Tender services compared: DTAD, Patterno and TenderScope

A tender service watches the portals where public authorities and publicly owned companies announce their contracts, and tells you about the tenders that fit your business. Without one, that means dozens of portals, every day, by hand.

This page puts three providers side by side. DTAD is the established provider, on the market for over 25 years by its own account. Patterno is an AI service with sector packages. TenderScope is our own service, which scores tenders from 27 countries against your company profile. About the other two, this page only says what they publish themselves, quoted and with the source.

The providers at a glance

Quotations translated from the German pages of dtad.com including its product page and of patterno.de including its product page, retrieved on 24 and 25 August 2026. "Not stated" means these pages say nothing about it, not that the feature does not exist. None of the three providers publishes prices.

The providers in detail

DTAD

DTAD is a tender database with search and daily notification, on the market for "over 25 years" and used by "more than 5,500 companies" by its own account. The platform bundles "over 12,000 sources" and advertises "99.99 % of all publicly published tenders" for Germany. The search is "AI-supported" with "semantic understanding", and there is a chat assistant. Tender documents "can be retrieved directly", and the assistant "reads documents, extracts suitability and award criteria".

What stands out is the scope beyond the bare notice: DTAD offers "planned construction projects", "historical award data", the "automatic monitoring of expiring fixed-term contracts" and answers the question "Which contracts did your competitor win most recently?". Matches come from search profiles with "daily notification", and "the search learns". For the construction sector this is a product of its own with a data set built over 25 years, and if construction projects are your core business, that depth is an argument.

Patterno

Patterno is an AI service with "100+ teams" as customers by its own account. It checks "every tender individually by AI against your profile" and advertises "not just vector similarity but genuine content assessment". It bundles "4,500+ portals" in 27 EU countries plus Germany and Austria. Tender documents are evaluated too, with "100+ pages in ~30 sec.", and onboarding is built for speed: "<5 min to the first tender".

Added to that are award data ("See which organisations won which tenders") and sector packages for pharma and construction. If your sector is covered there, that is an advantage over a generic profile.

TenderScope

TenderScope is an AI service for public tenders. You set up a company profile once: what you offer, where you deliver, for whom, and from which contract value. From then on TenderScope scores every new notice from 27 countries against that profile and emails you in the morning the ones you would bid on. The dashboard holds the tender documents, the checked exclusion criteria and a pipeline for everything you pursue.

The profile does not stand still. During the first three weeks you rate borderline cases, and your threshold settles from that. After that, every match you reject with a reason becomes a suggestion for how your profile should narrow. And your service catalogue is read by the assessment as a document, instead of you translating it into keywords.

Five things we do differently from the other two, and they are in the next section: We show you fewer matches. The profile learns from your feedback. We check the documents for exclusion grounds before you invest time. We find contracts before they are tendered. And we disclose what actually arrives.

What TenderScope does differently

Fewer matches that are right

Any system that assigns tenders by topical similarity produces the same kind of false match: formally plausible, practically worthless. An institution of this kind typically needs this kind of technology, so it might fit. Such reasoning sounds right, and that is exactly why it slips through.

That is the point where a tender service stops helping. Whoever goes through twenty matches every morning, two of which are serious, has not handed the work off but moved it.

With us, the assessment has to answer a question that cannot be paraphrased: Would you bid on this as the main contractor? Reasoning based on assumptions is thrown out mechanically, even when it sounds plausible.

So we show you less, and that is the service, not the shortfall.

A profile that learns

A company profile is a guess on day one. That is why every account starts with a calibration: for three weeks we deliberately show you borderline cases too, you rate them with one click, and from that the threshold settles at which a match reaches you. The stock of matches typically shrinks to a quarter in this period, and that is intended.

The learning does not stop afterwards. When you reject a match, we ask for the reason. From those reasons, concrete changes to your profile appear under "Suggestions from your feedback", such as a minimum contract value or an exclusion term. You accept them with one click or leave them. The profile narrows without you having to maintain it.

