.png?ts=1782305185)
E-commerce setup cost planning
The cost of an e-commerce setup is almost always underestimated, because the ongoing items are missing from the estimate. This breakdown separates one off from recurring cost and names the five items that get forgotten.
Why the first estimate is almost always too low
When people budget an e-commerce setup they think about the build: agency, shop system, design. That is the visible part. The invisible part starts on launch day and continues every month after it. That is exactly why first estimates regularly land at half the real figure. They contain the project but not the operation.
The decisive split for planning is therefore not cheap versus expensive, it is one off versus recurring. One off costs can be negotiated and spread. Recurring costs have to be earned by the shop every single month, which makes them the factor that determines how much revenue is actually needed.
One off costs
The figures below are ranges from projects in the German mid market. They move with catalogue depth and integration effort, but they work as a planning frame.
| Item | Range | What drives the effort |
|---|---|---|
| Concept and consulting | 2,000 to 8,000 EUR | Number of customer groups, process variety |
| Legal advice | 1,500 to 6,000 EUR | Digital goods, subscriptions, B2B business |
| Shop system and base setup | 5,000 to 15,000 EUR | Licence model, multi shop, languages |
| Design and templates | 4,000 to 25,000 EUR | Level of customisation, number of page types |
| Data import | 2,000 to 12,000 EUR | Product count, quality of the source data |
| Integrations | 5,000 to 25,000 EUR | ERP, CRM, logistics, accounting |
| Content and copy | 3,000 to 15,000 EUR | Number of categories and products, languages |
| Testing, training, launch | 2,000 to 8,000 EUR | Number of user roles, process complexity |
For a shop with a manageable catalogue and no ERP integration that gives a range of roughly 24,000 to 45,000 EUR. Once an ERP, B2B pricing and several languages are added, the effort realistically lands between 60,000 and 120,000 EUR. The spread is wide because the level of integration drives the effort more than the number of products does.
Recurring costs
This is where the real planning mistake sits. The monthly items are modest individually, but over three years they can easily exceed the build.
| Item | Per month | Note |
|---|---|---|
| Hosting | 50 to 400 EUR | Scales with traffic and data volume |
| Shop system and extensions | 0 to 300 EUR | Licences, subscription extensions |
| Payment providers | 1 to 3 percent | Revenue based, no fixed cost |
| Maintenance and development | 500 to 2,500 EUR | Security updates, small changes |
| Marketing and advertising | 1,000 to 10,000 EUR | The item with the widest range |
| Operating and maintaining data | 200 to 1,500 EUR | Product data, prices, images |
| Ongoing legal work | 100 to 500 EUR | Adjustments, updates |
Without the marketing line, running a mid sized shop costs roughly 1,000 to 3,000 EUR per month. That is the figure worth knowing before the start, because it determines the payback period.
Five items that almost always get forgotten
The same gaps appear in almost every project. They are not exotic, they are just systematically underestimated.
- Maintaining product data. Products get added, prices change, images get replaced. This work is permanent and needs either staff or a supplier. It does not stop after launch, it grows.
- Maintenance after launch. A shop system needs security updates, and updates regularly break extensions. Without a maintenance budget you end up with either an insecure shop or an outage that costs more than the maintenance would have.
- Keeping legal texts current. Legal pages are not a one off project. New services, new payment methods and changed requirements all need adjustments, and each adjustment is usually billable advice.
- Returns and rework. Every return costs shipping both ways plus time. Anyone planning a five percent return rate while their product group runs at twenty percent has budgeted for a loss without seeing it.
- Quality of the import source. The effort behind a data import depends less on the number of products than on the state of the source. A clean spreadsheet is migrated in two days. Five years of grown ERP data takes weeks.
Turning that into a budget
A budget holds up when three numbers fit together: the one off cost, the monthly fixed cost, and the contribution the shop can realistically earn after launch. The third number is the unknown, and it can only be estimated.
A simple calculation is enough for a first read. If the average order value is 120 EUR, the contribution per order is 40 EUR, and the shop expects 1,500 orders in its first six months, then 60,000 EUR of contribution is available. That has to cover the running cost and part of the build. Doing this arithmetic before the start shows early whether the project carries.
The content foundation behind it, meaning which pages, categories and texts are actually needed, is covered in the E-Commerce Setup seminar along a complete project plan.
Where saving is worth it
Not every line should be trimmed equally. Experience says saving works in three places and backfires in three others.
Saving makes sense on design, provided the template is clean, on features nobody needs before the hundredth order, and on projects that get reviewed again after six months. Saving becomes expensive on product data, on the ERP integration and on the legal framework. Those three come back with interest, because rework there always exceeds the original effort.
Frequently asked questions
What is the minimum cost of an e-commerce setup?
A working shop with a manageable catalogue, no ERP integration and no custom programming can be built for roughly 20,000 to 25,000 EUR. Add about 1,000 EUR per month for running, maintaining and updating it. Anyone significantly below that has usually outsourced items that will have to be done later.
Which costs scale with revenue?
Revenue based are mainly payment provider fees, marketplace commissions, shipping and the allocated cost of returns. All four scale with the business, which makes them more predictable than fixed cost. A shop system with a revenue based model behaves similarly.
When does a shop pay for itself?
For a shop costing 30,000 to 60,000 EUR to build with 2,000 EUR of monthly running cost, payback after 18 to 30 months is realistic, provided contribution per order exceeds 35 EUR and demand is stable. The calculation depends more on the repurchase rate than on first order value, because repeat orders spread acquisition cost across several years.
How much does the payment method matter to the budget?
A great deal. Invoice payment often lifts conversion noticeably but shifts default risk to the merchant, and it costs either a fee for credit protection or the losses themselves. At thin margins a single payment method can tip the calculation, which is why fees should be measured against margin and not against revenue.
What does the first year actually cost?
In the first year the one off build cost and every running cost from launch onwards add up. For a project costing 35,000 EUR to build with 1,800 EUR per month, going live after three months, that is 35,000 plus 16,200 EUR, roughly 51,000 EUR. That total is the figure liquidity planning needs, because it shows how much capital is tied up before break even. Budgeting only the build systematically underestimates the capital requirement in year one.
Should marketing sit in the build budget?
Yes, but separately. Marketing is not a build item, it is an ongoing investment that generates the revenue the build is paid back from. Hiding marketing inside the build budget makes the payback calculation look better than reality and the project gets corrected too late. A separate marketing budget with its own success measurement is the clean approach.