Measured
against the hours it replaces, not a feeling
Your tools
connected, not replaced
Exceptions
routed to a person, never silently swallowed
Handed over
documented so it is not a black box
Pricing
What automation costs.
Costed against the hours it replaces. If the maths does not clear inside a year I will say so before you spend anything.
Order confirmation flow
On requestone-off
- Every order confirmed on WhatsApp before it ships
- Automatic follow-up, then a call list for the silent ones
- Replies routed to whoever handles them
- Connected to your store and your carrier
- Return-rate reporting from day one
The single automation with the clearest payback in Egyptian and Saudi commerce.
Request this ↗One process, automated
On requestone-off
- One end-to-end flow: the trigger, the steps, the exceptions
- Connected to the tools you already pay for
- Exceptions routed to a person with the context attached
- A dashboard showing what ran and what did not
- Documented so your team can change the simple parts
Priced per process. Costed against the hours it replaces, and if the maths does not work I will tell you.
Request this ↗Systems talking to each other
On requestone-off
- Two or more systems connected: store, ERP, CRM, accounting, carrier
- Field mapping agreed in writing before a line of code
- Retries, failure alerts and a manual override
- Backfill of existing data where it makes sense
- Monitoring, so a silent failure is not discovered a month later
The right answer when the same data is typed into two systems by a person.
Request this ↗Add-ons — buy with any package or on their own
Automation audit
On requestfixed
- Every repetitive process in the business, listed and costed
- What to automate, what to simplify first, what to leave
- Ranked by hours returned per pound spent
- A plan you can execute in any order
Cheaper than automating the wrong thing, which is the usual outcome without it.
Request this ↗Care plan
On requestper month
- Monitoring, with alerts when a flow stops
- Changes as your process changes
- New flows at a reduced rate
- Monthly note: what ran, what failed, what it saved
Automation breaks quietly. That is the argument for someone watching it.
Request this ↗Compare
Side by side.
Whether you are fixing returns, a process, or two systems that disagree.
| What you get | WhatsApp flow | One process | Integration | Audit |
|---|---|---|---|---|
| Messages customers actually receive | Included | If the process needs it | Not included | Not included |
| Two or more systems kept in sync | Included | Sometimes | Included | Not included |
| Exceptions routed to a person | Included | Included | Included | Not included |
| Retries and failure alerts | Included | Included | Included | Not included |
| Monitoring dashboard | Included | Included | Included | Not included |
| Backfill of existing data | Not included | Not included | Included | Not included |
| Costed against the hours it replaces | Included | Included | Included | Included |
| Every process listed and ranked | Not included | Not included | Not included | Included |
| Documented for your team | Included | Included | Included | Included |
The audit is the cheap first move if you are not sure which of these you need — it is priced to stand alone and it often changes the answer.
Before we start
What your business needs in place.
What blocks these is access and ownership, not difficulty.
Admin access to every tool involved
Not a colleague's login — proper admin, and API keys where they exist. Half the delay on an integration is waiting for someone to grant access nobody wants to be responsible for.
The process written down, including the exceptions
Especially the exceptions. The happy path is the easy part; what happens when the customer does not reply, or the stock is wrong, is where the design actually lives.
A person who owns the exceptions
Automation does not remove judgement, it concentrates it. Someone must own the cases the machine hands back, or those cases quietly become a backlog nobody is looking at.
A WhatsApp Business sender, if messaging is involved
An approved sender and template through a provider — not a phone in someone's drawer. Approval takes days and it can be started before anything else.
An honest volume number
Orders or cases per day, and how long each takes now. This is what the price is judged against, and it is what tells us whether the automation is worth building at all.
Terms
How the money and the risk are split.
No retainer to start a conversation, and no invoice for work that never got delivered.
Half to start, half on delivery
Balance falls due when the work is done and handed over. Not delivered, not charged.
Fixed price, not an estimate
The number holds unless you change the scope. If I misjudged it, that is mine.
