What an e-commerce consultant does and what you should check

The week before a campaign, an order gets stuck between the website and the business system, and nobody knows who should find the error or who should fix it.

An e-commerce consultant helps you turn business goals into functions, integrations and a plan for product ownership after go-live. Judge the consultant by how you will own product data, pricing rules and ongoing management.

What an e-commerce consultant is actually responsible for

Picture a company that wants to bring its website, customer portal and reseller portal together around a shared product range. First, someone needs to pin down which channels are included, which customers will use them and how price lists and ranges differ. That is strategy and scope. If those decisions are put off, both the budget and the delivery become less certain.

The next part is technical delivery. The consultant needs to understand how product data, prices, stock and orders should move between the website and the business system, for example Business Central, Monitor or Jeeves. An order can look correct on the website and still be missing a customer number or a valid pricing rule when it reaches the business system. That is why responsibility and error handling need to be described for the whole flow.

Picture a company that wants to bring its website, customer portal and reseller portal together around a shared product range. First, someone needs to pin down which channels are included, which customers will use them and how price lists and ranges differ. That is strategy and scope. If those decisions are put off, both the budget and the delivery become less certain.

The next part is technical delivery. The consultant needs to understand how product data, prices, stock and orders should move between the website and the business system, for example Business Central, Monitor or Jeeves. An order can look correct on the website and still be missing a customer number or a valid pricing rule when it reaches the business system. That is why responsibility and error handling need to be described for the whole flow.

The third area decides whether the solution holds up over time. Someone needs to track versions, fix bugs, test compatibility and make sure the product information stays correct. The consultant's assignment can cover this, but it is separate from the supplier's responsibility for operations and from your own decisions about range, customer segments and pricing rules.

Start by deciding which area matters most to you. If orders and prices often create extra work, review integration and data ownership first. If the platform works technically but nobody takes responsibility after launch, start with the management setup.

Five competencies and how to test them

Judge the answers through concrete examples. A consultant who can describe a workflow from product field to confirmed order shows more than one who only lists platforms or tools.

For product data and PIM, ask how attributes are modelled, who owns them and how they reach the website, portal and resellers. A good answer follows, for example, a changed unit of measure step by step to each channel.

For integrations with the business system, ask what happens when an order fails to go into Business Central, Monitor or Jeeves, and who spots the error first. A good answer describes alerts, a named owner and how the customer is informed.

For customer-specific prices and permissions, ask how customer-specific prices, negotiated discounts and role-based permissions are handled. A good answer shows which rules apply to each role and how they are tested before go-live.

For experience and measurement, ask which steps in the buying journey are measured, from product view to confirmed order, and how performance is tested. A good answer names measurement points and a test plan.

For ongoing management and platform knowledge, ask which versions are supported, how compatibility is tracked during upgrades and which functions are already tested. A good answer includes a list of tested functions and an upgrade plan.

Then run three operational tests. Start with a normal week, continue with a simulated peak week and finish with a scenario where the business system responds slowly for a while. Watch what happens to queue length, API response time, error rate and data latency between the systems. It helps to separate the flows for product data, prices, stock and orders, so an error in one flow can be found and handled without guesswork.

Capacity should be tested before a campaign period so that campaign rules, stock levels and order status behave predictably under heavy load. Plan the test around your own volumes from previous campaigns.

Score each area by how concrete the answer is. A good answer contains an example, a clear owner and a way to test it. If all you get are general promises, you do not yet have a basis for judging whether the solution can be managed over time.

Working methods and pricing that reveal how the solution will be managed

A serious process starts with a review of the current state, moves on to scope, estimate, build and go-live, and then describes ongoing management. If you already have a solution that causes problems, the review should show where the code is unstable and which parts need replacing. A structured picture of the current state shows which functions and integrations exist and what they cost to maintain. It is also the starting point for Glanser's 72-hour analysis.

A fixed price per function and ongoing hourly billing spread the risk in different ways. A fixed price gives you a clearer cost within the agreed scope. It requires the functions and boundaries to be written down before work starts, otherwise every change becomes a discussion about extra work. Hourly billing suits you when scope changes often and you want close control of the work, but you take on more responsibility for keeping track of hours and total cost. With a fixed price the supplier carries the risk within the agreed scope. With hourly billing you carry it yourselves when the scope changes.

