Product Stewardship and Substance Compliance Software: Build vs Buy

What the off-the-shelf products actually do well

Buy, and buy the campaign capability even if you eventually build. Supplier chasing is a relationship business, and the platforms ask on behalf of many customers at once, which gets answers your own compliance manager will not.

Assent is the strongest of these at engaging a supply base and keeping declarations flowing. iPoint and Sphera come from the manufacturing compliance world and carry the regime knowledge that goes with it. Source Intelligence does campaign work at scale. Compliance and Risks, through C2P, is a content subscription rather than a data platform and does list monitoring properly. SiliconExpert and Z2Data are worth naming separately, because component intelligence on electronic parts is a different asset and hard to reproduce.

All of them monitor the regulatory lists, which is one of the better reasons to hold a subscription regardless of what else you own. The candidate list moves without any change on your side, and somebody watching it professionally is worth paying for.

If you carry a few hundred purchased parts across one product line into one regulated market, subscribe and stop reading. A build there will cost more than the licence and deliver less, and we say that to manufacturers with our own proposal on the table.

The rest of this page is about what a subscription cannot reach, which is your parts.

Where they stop: the rollup that lives in one person's workbook

A Tuesday email from a customer's procurement portal asks for a full material disclosure for assembly 4471-B, plus a substance statement, plus confirmation on two newly listed substances, by Friday.

The compliance manager opens the shared drive. Around nine hundred supplier declarations sit in it. Perhaps three hundred are more than three years old. Sixty are scans of scans. A dozen are your own template with not applicable written across every row. The structure for 4471-B changed twice this year because purchasing qualified an alternate connector, and the alternate has no declaration.

By Thursday afternoon the answer that goes back is a statement of belief, signed by somebody whose job depends on it being true, based on data nobody can fully evidence. If that statement is wrong, the first consequence is not a fine. It is a customer removing your part from their approved list, which means a product line loses a market.

The platforms are good at the campaign side. Where they are weaker is your parts. Their data model is theirs, your part numbering and approved vendor list logic is yours, and matching a supplier document referencing their part number to your internal number across revisions, alternates and manufacturer part numbers is where accuracy quietly disappears. That matching is the work being done by hand at six in the evening.

The second gap is that rollup logic is not arithmetic. Phantom assemblies exist in the bill of materials and not in the physical product. Alternates mean the part in the box depends on what purchasing had in stock that week. Process chemicals such as plating, flux, adhesive and marking ink are physically present and frequently absent from the structure entirely. Configurable products only resolve at order time.

Then there is threshold logic, which no vendor default should be deciding for you. Restrictions under the electronics substance directive apply at homogeneous material level rather than part level, so a declaration giving a percentage against a whole component is not directly usable. Article obligations under the European chemicals regulation use a threshold by weight, and how it propagates through an assembly is a question for your regulatory counsel about your products.

The arithmetic: per part and per campaign pricing versus a build

These platforms price on supplier count, part count, campaign volume or a blend, usually with a content subscription underneath. That model is fair for what it delivers, which is why the licence line rarely decides this.

Here is the crossover, stated in purchased parts. Below roughly 500 purchased parts with one product line and one regime, subscribe. Between 500 and 3,000 parts, subscribe and accept that one workbook will survive, because the rollup will still be assembled by hand. Above roughly 3,000 to 5,000 purchased parts, or three or more regimes, or configurable products where the answer depends on the configuration ordered, the subscription stops being the constraint and manual matching becomes it.

Then price the two things that appear on no invoice. First, the compliance manager's week: how much goes to opening documents, matching part strings and rebuilding customer packs, valued as loaded salary across a year. Second, the backup problem. If one person maintains the rollup workbook with no realistic replacement, price the fortnight when they are unavailable and a disclosure request arrives.

Then price the risk you cannot insure: a customer removing your part from their approved list. Do not invent a probability. Note only that the exposure is a market rather than a penalty.

What a custom build actually costs

Bands, from Digital Heroes delivery experience. A first release runs $60,000 to $140,000 over 12 to 16 weeks, covering declaration intake with extraction and part matching, the part level substance data model, integration to your bill of materials in the enterprise and engineering systems, and automated rollup producing a computed basis for each product. A full platform adding supplier campaign management, list change monitoring with re-evaluation, regulatory database and minerals template outputs, customer specific export templates and a supplier facing portal runs $180,000 to $400,000 phased across 6 to 12 months.

