You have two tabs open. One is a Fiverr gig offering a Wikipedia page for $300. The other is an agency quoting six times that. And somewhere in Slack, someone has volunteered the comms team to "just write it ourselves."
All three routes can produce a draft. The differences show up later — at review, at the first hostile edit, in month seven when a volunteer editor nominates the page for deletion. So this comparison runs the math symmetrically, including the two numbers vendor marketing leaves out: the probability each route fails, and what cleanup costs when it does.
One spoiler up front: the freelancer is sometimes the right answer. In-house almost never is — for creation. We will be specific about both.
TL;DR
- The question that decides the route is who is accountable after publication — not price. A Wikipedia page faces its real test weeks or months after going live.
- Every route carries the same disclosure obligation. Wikipedia's paid-contribution policy binds anyone compensated for edits — agency, gig seller, or your own employee. On the $300 route the disclosure usually just never happens, which is itself the violation.
- A $300 gig is not a discount version of a EUR 1,930 engagement. It is a different product: no notability assessment, no source map, no disclosure, no defense, no recourse.
- In our 2024 observation window, 95% of Fiverr-built pages were deleted within 90 days. On pages we accept after a source assessment, our verified pass rate is 93%. The difference is selection plus process, not magic.
- Use the Stakes Grid — notability strength × cost of failure — before choosing. One quadrant tolerates a vetted freelancer, one demands an agency, one says audit first, and one says do not build yet.
- Your in-house team should almost never create the page. It is often the best monitoring input the page will ever have.
The three routes, and the question that decides
The agency route. A firm that does Wikipedia work as its core business: notability assessment first, source mapping, neutral drafting, disclosed submission, monitoring, deletion defense. Cost for an English company page sits around EUR 1,930 — the pricing guide has the full table by language edition.
The freelancer route. One person, hired on a gig platform or directly. Price anywhere from $150 to $2,000. Capability ranges from "experienced editor in good standing who declines weak cases" to "account created last Tuesday, running a template."
The in-house route. Your comms or content team writes and publishes the page themselves. The cost looks like zero. It is not zero.
Most buyers compare these on price and writing quality. Both are the wrong axis — competent writers exist on every route. The real differentiator is accountability after publication: when the page is tagged, edited hostilely, or nominated for deletion, who is obligated to show up, and what do they lose if they don't?
Ask that question and the market reorganizes itself. The agency has a contract, a refund clause, and a reputation to protect. The gig seller has a five-star rating on a platform Wikipedia editors actively watch. Your in-house team has a conflict of interest written into their employment contracts.
What you're buying at each price point
The $300 gig and the EUR 1,930 engagement are not the same product at different margins. The anatomy of each:
The $300 gig, typically:
- A draft, often template-built, increasingly AI-generated
- Sources pulled from whatever Google returns — press releases included
- No notability assessment — every order gets accepted, because declined orders cost a gig seller their ranking
- Publication from an undisclosed account, frequently new, occasionally already blocked
- No disclosure on the article — which makes the edit a policy violation from minute one
- No monitoring, no defense, no refund that covers consequences
The EUR 1,930 engagement, typically (ours is described in full at Wikipedia page creation):
- A notability assessment before drafting money changes hands — with "no" as a possible and regular answer
- A written source map: which coverage counts, which does not, and what each source can support
- Neutral drafting against Wikipedia's content policies
- Disclosed, policy-compliant submission
- 90 days of monitoring, deletion defense when needed, and an 80% refund if a published page cannot be defended after three attempts
The price difference is not the writing. It is everything around the writing — and most of that "everything" exists to move one number: the probability the page is still standing a year later.