The clock starts at handover
From the day I have everything on the list, not from the day you pay.
One round of revisions included
Then hourly, quoted before anything starts. No surprise invoices.
You own everything at the end
Code, accounts, assets and credentials. Nothing is held hostage.
Proof
Automation running on my own money.
These are not client references. They are the systems my own businesses depend on.
Order confirmation at OurGrowth
WhatsApp confirmation before dispatch, follow-ups and a call list — built because my own cash-on-delivery returns were the problem.
our-growth.com ↗Egypt Post and Bosta
Carrier integration for labels, pickups and tracking, wired into the order flow rather than bolted beside it.
InstaPay collection
Payment links and reconciliation automated end to end, running against real settlement.
70 skills, self-optimising
The OurGrowth orchestrator: an automation platform, not a single flow.
our-growth.com ↗What you get
What running automation looks like.
Every one of these is measured against the hours it replaced, not asserted.
The flow itself
Trigger, steps, retries and the manual override — running against your real systems.
Exceptions routed to a person
With the context attached. Nothing the machine cannot decide is silently swallowed.
A view of what ran
What succeeded, what failed, what is waiting. Silent failure is the thing this prevents.
The monthly number
Hours returned against the build price, measured from where we started.
What you send
The handover, in one list.
Send it in one go and the clock starts.
- The process as it runs today, step by step
- Admin access and API keys for the tools involved
- Volume: how many per day, and how long each takes
- Who handles the exceptions, and how they want them
- Message templates or wording you already use
- Any compliance rule the flow must not break
How it runs
Six steps, and you see each one.
-
01
Map
Walk the process with whoever runs it, including what happens when it goes wrong.
-
02
Cost it
Hours per week today against the build price. If it does not pay back inside a year, I say so before we start.
-
03
Build
The flow, the integrations and the exception routing, in a test environment against real data.
-
04
Shadow run
It runs alongside the manual process for a week. Nothing is switched off until the numbers match.
-
05
Switch over
Manual step retired, monitoring on, and the team trained on the exception queue.
-
06
Measure
Back to the number from step two. Either it saved what we said, or we find out why.
Scope
What this is not, and who it is not for.
These are quoted separately, so the price above stays a price.
- Subscriptions to the tools being connected
- WhatsApp or SMS message fees
- Rewriting the systems being integrated
- Staffing the exception queue
- Data cleaning of historical records
- Process consulting beyond what the build requires
Right fit, wrong fit
Saves a call that would end in no.
A good fit
- Someone retypes the same data into two systems
- COD returns are eating the margin
- A process depends on one person remembering
- You want it measured, not just automated
- You already pay for tools that do not talk to each other
Not a fit
- The process changes every week and nobody can describe it
- Nobody will own the exception queue
- The volume is too low to pay back — I will tell you if it is
- You want AI in it regardless of whether it helps
After launch
Once it is running.
Automation fails silently — a token expires, an API changes a field, a carrier renames a status. Nobody notices for a fortnight, and then it is a fortnight of orders. The Care plan is monitoring plus somebody who knows the flow.
Questions
The ones people actually ask.
How do you know it will pay back?
We count first. Hours per week times the loaded cost of the person doing them, against the build price. If it does not clear inside a year I will say so, and that has happened more than once.
Will it replace people?
Usually it replaces the worst part of someone's day, not the person. Where headcount is genuinely the point, say so up front — it changes what is worth building and what needs to be watertight.
What happens when it breaks?
It alerts, it retries, and anything it cannot resolve goes to a person with the context attached. The design assumption is that it will fail sometimes, because it will.
Do we have to change tools?
No. The default is connecting what you already pay for. If a tool genuinely cannot do it, I will show you why rather than assert it.
Can our team change it later?
The simple parts — message wording, timings, thresholds — yes, and the handover covers exactly that. Structural changes are a conversation, and usually a small one.
Tell me what you have, and I will tell you what it takes.
The app, the market and the deadline is enough to get a straight answer on scope, price and timing — usually the same day.