What an ERP Implementation Actually Costs, and Where the Money Really Goes

Every week someone asks me what an ERP project costs, and every week I give the same slightly unsatisfying answer. The software is the cheapest part, and almost nobody quotes the expensive part properly. So here is the whole thing laid out, with the numbers I can actually verify and the variable parts explained clearly enough that you can build your own estimate this afternoon, before you talk to a single vendor.

The license is the small number

Start with what the software itself charges, because that part is public and fixed. The platform we build on publishes three tiers. The entry tier gives you a single app with unlimited users at no cost. The standard tier gives you every app for one per user monthly fee. The top tier adds a studio builder, multi company and external API access on top of the same all apps bundle.

At the time of writing, the published yearly rates sit around seven to nine dollars per user per month for Standard and around eleven to fourteen for Custom. Pricing differs by country, so open the official pricing page in your own currency rather than trusting any blog on this, mine included.

Run that for a ten person team and you are looking at roughly a thousand to seventeen hundred dollars a year in license. That is the number most founders fixate on. It is almost never the number that decides whether the project works.

The real cost is decisions, not software

Here is what actually consumes an implementation budget. Somebody has to decide what a customer record looks like in your business. Somebody has to decide when a quote becomes an order, who is allowed to discount, what happens to stock when a delivery arrives half short, how a returned item re-enters inventory, and which of your seven current spreadsheets is the real source of truth when two of them disagree.

The system will happily model any of those answers. It cannot invent them for you. Every hour a consultant spends waiting for you to decide who owns the pricing rule is an hour you pay for and get nothing from.

This is why two companies with the same headcount and the same modules receive quotes that differ by a factor of five. It is not the software. It is how much undecided business logic is still sitting in the founder's head.

The four things that genuinely move a quote

The first is the number of modules that have to talk to each other. Sales alone is a small job. Sales plus inventory plus manufacturing plus accounting is not four small jobs, it is one large one, because every handoff between them is a rule someone has to write down.

The second is data migration. Clean data in one consistent format is a short task. Eleven years of history across three systems, with duplicate customers and product codes that changed twice, is often the single biggest line on the invoice. If you do nothing else after reading this, go and count how many places your customer list currently lives.

The third is customization versus configuration. Configuration means using what the platform already does, through settings and Studio. Customisation means writing code. Configuration is fast and it upgrades cleanly. Custom code is slower to build and you carry it forever, because every future version has to be tested against it. Push hard toward configuration and be honest about the few places your business really is different.

The fourth is localisation. In the UAE, Qatar and Saudi Arabia that means VAT handled properly inside the transaction flow rather than bolted on at reporting time, Arabic and right to left layouts if you need them, and e-invoicing readiness. That work is real and it is not optional, so make sure any quote you receive names it explicitly instead of hiding it inside a round number.

Where projects quietly go over

Three patterns cause most overruns, and all three are avoidable.

Scope that grows one small request at a time is the most common. Nobody ever asks for a second phase. They ask for one more field, one more report, one more approval step, and eight weeks later the go live date has moved. Fix this by writing down what phase one includes and, more importantly, what it excludes, then putting every new idea on a phase two list instead of into the build.

Training treated as an afterthought is the second. A perfect system that your team works around is a failed project. Budget real hours for training and expect the first two weeks after go live to be slower than your old process, not faster.

The third is having no owner inside the business. If nobody on your side is accountable for answering questions within a day, your implementation partner stalls and bills you for the wait. Name that person before you sign anything, and give them the authority to actually decide.

How to scope your own project this afternoon

Take an hour and write four lists. First, the processes that hurt today, in the order they cost you money. Second, every place your data currently lives, including the spreadsheets and the WhatsApp groups. Third, how many people will genuinely log in, which is usually far fewer than your headcount. Fourth, the three reports you would look at every Monday if you could.

Those four lists are ninety percent of a proper scope document. Send them to any implementation partner and you will get a far more accurate quote, faster, and you will be able to tell immediately who read them and who just sent you a template.

What a sane first phase looks like

Pick the one process that hurts most and put only that into the system. Get it live, get people using it daily, then add the next one. A first phase that goes live in weeks and is genuinely used beats a twelve month program that arrives complete and gets quietly abandoned.

This is also the cheapest way to buy information. After one live process you will know how your team actually behaves in the system, and every estimate you make after that is grounded in something real rather than in a workshop.

For reference, we run fixed scope ERP setups at Talent Alliance Hub starting at 595 dollars for a first build, precisely because the first phase should be small enough that the decision is easy and the result shows up quickly.

One honest next step

If you have those four lists, or even just the first one, send them over and I will tell you straight what your project looks like, what it does not need, and what it would cost. If it turns out you do not need an ERP yet, I will tell you that too. Start the conversation on our contact page.

Related reading

If this was useful, you may also want what a CRM setup actually costs a small business, the 30 day spreadsheet to ERP migration plan and UAE e-invoicing for SMEs and what to fix before July 2027.