The problem is not the practice system. It is what sits in between
When an accounting firm calls us about “a software problem”, in the great majority of cases the practice system works perfectly well. What does not work is everything around it.
An average firm uses three or four different applications — accounts, payroll, tax filings, document management — often from different vendors, each perfectly in order on its own. Between one and the next there is nothing, and that nothing gets filled by a person rekeying data, by a spreadsheet somebody keeps updated, and by a habit nobody has ever written down.
The cost of that nothing appears on no invoice, and it is high: hours of qualified work spent copying, and it is also the point where a figure can go wrong without anybody noticing.
Where data gets lost in a firm
It is always the same four points, and you will recognise them immediately.
Between accounts and payroll. The same client records and the same transactions, entered twice into two systems that do not talk.
In the documents arriving from clients. Invoices, receipts, contracts, in every imaginable format, which somebody opens, looks at, renames and files by hand. It is the most repetitive and most easily automated work there is in a firm.
Towards the authorities’ portals. Filings, submissions, receipts to download and reconcile. Much of this is already automatable and almost nobody automates it.
In the client records. The same client company exists in three systems with three versions of the data, and nobody knows which is the right one. It is the dullest problem and the one that generates the most errors downstream.
What we do, in order
First we look at the flow, not the software. We follow a document from the moment it enters the firm to when it leaves, noting every time a person touches it. It is a few days’ work and it almost always produces the first real map of how the firm actually functions.
Then we verify what is technically possible. Some practice systems expose a way of being queried from outside, others do not, as a commercial choice. This has to be established before promising anything, and the answer changes the options.
Then we automate the stretch with the best return. Not everything: the step that consumes the most hours and has the fewest exceptions. An automation that works in eighty per cent of cases and leaves the rest to a person is worth far more than one that tries to cover everything and breaks.
Custom only where needed. We build something new — a portal, a web app, an internal tool — when no standard product covers that process. Not as a matter of principle, and not before verifying that it does not already exist.
The client portal, when it makes sense
It is the most frequent request after integration, and it is worth being honest about when it pays.
It pays if the firm exchanges documents with many clients and does so today by email — which is awkward to search, heavy to file, and questionable when third parties’ data is inside. A portal gives each client their own space, with the right permissions and a record of who uploaded what and when.
It does not pay if there are few clients, or if they will not use it. A portal nobody opens is worse than email, because it adds a channel instead of replacing one. The question to ask beforehand is how your clients actually work, not how elegant the tool is.
The rest of the firm’s IT
A project like this rests on things that have to be in order already: individual logins rather than one shared account, client credentials kept in a manager rather than a spreadsheet, backups that genuinely restore and that capture the practice databases properly.
See also IT, AI and security for accounting firms, AI for accounting firms for the document side, and custom software development for the method.
The first step
We follow a document from entry to exit and count how many times somebody touches it. Out of that comes the list of what can be taken out of the way, with priorities. Free and without obligation.