Dealer portal cost: the five-year total at 25, 100 and 250 dealers

Almost every article on this question answers with a range so wide it is useless: a dealer portal costs somewhere between $15,000 and $300,000. That range is not invented, and the published vendor figures in the table below do span it, from Box Software's $15,000 basic tier to ScienceSoft's $300,000 ceiling. It is also true and helps nobody. The range is wide because it silently mixes five different cost lines and three different billing models into one number. This page separates them, then does the arithmetic for three specific network sizes.

Every competitor figure below is linked to the page it came from. Where a vendor publishes nothing, that is recorded as "not published" rather than filled in with a guess. We build these systems, so we are not a neutral party; the defence against that is that you can check every number yourself.

Where a dealer portal cost actually comes from

Five separate lines make up the total, and quotes routinely present one of them as if it were all five.

1. Licence or build. The recurring subscription, or the one-time project fee. This is the only line most articles discuss, and on a five-year view it is frequently not the largest.
2. Implementation. Configuration, catalogue and price-list loading, dealer account setup, historic data migration. Orderwerks publishes onboarding "starting at $2,500". ITQlick puts Sana Commerce's onboarding at $5,000 to $20,000. A custom build folds this into the project fee, which is one of the few genuine simplifications of the build route.
3. Integration. The link to your ERP or accounting system. This is the line that moves timelines most: Yunsoft publishes 3 to 5 months for a foundational portal and 6 to 9 months once heavy ERP integration is in scope. Roughly doubling a schedule for one line item tells you how much of the work it is.
4. Training and adoption. Getting dealers to actually use it. Note that the question "what if my dealers resist the portal" appears in the published FAQ of both Orderwerks and Yunsoft, two vendors selling opposite things. When competing vendors both pre-empt the same objection, it is the standard failure mode rather than an edge case.
5. Change. Everything you ask for after go-live. On a subscription this is either free because the vendor already built it, impossible because they will not, or a bespoke development quote. On a build it is a rate. Neither is inherently cheaper; they fail differently.

What each vendor actually publishes

Before any arithmetic, here is what is on the record. Figures are as shown on the linked page at the time of checking.

The assumptions behind the five-year table

A five-year total is only meaningful if you can see what went into it. These are the assumptions. Replace them with your own before you use the result.

• Horizon: 60 months, prices held flat. No subscription contract guarantees this, so treat the rented rows as optimistic.
• Dealer logins: two per dealer on average, so 50, 200 and 500 dealer-side users at the three network sizes.
• Internal staff on the system: 4, 8 and 14 people respectively. This is the line most worth adjusting to your own reality, because it drives every per-seat figure below.
• Scope held constant: per-dealer price resolution, stock visibility rules, order entry, order and delivery status, account balance, and a one-way ERP link. Not storefront checkout, not card payments, not a native mobile app.
• Custom hosting: $6,000 across five years, the bottom of Box Software's published $100 to $2,000 per month range. A busier portal sits higher.

Five-year total by dealer count

The rented rows exclude the customer-account tier, which Orderwerks prices separately and does not publish, so they are floors rather than totals. The built rows include five full years of optional maintenance, which is the pessimistic reading of our own model. The KodDelta figure at 250 dealers assumes a build at the start of the $20,000 and up band; scope above that band costs more, and the "and up" is not decoration. One caution on the second row: ITQlick's anchor is 100 licensed users, not 100 dealers. At two logins per dealer, a 100-dealer network is already 200 users plus internal staff, so that cell is conservative rather than alarmist.

Why our band is so much lower, stated plainly

A reader who has just seen $180,000 and then $39,750 attached to the same two words is right to be suspicious. Three reasons, none of them magic:

• The scopes are not the same. Box Software's enterprise tier explicitly includes SOC 2 and HIPAA compliance work, white-label capability, custom mobile applications and AI features. We do not hold ISO 27001 or SOC 2, and we do not sell a compliance certification programme. If your buyer requires one, our band does not cover it and nobody should pretend otherwise.
• The cost base is different. Development is done in Türkiye. That is a structural difference in engineering cost, and it is the entire reason the nearshore model exists.
• We publish a floor, not a ceiling. "$20,000 and up" means exactly that. A 250-dealer network with two-way ERP synchronisation, multi-currency pricing and a returns workflow is a bigger project than the bottom of that band, and the quote will say so in writing.

The single biggest lever: are dealer logins licensed users?

Compare two published positions. Orderwerks states that customer accounts are "tiered by count, not per-user", meaning the staff who administer the portal consume seats and the dealers who log into it do not. ITQlick's figures for Sana Commerce scale with users, which is how a deployment reaches over $100,000 a year at 100 of them.

Those two structures produce completely different five-year outcomes from the same headline monthly price. If dealer logins are free, growing from 25 to 250 dealers roughly triples your cost, and the increase comes from hiring internal staff rather than from the network itself. If dealer logins are licensed, the same growth multiplies the bill by ten and keeps going.

