All case studies

Use case, Planning firm, 2026

Tenders at planning firms: 30 minutes of review instead of 10 hours of bid work a day

How an agent tuned to a specific firm's profile can cut ten staff hours of daily bid work down to thirty minutes of review, without relying on generic tender AI. A use case study based on our production-ready PoC.

Auf Deutsch lesen

Potential effect

20×

Less time on bid work (projected from the firm owner's baseline)

Industry
Planning firms (architecture and engineering services), mid-market from about 50 employees
Scope
Fully built PoC; projection based on the baseline figures the firm provided
Technologies used
Claude Sonnet, Next.js, Postgres, TED, DÖE

Starting point

Several staff work on tenders every day: they screen notices on TED and DÖE, sort them, write bid documents, pick the right people by competency and contact partner firms. Filtering by keywords alone returns hit lists that match thousands of firms at once and none of them really.

What we built

The agent doesn't rely on a generic keyword filter; it works with the firm's own profile: its references, its staff's competencies, its regular partner firms and the target contract sizes it has set for each practice area. Claude Sonnet sorts incoming notices from TED and DÖE against this profile and drafts the bid document in Word automatically.

Focus

  • Tuned to your firm profile

    A tender AI that doesn't know your firm can only return general hits. We index your reference list, your staff's skills, your regular partner firms and the contract sizes that make economic sense for you. The results then show what your firm can actually deliver, not what fits ten thousand other firms just as well.

  • From screening to bid presentation

    The agent handles the entire bid workflow: it filters notices from TED and DÖE, sorts them, drafts the Word document, finds matching staff in your personnel database and suggests partner firms. If a bid is shortlisted, the same agent creates the presentation on the same principle.

  • Compliance stays in your firm

    We don't build AI that makes procurement-law judgments. The agent only does the groundwork: screening, sorting, drafting, matching staff by competency. The legally compliant finalization (self-declarations such as the ESPD, procedural compliance) stays in your existing workflow. VgV, GWB and HOAI risk stays in your hands.

Why tender AI needs to know your firm profile.

Tender software has been around for years. What's been missing is tender AI that understands what your firm stands for. A plain keyword filter doesn't know your firm profile. It returns a hit list that fits thousands of firms at once and none of them really well. Bid managers then end up sorting through the notices themselves again. In our use case study for upper-mid-market planning firms, we tuned the agent to the firm's own profile instead: references, staff competencies, regular partner firms and target contract sizes per practice area. Projected, ten staff hours of bid work per day become 30 minutes of review, a factor of 20. That's why we don't recommend an off-the-shelf tool. We built exactly this approach as a working proof of concept for a planning firm: we indexed the reference list and the staff database, ran the classification prompt against live notices and matched the pre-selection to the firm's own services. The firm never commissioned the full build. The prototype works, and it is the starting point for the next firm.

The fair concern

Another tender filter that throws out everything above €5 million without ever checking whether your firm profile actually fits.

The industry pattern

Mid-sized planning firms track public tenders every day on TED and DÖE, Germany's central notice service. Their bid process has grown over the years. Typically, several staff spend two hours each per day on different steps: screening and sorting notices, writing bid documents, picking staff from the personnel database by competency, contacting partner firms.

That adds up to about 10 hours a day, more than one full-time role, spent on bidding instead of planning.

A filter that only searches for keywords doesn't know what a firm actually builds. So it returns lots of notices that don't fit, and a person has to sort them out again.

Architecture

The agent doesn't use a generic keyword filter; it works with the firm's own profile. For that, we index:

  1. Indexing references: the firm's references are tagged by practice area, contract size, project phase and regional distribution.
  2. Capturing competencies: staff competencies come from the personnel database: specialties, software knowledge, certifications, typical project sizes.
  3. Partner firms: regular partners are tagged by practice area, along with the track record of past collaboration.
  4. Target contract sizes: for each practice area, the firm sets the contract size below which a bid isn't worth pursuing.

The agent (Claude Sonnet) sorts incoming notices from TED and DÖE against this firm profile. Hits with a high rating go straight into automatic Word drafting. Hits with a medium rating land on a review list for a single bid manager. If a bid moves on to the next round, the same agent creates the presentation using the same criteria from the firm profile.

Compliance: the agent only does the groundwork. Self-declarations such as the ESPD (European Single Procurement Document) and running the procedure in a legally compliant way stay in the firm's existing workflow. VgV, GWB and HOAI risk remains entirely in the firm's hands.

Model calculation

The projection starts from the baseline the firm owner stated (10 staff hours/day on bid work) and is validated against the TED and DÖE notices the PoC processed:

  • 10 staff hours/day → 30 minutes of review in bid work (factor of 20)
  • Built into the architecture: once the sorting bottleneck is gone, the firm can handle more bids without adding staff.
  • Built into the architecture: competency matching becomes more consistent because the agent uses an indexed staff database instead of manual assignment.

The time freed up goes back into actual planning work. More than one full-time role moves from bid work into planning, with the same people.

Methodology in detail

To keep the factor of 20 traceable, here is the math for a generic mid-market planning firm. It builds on the firm owner's baseline and the target processing time we validated in the fully built PoC.