Data migration adds 10 to 25 percent, and the variable is historical backfill: how many years of declarations you want extracted and matched rather than starting clean. Extraction is the honest use of machine assistance here, reading the substance table, scope, signatory and date out of a hundred layouts. Every value needs a provenance link back to its page, and a human confirms before it counts.

Year two runs 15 to 20 percent of build cost annually. Lists move, exemptions expire on published schedules, a new regime arrives, and each new division brings its own part numbering.

What pushes the number up: the number of regulatory regimes, since each carries its own threshold logic, exemption handling and output format. Bill of materials complexity, especially configurable products and heavy use of alternates. Engineering system integration, which varies enormously between Teamcenter, Windchill, Arena and a home grown structure database. And multi division rollout where each division numbers parts differently and insists it is correct.

What keeps it down: one product line, one regime, and your top 500 purchased parts by spend. That covers most customer requests and proves extraction accuracy before you scale it.

The four situations where building wins

• Regulatory fit. The formats are published and machine readable, which is why owning the pipeline pays. IPC-1752A defines a declaration format and IEC 62474 defines a substance data exchange, and a minority of your suppliers will use them. Minerals reporting runs on the industry template. Notification obligations for articles containing listed substances placed on the European market have their own database and submission shape. Each is one internal data set rendered into a different output, which creates no value beyond keeping accounts open, and is exactly what should be automated.
• Scale economics. Not seats but parts. Once the manual matching of supplier part strings to your approved vendor list consumes more compliance hours than the subscription costs, you are paying twice for one capability and getting the weaker half by hand.
• A workflow that is your competitive advantage. If customers keep you on their approved list partly because your disclosures arrive complete and on time, the rollup engine is the business. The version that protects the relationship best is statement versioning: keep every statement you have issued as a versioned object with the data it was computed from, so when a list updates you can tell a customer their statement is now stale before their auditor does.
• Integration sprawl across three or more systems. The enterprise system holding purchasing and bills of materials, the engineering system holding structures, the shared drive of declarations, the rollup workbook and the subscription platform. Not one of them holds the join between a structure revision, the supplier part actually bought against each line, the declaration covering that part on that date, and the statement you issued last quarter. That join is the entire job.

Two of those true is a build, and the hybrid is common: subscribe for campaigns and list monitoring, build the matching, rollup and output layer.

How to decide in a week

Take one assembly and try to produce a full material disclosure with a computed basis. Not a compliance statement. A disclosure, with every contributing part named.

Monday: choose an assembly a real customer has asked about. Pull its current structure revision from the engineering system and the as bought supplier parts from purchasing. Note where the two disagree.

Tuesday and Wednesday: for every line, find the declaration that covers the part actually bought. Record how long each takes, and mark each line covered, assumed or unknown. Include process chemicals, and note whether they appear in the structure at all.

Thursday: check the thresholds. For the lines you have data on, confirm whether the percentage you hold is against the whole part or against a homogeneous material, and whether that is the basis the regime requires. Count the lines where you cannot tell.

Friday: put three numbers on a page. Hours spent, the share of lines genuinely covered by a current declaration matched to the part you actually buy, and the count of alternates with no declaration at all. Above eighty percent coverage in under a day means your subscription and your process are working. Under half, with alternates uncovered and process chemicals invisible, means the statement you send customers is a belief rather than a computation, and no amount of supplier chasing changes that, because the failure is in the join.

What follows is a paid discovery phase rather than a proposal. Two to three weeks, fixed fee, producing a signed product requirements document covering the data model with structure revisions and alternates, the threshold basis per regime, the extraction and matching design and acceptance criteria. You own that specification whoever builds it, and you can hand it to three firms and finally get comparable quotes.

Who we are wrong for: manufacturers with one product line and a few hundred parts, anyone shopping purely on hourly rate, and anyone who wants software before their engineering and purchasing systems agree on what a part is. Digital Heroes writes that requirements document before any code, with more than fifty specialists and India LLP, US LLC and UK LTD entities so intellectual property assigns under your own law. ShopScore, HeroCheckout and Section Vault are our own products, over 2,000 projects sit behind us, and you meet the named team before signing. We are listed on Clutch, Trustpilot, Fiverr Vetted Pro and D-U-N-S.

Book a 30-minute call with Digital Heroes and get a written plan and a fixed quote within 48 hours.