So this is what to put in writing before you look at a single feature list:

• Does a dealer login consume a paid licence?
• If a dealer has three buyers, is that one account or three?
• What is the price at dealer number 101, and at 251?
• Is that in the contract, or is it "current pricing"?

A vendor who answers all four in writing is offering you a budget. A vendor who answers none is offering you a demo.

When renting genuinely beats building

We build custom systems, and there are situations where we will tell you not to. These are them, and the arithmetic above supports every one.

1. Under about 25 dealers with a single published price list. $16,900 over five years on a per-seat product beats the bottom of our own five-year figure of $18,800, and it is live in 2 to 4 weeks against 4 to 8 weeks for our multi-department band. There is no argument to make here.
2. Your process is still moving. If the discount rules, the approval chain or the delivery model have changed twice this year, you are still discovering the process. Building a system around a process that has not settled is how a project ends up rewritten. Rent while it settles, then build.
3. You need it live this month. Nothing custom goes live in two weeks with an ERP link attached. If the deadline is external, take the packaged product.
4. Nobody internally will own it. A custom system without an owner degrades. A subscription at least has somebody else's roadmap keeping it alive.
5. You do not yet know whether your dealers will use a portal. Given that dealer resistance is the objection both major vendor FAQs pre-empt, this is a live risk. A cancel-anytime subscription is a cheap way to find out, and the answer is worth more than the licence fee.
6. Fewer than about five gaps against a published feature list. If a packaged product does almost everything, configure it. Building in order to close five gaps is the most expensive way to close five gaps.

The mirror image is worth stating too. Building tends to win when per-dealer pricing is the core of your commercial model, when dealer logins are licensed users and the network is growing, when the ERP link has to run both ways, and when you expect to keep changing the system for years. In that last case the deciding factor is often not price but permission: on a build, a change is a decision you make rather than a request you file.

The lines that do not appear in quotes

These are the items that turn a signed number into a different actual number. Ask about all nine in writing, whichever route you take.

1. The dealer-account tier. Orderwerks publishes seat pricing but tiers customer accounts separately without publishing those tiers. This pattern is common, and it is the line most likely to move your total.
2. Onboarding as a separate fee. Published where it exists: Orderwerks from $2,500; ITQlick puts Sana Commerce's at $5,000 to $20,000.
3. Data migration. ITQlick lists data migration among the hidden fees for Sana Commerce, at $1,000 to $5,000. Loading five years of dealer order history is a project, not a checkbox.
4. ERP integration. Yunsoft's own published timelines move from 3 to 5 months to 6 to 9 months for this single line. Whatever it costs, it is not a rounding error.
5. Hosting and infrastructure. Bundled inside a subscription, a real line on a build. Box Software publishes $100 to $2,000 and above per month.
6. Maintenance. Box Software publishes 15% to 25% per year for custom portals. We publish 12% to 25% and make it optional. On the rented side there is no separate maintenance line at all, which is a genuine advantage of that model.
7. Training and dealer onboarding. Every dealer who has to be walked through a first order is time. At 250 dealers that is a project with a schedule, not an afternoon.
8. Change requests after go-live. Ask for the day rate, the lead time, and what happens when the answer is "that is on the roadmap".
9. Renewal increases and exit. Our table holds prices flat for 60 months, which no subscription guarantees. Separately: what does it take to get your catalogue, price lists, dealer accounts and order history out, in what format, and who pays for the export? Settle that at the start rather than at the end.

When you should not build a portal at all

The most useful thing a build shop can tell you is when not to hire one. Here is the test, and it is arithmetic rather than opinion.

Count the orders your dealers place in a typical month, then multiply by the minutes it takes to key one in from an email or a phone call. If your dealers place around 60 orders a month and each takes five minutes, that is five hours a month, or roughly 300 hours across the same five-year horizon as the table above. Almost no version of a portal pays for itself against 300 hours. A shared price list, a standard order form and a rule that everything arrives by email will serve you better and cost nothing.

Four conditions flip that. Any one of them is usually enough:

• Prices differ per dealer. The moment a human has to look up which price applies, you have an error rate, and price errors cost more than the keying time ever did.
• Order keying passes about a day a week. Roughly 100 orders a month at five minutes each. Past that it is a job rather than a task, and it is a job nobody was hired for.
• Balance and delivery questions arrive several times a day. This is the cheapest thing a portal ever fixes, and usually the reason the phone stops ringing.
• You cannot push a price change without telephoning people. If a campaign takes a week to reach the network, the portal is a commercial tool rather than an administrative one, and it should be costed as one.

If none of those are true yet, keep the spreadsheet. Come back when one of them is.

Run it on your own numbers in three questions

1. How many dealers will you have in three years, and how many of their people need a login? That second number, not the first, is what a per-user vendor bills. Have it before you ask for a quote.
2. How many of your commercial rules are missing from the published feature list? Write them as sentences, not as feature names. Zero to five means configure a product. More than fifteen means you are about to rebuild your commercial model inside somebody else's framework, and you should at least price building it properly.
3. What is your integration budget, separately from the licence or the build? If you cannot answer this, the comparison is not ready. It is the line that most often doubles a project and the one nobody publishes.

