Custom configurator for PLN 60,000 one-time. Enterprise CPQ SaaS for PLN 2,500/month per user. PandaDoc template for PLN 500/month. Which option is for you? The answer is not universal; it depends on the number of merchants, offer values, and your ERP. In this article: a 10-dimensional comparison, 3-year TCO with real numbers, hidden SaaS costs.
You have received three offers for implementing a configurator. The first is from a software house offering a custom solution for a one-time payment of PLN 60,000 and a 10-week implementation period. The second is an enterprise CPQ SaaS (Epicor or Tacton) with an PLN 80,000 setup fee plus a monthly subscription of PLN 2,500 per user. The third option is PandaDoc or Proposify with template customization for PLN 500 per month.
The board asks: „which one are we taking?”. The IT Director says, „Custom,” the Finance Director says, „Template, cheaper,” while the Sales Director says, „SaaS, they have five hundred companies on their reference list.” As a result, everyone is right – and everyone is only partly right.
This is why this article breaks down the decision into three dimensions: Process fit, Total Cost of Ownership over 3 years, lock-in in a 5-year perspective.
The startup cost says one thing. The TCO after 3 years says another. The lock-in after 5 years tells the truth.
Are you comparing three offers and don't know which one to choose? A 30-minute conversation, and let's be upfront – even if the answer is „take Epicor, not us.”. | Arrange a call |
Three categories of solutions (not three products)
Check if the configurator makes sense on your end
Survey 15 questions (3 min). After filling it out, we'll both know if there's anything to talk about.
First and foremost, we need to distinguish between three very different categories of tools, all of which are called „configurators.”.
CATEGORY A · CUSTOM
Custom configurator from a software house
In other words: a custom system, written from scratch for your process. In Poland, JSON Crew (us), Ideo, LemonMind, Webtom, Software Mill, and Hicron do this. The stack differs per vendor – we use Next.js + Three.js + Nest.js.
- Dinner 35,000 – 300,000 PLN one-off
- Timeline: 8-12 weeks MVP, 4-6 months full functionality
- Match 100 days% under your process
- Code ownership: Yours forever
- Maintenance ~20% project value annually
CATEGORY B · ENTERPRISE CPQ SAAS
Global CPQ Platforms
Ready-made solutions for manufacturing enterprises. Main players: Epicor CPQ, Tacton CPQ, Configure One Cloud, Salesforce CPQ. Plus Threekit for 3D-heavy projects.
- Dinner setup $20-100k + subscription $75-200/user/month
- Timeline: 2-4 months (setup + configuration + training)
- Match 70-80%%, the rest are workarounds
- Ownership Subscription (you lose access when you stop paying)
- Integrations popular systems OK, legacy / Polish ERP – problem
CATEGORY C · PROPOSAL TEMPLATE
Offer Templates / Low-code
However, these are not configurators in the technical sense. Essentially, they are offer templates with drag-and-drop: Proposify, PandaDoc, Qwilr, Better Proposals, Kickflip. Rather beautiful PDFs than real configurators.
- Dinner 30-100 USD per user/month (no setup)
- Timeline: 1-2 weeks
- Match 30-50% (ok dla prostych produktów)
- Restrictions No real configuration, no ERP integration, vendor branding
Comparison Table — 10 Decision Dimensions
| Dimension | Custom | Enterprise SaaS | Proposal template |
| Launch price | 35-90k PLN (MVP) | 80,000-400,000 PLN | 2,500-5,000 PLN per year |
| Ongoing cost | ~20% values/year | $75-200/user/mies | $30-100/user/months |
| Matching | 100% | 70-80% | 30-50% |
| ERP Integration | Full (API, custom) | Restricted to partners | Brake / webhook |
| 3D / Visualization | Three.js, full | Threekit addon (personal) | Brak |
| Implementation time | 8-12 weeks MVP | 2-4 months | 1-2 weeks |
| Code ownership | Yours forever | Subscription (lock-in) | Subscription |
| Changes to the rules | Sprint dev (2 weeks) | Consultant$$$ or checkbox | Fast, but limited |
| Scalability | Unlimited | Dependent on pricing | Limited |
| For whom | 5-20 salespeople, unique process | Enterprise 50+ salespeople | Simple products, <5 offers/month |
Total Cost of Ownership - 3-year real-world scenario
For example, let's take a typical company from our portfolio: an agricultural machinery manufacturer, 5 salespeople, 10-15 quotes per month per salesperson, average quote value PLN 80,000. They need a configurator with pricing rules + 3D visualization + CRM integration.