The Stakes Grid
Place yourself on two axes. First, notability strength: is your independent coverage deep and obvious, or borderline? Second, cost of failure: if the page is deleted, tagged as undisclosed paid editing, or written up in a "brand caught astroturfing Wikipedia" piece, what does that cost you?
| Low cost of failure | High cost of failure | |
|---|---|---|
| Strong notability | Freelancer territory. A vetted, disclosed freelancer can do this job. The case is hard to lose; you are paying for competent execution. | Agency territory. The case is winnable, but you need contractual aftercare, defense capacity, and recourse. Funding rounds, IPO windows, and regulated industries live here. |
| Borderline notability | Audit first. Do not buy creation from anyone yet. Buy an honest read — a notability audit — and consider smaller moves before a full page. | Do not build yet. Weak sources plus high visibility is how brands end up with a deleted page and a public paper trail. Build coverage first; the page comes later. |
The grid makes two things obvious. The freelancer is only right in one quadrant — and only the disclosed, vetted kind. And one quadrant's correct vendor is no vendor at all: there, every route fails, just at different prices.
Head-to-head: the seven rows that matter
| Agency | Gig-platform freelancer | In-house | |
|---|---|---|---|
| WP:PAID disclosure | Contractual, on-wiki, verifiable | Rarely offered; gray sellers refuse to name their accounts | Required but routinely missed — employees count as paid editors |
| Whose account history is at risk | The agency's editors — years of standing they will protect | A disposable account; the seller loses nothing | Your staff's accounts and your office IP range, traceable to you |
| Draft quality | Specialist, policy-shaped | Anywhere from competent to template spam | Usually well-written, almost always promotional in tone |
| Deletion-defense capacity | Knows the AfD process, argues from sources | Gone — the gig closed weeks ago | Cannot argue without deepening the COI exposure |
| Aftercare | Defined monitoring window, then optional support plans | None | Accidental — whoever remembers to look |
| Total cost incl. failure | EUR 1,930 with bounded downside: 93% pass rate on accepted cases, refund clause | $300 plus a ~95% deletion probability plus a rebuild from a worse position | "Free" plus a high tag-and-revert probability — and the cleanup is also you |
| Recourse if it fails | Refund clause, contract, a firm with a reputation to defend | A platform dispute over $300 | None — you cannot file a complaint against yourself |
The table is the article. Everything below is the mechanics behind the three columns.
The gig-platform trap, mechanically
The buyer confusion is real — there are entire Quora threads asking whether Fiverr or Upwork is the better place to order a Wikipedia page. It is the wrong question, and here is the mechanism.
Wikipedia's volunteer community treats undisclosed paid editing as a threat to the project, and acts on it. As Reputation X notes, experienced editors monitor the gig listings on these platforms, and hiring through them can end in the page deleted and the editor banned. The listings are public, and a new page on a commercial subject from a fresh account is precisely the pattern Wikipedia's patrol tools are built to catch.
Now look at the incentive split. If the page is deleted and the account blocked, the seller loses a $300 order and registers a new account tomorrow. You lose the page, possibly the title — repeatedly recreated pages get protected against recreation — and the clean version of your story, because the next attempt happens with a documented violation history attached to your brand. The seller's downside is an inconvenience. Yours is structural.
And the disclosure obligation never went away. WP:PAID requires anyone compensated for edits to disclose their employer, client, and affiliation. Gray-market sellers will not — Reputation X observes they typically refuse to share their usernames at all, because disclosure would end the business model. Which means the $300 includes a policy violation committed on your behalf, with your company as the named beneficiary.
The freelancer who is worth it
Legitimate freelance Wikipedia editors exist, and in the strong-notability, low-stakes quadrant they are a rational choice. You recognize one by the signals a gig listing cannot fake:
- A disclosed username, offered without being chased, with paid contributions declared on their user page.
- A visible edit history — years of contributions across many topics, not a cluster of corporate pages created in the last 90 days.
- They decline bad cases. A freelancer who says "your sources don't support a page yet" is showing you their selection filter — the same filter that keeps their pass rate, and their account, alive.
- They price like professionals. Accountable independent editors are not cheap; Reputation X notes some charge $10,000 or more for a single edit at the top end. Honest freelance pricing lands far closer to agency pricing than to gig pricing, because the work is the same work.
What the legitimate freelancer still cannot give you: defense capacity in month seven (one person, other clients, no contract), structured aftercare, or recourse. You are buying competent creation, not a service relationship. In the right quadrant, that is enough.