Bring those three numbers and the decision usually resolves in one conversation. If they point to a subscription, we will say so. We would rather lose a project than deliver one that should not have been built.

Related reading

• What a B2B dealer portal contains, module by module, including per-dealer price resolution and stock visibility rules.
• ERP customer portal versus dealer portal: where the ERP module stops and a separate portal starts.
• A B2B ordering portal for international distributors: what belongs in the first release.
• Dealer portal software: buy it or build it, the same vendors judged on fit rather than on five-year cost.
• ERP pricing 2026: the same published-numbers treatment applied to ERP.
• KodDelta pricing and delivery bands, published in full.

Sources

• Orderwerks Pricing (orderwerks.com): $60 per user per month base platform, onboarding starting at $2,500, no transaction fees, customer accounts tiered by count
• Orderwerks Dealer Portal (orderwerks.com): "Pricing starts at $100/month", "live within 2-4 weeks", "No per-transaction fees"
• ITQlick, Sana Commerce pricing (itqlick.com): from $10,000 per year, licence $15,000 to $30,000 a year, onboarding $5,000 to $20,000, hidden fees $1,000 to $5,000, one-time implementation, customisation, training and migration $5,000 to $50,000 and above, over $100,000 a year at 100 users
• Capterra, Sana Commerce listing (capterra.com): "Contact vendor for pricing", no free trial, no free version
• Growmax, Best Dealer Portal Software 2026 (growmax.io): states $199 per month and 2 to 4 weeks implementation for itself
• Growmax ARC pricing page (growmax.io): "Custom Pricing, Book a Demo"
• Yunsoft, B2B dealer portal development (yunsoft.com): 3 to 5 months foundational, 6 to 9 months with heavy ERP integration, no price published
• ScienceSoft, custom web application development (scnsoft.com): customer-facing web portal $70,000 to $300,000, enterprise web app $50,000 to $250,000
• Box Software, cost to build a customer portal in 2026 (boxsoftware.net): $15,000 to $30,000, $35,000 to $80,000, $90,000 to $180,000 and above; maintenance 15% to 25% per year; hosting $100 to $2,000 and above per month
• Capterra, b2b.store reviews (capterra.com): reviewers cite "Lack of integrations with other software" and adding products through "cumbersome import and export via CSV"
• KodDelta pricing and delivery bands

Sources checked in August 2026. Vendors change prices without notice and regional rates differ; open the linked page before deciding anything. ITQlick figures are that publisher's estimates rather than Sana Commerce's own published prices, and are labelled as such throughout. Product names belong to their respective owners. KodDelta is an independent third party and is not a partner, reseller or certified implementer of any vendor named on this page.

Frequently asked questions

How much does a dealer portal cost?

It depends far more on how it is billed than on what it does. Published rates in this market run from $60 per user per month (Orderwerks) up to an estimated $15,000 to $30,000 a year in licence alone for a mid-market B2B commerce platform (ITQlick, on Sana Commerce). Published custom build bands run from $15,000 to $180,000 and above (Box Software) and $70,000 to $300,000 (ScienceSoft). KodDelta publishes $3,000 to $6,000 for a single module, $8,000 to $15,000 for a multi-department system and $20,000 and up for an end-to-end platform, as a one-time fee with source code delivered.

Is it cheaper to build a dealer portal or subscribe to one?

Below roughly 25 dealers, subscribing usually wins on five-year total and always wins on time to live. Above that, the answer turns on one question: does a dealer login consume a paid licence? If dealer accounts are tiered separately from user seats, the subscription stays flat as your network grows and renting stays competitive for a long time. If dealer logins are licensed users, the subscription curve rises with dealer count and a one-time build overtakes it within a few years.

What does a dealer portal cost per year to maintain?

The commonly published figure for custom software is 15% to 25% of the original build cost per year; Box Software states that range for customer portals specifically. KodDelta publishes 12% to 25% and makes it optional rather than a condition of the licence. On top of that sits hosting, which Box Software puts at $100 to $2,000 and above per month depending on scale.

How long does it take to build a dealer portal?

Published timelines vary by an order of magnitude because the same word covers very different things. Orderwerks states most manufacturers are live on its product within 2 to 4 weeks. Yunsoft states 3 to 5 months for a foundational custom portal and 6 to 9 months once heavy ERP integration is involved. Box Software states 2 to 3 months for a basic portal and 7 to 12 months and above for an enterprise one. KodDelta publishes 2 to 4 weeks for a single module, 4 to 8 weeks for a multi-department system and 3 to 6 months for a full platform.

How many dealers do you need before a portal is worth building?

There is no universal number, but there is a test. Count the orders your dealers place in a month and multiply by the minutes it takes to key each one in. If the answer is under about five hours a month and your prices are the same for everyone, a shared price list and an email order form is honestly the correct answer and no portal is justified. The economics turn when per-dealer pricing exists, when order keying passes about a day a week, or when balance and delivery questions arrive several times a day.