| Scenario | Year 1 | Year 2 | Year 3 | 3-year TCO |
| Custom (60k MVP + 20%/year) | 72 000 | 12 000 | 12 000 | 96,000 PLN |
| Enterprise SaaS (setup + 5 users) | ~184 000 | ~24 000 | ~24 000 | ~232,000 PLN |
| Proposal Template (5 Users) | ~12 000 | ~12 000 | ~12 000 | ~36,000 PLN* |
But scenario 3 assumes that the proposal template is sufficient for you from a business perspective. Consequently, if your product has more than a few variants, you will end up buying a configurator in a few months anyway. If your product has more than a few variants, it won't be enough, and you'll be buying a configurator in 6 months regardless.
Break-even
Custom vs. Enterprise SaaS: ~18 months with 5 traders. With 10 traders, custom is cheaper in the first year.
Hidden costs that SaaS sellers don't tell you
1. Customization fees
First of all, any change to pricing rules that doesn't fit in the admin panel – consultant hours. Example: you want to add a rule „customers from Germany get 8% discount, but only for orders >50,000 EUR”. In Epicor / Tacton – it's either in the template, or you pay consultants $200-400/h. In custom – sprint dev, 5-15 thousand PLN one-time.
2. Per-user pricing puzzle
W praktyce SaaS skaluje koszt liniowo z liczbą użytkowników. Przy 5 handlowcach ($100/mies per user = $6k/rok) – rozsądnie. Przy 50 handlowcach ($60k/rok) – boleśnie. Custom koszt utrzymania nie rośnie z liczbą userów.
3. Integration lock-in
Specifically, CPQ SaaS integrates with popular systems – Salesforce, Dynamics, SAP, MS, NetSuite. If your ERP is Comarch XL from 2015, Subiekt GT, or a proprietary .NET system from 2011, there is no native integration. Options are: (a) customize the ERP, (b) write custom middleware, (c) abandon the dynamic pricing.
4. Contract lock-in
Namely: you stop paying for the subscription = you lose access to customer data, configuration history, and offer templates. Exporting to CSV rarely solves the problem – half of the value is in the logic, which you cannot export.
5. Pricing volatility
Moreover, the SaaS provider is changing their pricing – you have no choice. In 2023 Salesforce raised CPQ prices by ~12%% for most contracts. Custom – once the pricing is set, it only changes when you commission an expansion.
When to choose what — decision tree
Go to CUSTOM if...
- You have >50 product variants or complex exclusion rules
- Offer value >50,000 PLN
- 5-20 salespeople
- Integration with your own ERP (not Salesforce/SAP)
- Unique pricing rules
- Configurator = strategic asset
Bet on ENTERPRISE SAAS if...
- >500 employees, IT budget >500k/year
- Standard Manufacturing
- 50+ sales representatives
- No in-house dev team
- Do you accept a 5+ year lock-in?
- Do you already have an enterprise stack (Salesforce, SAP)?
TEMPLATE is enough when...