And whoever can describe what they offer does not have to translate it into keywords: you upload your service catalogue as a PDF or Word file, and the assessment reads it along, in the bidder test, in the exclusion criteria and in the offer draft.

DTAD also writes that "the search learns and gets better and better over time". What sits behind it with us are the three steps above, and you can see each of them in the dashboard.

You know before the effort whether you are excluded

For every match we fetch the tender documents and read them. What counts there is not the service description but the exclusion criteria: required certificates, minimum turnover, reference conditions, prequalifications. You get them quoted verbatim with the passage, before you invest the first day of bid work.

The other two evaluate documents as well; DTAD's assistant extracts suitability and award criteria. The difference is the conclusion drawn from it: we hold the criteria against your profile and your service catalogue and tell you "you lack the required certification here, this will not work" before you find out yourself.

You see the contract before it is tendered

We hold 8,375 signals of planned procurements: prior information notices and market consultations from contracting authorities, as well as investment plans we extract from municipal budgets. A municipal budget is fixed months to years before the first notice, it is public, and it sits as several hundred pages in the council information system.

That changes the moment you enter the race. When the notice appears, everyone knows it, and the specification is often modelled on the incumbent. Whoever knows beforehand that something is coming can talk to the contracting authority while the specification is still open.

DTAD offers something comparable with construction projects and the "automatic monitoring of expiring fixed-term contracts". The difference with us: all sectors, and every signal is scored against your profile like a tender. You do not search a second database; it is in the same email.

With the Pre-Tender Radar come 21,750 running contracts with a recorded end date, computed from three years of award notices, plus positions from procurement plans. With that you see who delivers today and when their contract runs out.

You ask the right questions before the deadline passes

From the tender documents, a draft of your bidder questions to the contracting authority emerges: contradictions in the specification, unclear requirements, criteria that could exclude you. The deadline for bidder questions regularly passes before anyone has really read the documents.

And you can check whether we deliver

The above is worth nothing if the tender never arrives in the first place. That is why we publish what actually comes in, per country, computed from our own corpus and refreshed monthly: Coverage.

There is no portal count on that page. A portal count is the wrong figure, because we connect to the statutory publication layers rather than to individual portals. In Austria, a single connection delivers ANKÖ, VEMAP, PROVIA, wien.gv.at, ausschreibung.at, the Vorarlberg notices and more in one stream. So what gets counted is what comes out at the end: contracting authorities and notices.

What TenderScope is not built for

A comparison written by a provider is only worth as much as what it leaves out. Three things we deliberately do not do:

No filter sets to maintain yourself. You set up a company profile, and every notice is scored against it. That is a decision against CPV filters, because CPV codes are unreliable for composite services: the same service appears under software one time, under engineering services the next, under electrical installation the third. A filter on the wrong code loses it silently.

No percentage promise on completeness. We could name a figure like others do. Instead we publish the measured one, so that you can check it. Whoever prefers a round percentage to a verifiable number is in the wrong place with us.

No AVA software. TenderScope finds, scores and supports the bid. Producing tender documents, as AVA software for contracting authorities does, is not part of it.

Which service suits whom

DTAD, if construction projects are your core business and you want a dedicated product with 25 years of data for that, or if you want to work with saved search profiles and daily notifications.

Patterno, if your sector is pharma or construction and you benefit from its sector packages there.

TenderScope, if you prefer a few checked matches to many plausible ones, want to know before the first day of bid work whether you are excluded, want to see contracts before they are tendered, or want a profile that learns from your feedback instead of being maintained. And if you bid in several European markets and want to check what a provider actually delivers there.

The easiest way to check

Tell us what you bid on. We show you five matching tenders from the live corpus, without an account and at no cost. If none of them fits, you know it in five minutes instead of after a trial period.

See five matching tenders