts, evaluation sets, logs and documentation — all in your name from day one, and still yours if we never work together again.
- Running costs are yours and visibleAPI usage is billed by the provider directly to you. We never resell tokens or mark up usage, and you see the real number.
- The retainer is month to month30 days’ notice, no exit fee. Stop it and your automations keep running — you are simply maintaining them yourself.
- Human approval is the defaultAnything customer-facing or irreversible needs a person until the evaluation data justifies otherwise, and that decision is yours to make, not ours.
These are the terms as they appear in the agreement itself — nothing here is softened for the website. The full wording lives in our terms and conditions, and you get the agreement to read before anything is signed or invoiced.
E-commerce AI Automation questions, answered
Related AI Automation pages
Same service, different angle — by market, by service and by industry.
Want the forty-a-day question handled?
Free 30-minute call · No obligation · Response within 24 hours
Get in touch with Ghalib Ashrafi HQ
Ghalib Ashrafi takes on work for brands in six markets — the UK, USA, UAE, Saudi Arabia, Australia and Pakistan. Every enquiry gets a reply within 24 hours.