Typical industry assumptions

  • TED and DÖE volume, Germany: we assume 800–1,200 relevant notices per week across all practice areas. TED is the EU-wide platform; DÖE (Datenservice Öffentlicher Einkauf) bundles notices from German federal, state and local authorities.
  • Hit rate after indexing the firm profile: typically 0.5–2 % of incoming notices are real matches for a specialized firm. The rest fits the practice area, but not the specific profile.
  • Manual screening effort without pre-sorting: roughly 3 min per notice × 200 notices/day = 600 min/day = 10 staff hours/day (baseline stated by the firm owner).
  • HOAI fee zones: specialized firms are mostly interested in contracts in the upper fee zones IV and V, where fee rates are higher.
  • Procurement law: self-declarations such as the ESPD stay in the firm's existing workflow. The AI makes no procurement-law assessment. VgV, GWB and HOAI risk stays in the firm's hands.

The math

  • Factor of 20: 600 min/day (baseline stated by the firm owner) / 30 min/day (target processing time validated in the PoC for reviewing the filtered top hits) = 20×. We measured the target time in the fully built PoC against real TED and DÖE notices; it is not modelled.
  • In headcount terms: at 8 hours per working day, 10 staff hours equal roughly 1.25 full-time roles in bid work. Bid work earns no fees, planning does. Every hour that moves from bidding into planning becomes billable.

What moves the numbers

  • In smaller firms (under 50 staff), the absolute number of hours saved shrinks. The factor stays, but the business case for the pilot looks different.
  • In highly specialized firms (e.g. structural engineering only, HOAI zone V), typical TED volume allows even more selective filtering. The factor can rise to 30–40×, because the hit rate is below 1 % and manual screening is correspondingly less efficient.

The key takeaway

A tender AI that knows nothing about the firm can only return general hits. What matters is not the model but indexing the firm profile. Without that step, even a strong model like Claude Sonnet only returns average hits.

That is exactly what we built: a working PoC that indexes the firm profile and runs against live TED and DÖE notices. The full build was never commissioned, but the prototype runs. For the next firm, it is the starting point for a PoC sprint, not a concept on paper.

How to start

The fastest way to apply this to your company is a PoC sprint on your real data. It runs within days.

  • PoC Sprint

    Days to two weeks

    You have an idea and want to know whether it holds up before you invest.

    Running prototype, demo, architecture note, build plan

  • Product Build

    Weeks, not quarters

    An idea or a prototype needs to become something that runs day to day.

    Running system, design, cloud setup, documentation, handover

Compliance

We don't build AI that makes procurement-law judgments. The agent only does the groundwork: screening, sorting, drafting, matching staff by competency. The legally compliant finalization (self-declarations such as the ESPD, procedural compliance) stays in your existing workflow, with your legal counsel where needed. That keeps VgV, GWB and HOAI risk entirely in your hands. We host in EU regions, sign the DPA before the project starts and can run the system on your own servers if your firm's data protection concept requires it.

FAQ

Which tender platforms do you integrate with?

By default, TED (Tenders Electronic Daily), the EU-wide platform, and the Datenservice Öffentlicher Einkauf (DÖE), which bundles notices from German federal, state and local authorities. Together they bring most of the notices relevant to planning firms in Germany into one place. We connect further platforms such as evergabe.de, Vergabe24, subreport or DTAD if your firm pursues tenders from those sources. We check the specific integration on day one of the PoC sprint.

How is this different from generic tender tools?

A tool that doesn't know your firm can only filter public notices by keywords. The result is lots of hits that don't fit: too small, too large, wrong practice area, no matching competencies. We take a different approach. In the PoC sprint, we index your reference list, your staff competencies, your regular partner firms and your target contract sizes. The agent sorts incoming notices against this firm profile and returns a short, filtered hit list instead of an unsorted stream.

How quickly do we see results?

The PoC sprint takes a few days to two weeks: the agent sorts live notices from TED and DÖE, matched against your firm profile. In the full build, automatic drafting and competency matching from the staff database follow within a few weeks. Bid managers see the effect from the first round of notices after launch. Projected, the daily effort drops from 10 staff hours to 30 minutes of review, a factor of 20.

Who keeps the system tuned to our firm profile?

For the first six months, we do: once a month we tune the prompts based on your feedback on the hits. After that, we hand the tools over to your bid team: an interface for tagging references, a way to maintain skills in the staff database and a versioned prompt editor. If you like, we stay on as your point of contact under a maintenance contract. For ongoing strategic guidance, there's our Fractional CTO offer.

How large does our firm need to be for this system?

The system makes economic sense from around 50 employees and about 30 bids per year. For firms with 20 or fewer staff, the internal TED filter is often enough. From 150 employees, the effect gets particularly large: a projected 1.5–2.5 full-time roles freed up in bid work, which move into actual planning work instead of being cut.

The same for your company?

The fastest way to find out is a PoC sprint on your own data.

Let's talk.

A free 30-minute intro call with no strings attached. We listen and tell you straight whether and how we can help.

On working days we usually reply within 24 hours.