What I automate / Reports that build themselves
The report that builds itself
Almost every business has one: a report somebody assembles by hand on a fixed day, from figures that live in three or four places. It takes a morning. It is late whenever that person is on leave. And because it is assembled under time pressure, nobody fully trusts it — which is why the meeting it feeds usually starts with someone questioning a number.
The work is not the thinking. The thinking takes ten minutes. The work is fetching, pasting, reformatting, checking that the totals tie, and chasing the one person who has not sent theirs. All of that is mechanical, and all of it can be done on a schedule.
What you get back is not just the hours. It is a report that arrives at the same time every week whether or not anyone is available, calculated the same way every time, with its numbers defined in one place instead of in somebody's head.
You probably know this is you if
- The same file gets rebuilt every week, from the same sources, by the same person
- You chase two or three people for their figures before you can start
- Nobody can fully explain how one of the numbers is calculated any more
- The report is late or missing whenever one person is away
- Two teams arrive at a meeting with different versions of the same figure
The jobs in this area
Rebuilding the same report every week
The numbers are pulled, checked and formatted on a schedule, and land in your inbox before you ask for them.
Chasing four people for their figures
Each person fills one short form, or the figure is read straight from the system it already lives in. No chasing.
Copying yesterday's totals into a slide
The slide or dashboard updates itself from the source, so the version you present is never the stale one.
Nobody agreeing which number is right
One definition per number, written down in the job that calculates it, so two teams stop arguing about whose export is correct.
Finding out about a bad week on Friday
A threshold you choose, checked daily, that messages you the moment something moves the wrong way.
How a build like this goes
- 1
First we agree the definitions — what exactly counts as a sale, an active customer, a completed job. This takes longer than the code and is usually the most valuable part.
- 2
Then the fetching: from the systems that have an export or an API, from the sheets people maintain, from the files that arrive by email.
- 3
Then validation, which is the part most home-made reporting skips. Totals that must reconcile, values that must exist, rows that must be unique. Anything that fails is set aside and named, never quietly averaged away.
- 4
Then the output — an email, a PDF, a sheet, a dashboard, whatever people will actually read — on a schedule, with an alert if a source is missing or the job does not run.
A worked example
A four-branch weekly sales report
- Before
- One person, every Monday morning, about three hours: four files, three date formats, branch names typed differently, one invoice sent twice.
- After
- It runs at 06:00 on Monday. The report is in four inboxes before anyone arrives, with three rows flagged as needing a human.
3 hours a week × 52 = 156 hours a year. At a loaded ₹200/hour that is about ₹31,000 a year, for a build in the ₹18,000–₹45,000 band. It pays for itself somewhere between eight and eighteen months on hours alone — and that ignores the late reports and the arguments about whose number is right.
Those are illustrative figures, not a quote and not a promise. Your own numbers are the only ones that matter — the calculator does the same arithmetic on them.
What this normally costs
A job in this area is usually a quick win: ₹18,000 to ₹45,000, ready in 3 to 7 working days. The exact number comes from the audit, in writing, before anything is built.
The audit itself is ₹9,999 and comes off the first build in full. If it turns out this is not worth automating for you, that is what the report will say.
Questions about this kind of work
Our figures come from software that has no API.
That is the normal case. If it can export a CSV or Excel file, print a report, or reach a database, it can be automated around without touching the software or breaking your support contract.
What if the report needs commentary, not just numbers?
Then the automation does the numbers and leaves you a draft to write the commentary into. The judgement stays with a person; the fetching and formatting does not.
Can it go into our existing dashboard instead of an email?
Yes, if that is where people actually look. Most teams read an email and ignore a dashboard, so it is worth being honest about which is true for you before building the prettier option.
Tell me about your version of this
Tell me the job you are tired of
One call, half an hour. You will get a straight answer about whether it is worth automating — including 'no' if that is the answer.