An estimate should show a range and state which questions could affect the scope. Ask for assumptions and exclusions in writing. Without them, a quote can look clear while the risk of add-ons ends up with you. Three items often missed in the budget are the costs of ongoing management, upgrades and bug fixing. Read how Glanser Care describes management responsibility and compare it with what is actually included in your proposal.

We describe our way of working from current state to ongoing management openly, so you can compare our approach with other suppliers'. Ask everyone for the same material: scope, assumptions, the monthly cost of management and a clear answer on what an upgrade costs. Glanser builds solutions with the Function Library: ready-built, tested and compatible functions and integrations with a fixed scope and a fixed price. Functions can be delivered at three levels, Standard, Adapted and Customer-unique, where a customer-unique function is isolated so it does not affect the rest during upgrades.

AI needs clean product data before it is allowed to publish

It is easy to start with a chatbot because customers can see it. But an AI tool that writes product copy can cause problems if it lacks clear rules for product fields, prices and stock status. So start by deciding which information may be changed automatically and who approves the result.

According to Statistics Sweden's figures on AI use in businesses, AI use among Swedish companies with at least ten employees rose from 25.2 percent in 2024 to 35.0 percent in 2025. That shows why the AI question can reach the e-commerce team before responsibility for data and workflows has been assigned.

Ask the consultant to describe which product fields AI may enrich, which changes require manual approval and how the source of each change is recorded. Prices and stock levels should come from governed sources, with clear permissions that stop incorrect data from being published. For customer answers in a portal, the rules also need to stop one user from receiving information that belongs to another customer. Ask for a written description of what data the tool stores and how it handles personal data, and compare it with the Swedish Authority for Privacy Protection's guidance on GDPR and AI.

A PIM can collect, enrich and distribute product information to the website and other channels. See how Glanser describes structured product information, and read about why product data needs a clear owner. Start with version-controlled product data and clear responsibility across marketing, product, IT and the business system before you put a chatbot into production.

When an e-commerce consultant is not the right choice

If your range is small, you sell through one channel and no integration is needed, a standard solution may be enough. An internal owner can then handle updates and keep an eye on the order flow without a larger consultancy setup.

Unclear business rules are another clear obstacle. If you cannot describe price lists, customer segments and approval flows, every estimate becomes uncertain. A quote can structure the work, but it cannot make the decisions for you. So bring the right people together before you ask for a fixed price.

A fixed price is a poorer fit when scope changes often. If you need to rethink functions as you go, hourly billing or a smaller first delivery can give you more room to manoeuvre. That also requires someone on your side to keep track of priorities and cost.

A partnership needs a product owner who can make decisions within a working week, a person who owns the product data, access to a test environment for the business system and time for acceptance testing. If your internal team has the right skills and time, you can also manage the solution yourselves. A consultant can then provide limited support without taking over product ownership.

Three questions that reveal how the supplier works

Ask which parts are built on existing, tested functions and which are developed specifically for you. A vague answer makes it hard to understand how much of the cost is predictable.

Ask who owns the product data, pricing logic and order flow if the contract ends or you change supplier. If the answer does not say how you get access to data and documentation, a future switch may become more complicated.

Ask what happens when the business system gets a new version, which tests are run, who pays and how long the upgrade takes. An unclear answer means responsibility for compatibility may land with you without showing up in the budget.

Decide who owns your data before you choose

Decide who owns the product data, pricing rules and order flow before you compare quotes. That decision affects both the technical solution and how easily you can change your way of working later. Write down the suppliers' answers and compare them against the same five competency areas, so it becomes easier to see the differences behind similar-looking presentations.

Ownership matters most when the business system and the web channel need to share product and order information. A walkthrough of how e-commerce and the business system work together can help you pin down which flows you need to describe before deciding.

We deliver websites, product websites, PIM and portals with long-term product ownership, so you can start with one solution and keep building on the same foundation. You can start with a free 72-hour analysis: you set aside two hours, and within 72 hours you get an inventory of functions and integrations, a prototype and a cost estimate.