- The offer is 3-5 variants
- Simple, static price list
- <5 offers per month
- Budget <500 PLN/month
- You don't need ERP integration
- Do you accept provider branding
When is custom not worth it
To summarize honestly, not every company should build custom. If you fall into any of these categories, a ready-made SaaS will be cheaper and faster:
- Do you have it yet Salesforce Enterprise and you need native integration – then Salesforce CPQ is a better choice
- You are in the business automotive / aerospace with very standard processes – Configure One has preset templates
- Do you want to start in Two weeks (custom requires min. 8 weeks MVP)
- Your budget is Under 20,000 PLN (custom pricing doesn't make sense in this context)
- Your product is commodity without unique rules – just a price calculator
We don't sell custom projects to companies they don't actually fit. a separate article about the configurator's price We have outlined when it's not worth using configurators at all.
How do we do it at JSON Crew
First off: custom configurators are our core. We've implemented three: Akpil (57 types of agricultural machinery, 136 price list pages), Forest (Hunting weapon – model + stock + caliber + accessories), plus a construction industry project. Additionally, ten manufacturing companies are using our platform. JSON Hub, which combines a configurator + CRM + digital quoting in one place.
Stack: Next.js + Three.js + Nest.js, Fixed Price for MVP, 8-12 weeks. 5 most common implementation problems we've detailed separately - it's required reading, regardless of which provider you choose.
JSON Hub as an alternative to Enterprise CPQ
However, for companies considering Enterprise CPQ but not needing the scale of Epicor / Tacton, we offer JSON Hub. This is our SaaS platform for manufacturing companies, which connects:
- Product configurator (like Epicor)
- CRM with leads (like Salesforce)
- Digital quoting (like PandaDoc)
- Dealer panel (like Configure One)
In one tool, with native integration with Polish ERP systems (Comarch, Subiekt, SAP Business One). Pricing per company, not per user. Ten manufacturing companies are already using it.
The questions we hear most often
„If Tacton is more expensive, why do so many companies use it?”
This is because for an enterprise with 50+ salespeople and a mature IT stack, Tacton is a rational choice. The setup fee is amortized across many users, the subscription model fits their operational budget (opex, not capex), and advanced constraint-solving handles truly complex products. For a Polish company with 8 salespeople and a single production plant, it's overkill.
„Does JSON Hub replace Epicor?”
Not to its full extent. Above all, Epicor CPQ has 20 years of history, thousands of enterprise clients, and deep integrations with SAP and Oracle. JSON Hub targets a different segment – Polish manufacturing companies with 50-500 employees and unique sales processes. For this segment, JSON Hub is often better (cheaper, more flexible, Polish-language support, integration with Polish ERPs). For General Electric – no.
„We already have Salesforce. We're getting Salesforce CPQ, right?”
On one hand: if your sales team already uses Salesforce – native integration. But remember: Salesforce CPQ costs $$75-150/user/month, requires additional configuration, and rarely has ready-made integrations with Polish systems (Comarch, Subiekt). With 10 salespeople, that's $$900-1,800/month – PLN 43,000 – 86,000 annually. A custom configurator costs less to start, and after 2 years, you're on your own.
„Can we start with SaaS and then move to custom?”
Yes, but in practice, it's expensive. Namely: migrating from SaaS to custom isn't export-import. It's rewriting business rules, which in SaaS reside in a proprietary format. We've seen this with a client – two months of team work plus access to historical data, which the SaaS provider reluctantly provides. It's better to decide from the start.
What to do if you're comparing options
Order a diagnostic call 30 minutes, free. Come with three answers: 1. How many salespeople and how many offers/month? 2. What is the average offer value? 3. What ERP do you use? | Fill out the form |
As a result, we will state it directly: custom, enterprise SaaS, or proposal template. Even if the answer is „Epicor CPQ, because you have Salesforce and 80 salespeople” – we will say it and recommend a good integrator.
P.S. The most common reaction after a TCO comparison is: „I didn't realize custom would be cheaper in 18 months.” It's not a secret; it's math that no one selling SaaS does. The salesperson shows you the monthly subscription. You have to multiply it yourself by 12 x 5 years x number of traders, add the setup fee, customization hours, consultant dependencies, and the hidden cost of lock-in. Then the comparison looks different.