In-house, honestly
For page creation, in-house is the weakest route — not because your team writes badly, but because Wikipedia's conflict-of-interest rules are aimed squarely at it. Employees editing the employer's article are paid editors under WP:PAID. The tone drifts promotional, reviewers spot it quickly, and the edit history traces to you permanently. Full mechanics in why you shouldn't edit your own Wikipedia page.
But in-house has a real, underused role on the other side of publication:
- Source gathering. Nobody assembles your coverage history faster than your own team, and a clean source list cuts any vendor's assessment time.
- Edit requests. Wikipedia supports COI-compliant participation: factual corrections proposed on the article's talk page, declared and specific, get reviewed and applied by neutral editors.
- Monitoring inputs. Your team notices a hostile edit or an outdated figure within hours; a professional monitoring service responds to it properly. The combination beats either alone.
The line to hold: in-house contributes inputs and watches outputs. It does not touch the article directly. The moment someone on payroll edits the page itself, "free" converts into the most expensive route on this page.
The failure math, worked
Take a strong-notability company deciding between the $300 gig and the EUR 1,930 engagement, and price both symmetrically.
Route one: the gig. In our 2024 observation window, 95% of Fiverr-built pages were deleted within 90 days. So the realistic cost of the $300 route is $300 plus a roughly 95% chance you then need the page built properly anyway — except a rebuild after deletion costs more than building right once: the title carries deletion history, reviewers look harder, recovery adds weeks. Expected total: somewhere past EUR 2,200 and a quarter lost, to arrive — at best — where EUR 1,930 would have placed you directly. That is the friendly scenario. The unfriendly one adds an undisclosed-paid-editing finding with your company named in it.
Route two: once, properly. EUR 1,930, a 93% verified pass rate on cases accepted after source assessment, 90 days of monitoring, and a bounded downside: 80% refunded if the page cannot be defended after three attempts. The five-year picture, including maintenance, is broken down in how much a Wikipedia page costs.
Read that way, the gig is not cheaper. It is roughly the same money carrying a 95% surcharge probability — and a worse starting position for the rebuild.
Recourse: who answers in month seven
Deletion nominations do not arrive on publication day. They arrive when a patrolling editor crosses the page — week three, month seven, year two. The discussion then runs about a week, and arguing it well requires knowing the source base, the notability guidelines, and the register Wikipedia editors respond to.
Month seven is past every gig platform's dispute window; the seller is unreachable. Your in-house team can read the discussion but cannot safely argue in it. The agency is the only route where someone is contractually pointed at exactly this moment — which is why aftercare, not drafting, is most of what the price difference buys.
The decision checklist
Four steps, in order:
- Place yourself on the Stakes Grid. Borderline sources mean audit, not creation. High stakes mean agency. Strong case at low stakes — a vetted freelancer is honestly fine.
- Apply the disclosure test to any vendor. Whoever edits must disclose under WP:PAID. A vendor who will not name their account has answered your most important question.
- Price the failure, not the invoice. Weight each quote by its survival odds and add cleanup. The lowest invoice rarely wins this math.
- Decide who owns month seven. If the answer is "nobody," you have bought a draft, not a page.
And six questions that expose a gig seller in one email:
- What username will publish the page, and where will your paid-contribution disclosure appear?
- Can you show three pages you created that are still live after 12 months?
- Will you assess notability first — and have you ever declined a paying client?
- Who responds if the page is nominated for deletion six months from now, and at what cost?
- What happens to my page if your account is blocked?
- What exactly does your refund cover if the page is deleted within 90 days?
A legitimate professional answers all six in writing without flinching. A gig seller goes quiet, usually around question one. For the longer vetting playbook, see how to hire a Wikipedia agency.
Download: PDFthe printable vendor scorecard (PDF) — score any agency, freelancer, or internal proposal on the same rows before you commit budget.
If you want the agency column itself compared across real providers — pricing, disclosure practice, monitoring scope, refund terms — we maintain an honest side-by-side at Wikipedia agency comparison, with WikiBusines scored on the same criteria as everyone else.