AI Quoting: How Tradespeople Create Quotes 80% Faster

An electrician from the Eifel region recently showed me his quoting process. Three hours per quote. Five quotes per week. That is 15 hours he does not spend on-site. Not with the customer. Not with his family. Instead, he sits at a desk typing material lists, looking up prices, and totalling line items.

I asked myself: does it really have to be this way?

At DHL, Instacart, and Delivery Hero I saw first-hand how companies used structured data and automation to speed up processes by 80 or 90 percent. Not through magic, but through smart systems that handle routine work. And that is exactly what is happening in the trades right now. Few people talk about it because most vendors focus on large corporations.

This article is for trade businesses who want to know what AI can actually do for quoting today. And what it cannot.

What AI Can Do for Quoting

Let us start with what works. AI excels when patterns need to be spotted in large data sets. For quote calculations, that means:

Material recognition from photos. You photograph the room, the job site, the facade. An AI analyses the image and identifies which materials are installed or needed. Tile type, cable gauge, window dimensions. This works surprisingly well for standardised components. For custom builds, AI has its limits, which I will cover below.

Quantity take-off. From floor plans, site photos, or CAD files, AI can calculate areas, lengths, and quantities. Anyone who has manually worked out the wall area of an L-shaped room with three windows and a sloped ceiling knows how error-prone and time-consuming that is.

Price matching against material databases. Once the material list is ready, an AI system can pull current prices from suppliers. No more manual searches through catalogues or websites. Prices flow straight into the calculation.

Where the Limits Are

AI does not replace the expertise of a master tradesperson. It cannot assess building fabric beyond what is visible in a photo. It does not know the plaster behind the wallpaper is damp. It does not remember that the customer wanted a specific tile brand last time and the price should reflect that. And it cannot judge whether access is difficult and scaffolding costs apply, or whether the neighbour calls the police if there is noise after 5 pm.

Then there is the human factor. A good tradesperson senses during a conversation whether the customer wants the cheapest option or is willing to pay more for quality. That instinct cannot be automated.

The combination matters: AI does the legwork, the tradesperson does the judgement call. That saves time without sacrificing quality.

Before and After: The Quoting Process

I spoke with several businesses and recorded typical timings. Here is the comparison for a mid-sized quote, for example a bathroom renovation or a residential electrical installation.

Without AI Support (approx. 2.5 to 3 hours)

1. Read customer enquiry, clarify follow-up questions: 20 min.
2. On-site visit, take measurements: 45 min.
3. Research materials and look up prices: 40 min.
4. Calculate quantities, compile line items: 30 min.
5. Format quote in a word processor: 25 min.
6. Proofread, correct, send: 20 min.

That adds up to roughly three hours. And that is optimistic. Many businesses take longer because they are on-site during the day and finish the quote at the kitchen table in the evening. On top of that, when the customer calls back two weeks later wanting a change, half the process starts over.

With AI Support (approx. 30 to 45 minutes)

1. Customer enquiry is automatically pre-structured: 5 min.
2. On-site visit, photos and measurements via app: 30 min.
3. AI generates material list from photos and measurement data: 2 min.
4. Prices are pulled automatically from the database: 1 min.
5. Quote is generated from a template: 2 min.
6. Tradesperson reviews, adjusts, and sends: 10 min.

The on-site visit stays the same. Of course. But everything after it speeds up dramatically. The step where the most time is lost is not the customer visit. It is the desk work that follows.

Checklist: Is My Quoting Process Ready for AI?

Not every business benefits immediately. These questions help you gauge readiness:

1. Do you create more than 3 quotes per week?
2. Do you use recurring material lists or service items?
3. Are your prices available digitally (Excel, ERP, supplier portal)?
4. Do you already use a tablet or smartphone on-site?
5. Do you have a recurring quote type that accounts for 60% or more of your volume?
6. Do you or your staff regularly spend evenings writing quotes?

If you answer yes to four or more, getting started is worthwhile. If you only have one or two yeses, it can still make sense, but the return on investment comes more slowly.

What Tools Are Available?

The market is still young. Many solutions originate in the English-speaking world and do not translate directly to German trade businesses (different standards, materials, and calculation methods). Still, there are approaches that work today.

Trade Software with AI Modules

Programs like Taifun, Craft, or Plancraft increasingly offer AI features. Usually as add-on modules: text recognition for measurements, automatic bills of quantities, price suggestions. The advantage is you stay in your familiar software. The downside is the AI features are often limited and work best within the tool's own data.

Specialised AI Tools

A new generation of tools focuses exclusively on AI-powered quoting. Most use image recognition and language models to turn photos and notes into calculations. Some claim you only need to upload a few pictures and have a finished quote 10 minutes later. In practice, that works for simple, standardised jobs. For more involved projects, the result is a starting point, not a finished product. Some of these tools are still in beta. Test before you buy.

The DIY Approach

For tech-savvy businesses: with a structured Excel template (or better, a simple database) and an AI assistant like ChatGPT, you can already automate parts of the calculation today. For example, you enter your measurement data and the AI suggests material lists and calculates quantities. This is not a turnkey solution, but a first step at zero software cost. I know a tiler from the Eifel region who does exactly this. He has his 20 most common tile formats with prices in a spreadsheet and lets ChatGPT verify the quantity calculations. He says it saves him 30 minutes per quote.

If you want to know what Excel-based processes really cost over time, it is worth looking at the risks too.

Custom Development

If your requirements are specific and no off-the-shelf tool fits, a custom solution can make sense. It sounds more expensive than it is when done right. At Plexito we build exactly these systems: material recognition, price database integration, automated quote generation. Tailored to your business and your region.

Where AI in the Trades Stays Hype

I want to be honest here, because I think too much is promised and too little nuance is offered in the AI debate.

AI does not replace a master tradesperson. Full stop. Anyone who tells you AI will soon handle the entire calculation and the human just clicks "Send" is selling you something. The on-site assessment, the conversation with the customer, the experience of knowing whether a certain material works with a particular building fabric: that stays human. For the foreseeable future.

Fully automatic quotes only work for standard jobs. If you always do the same thing (for example solar panel installations of a certain size), then yes, a lot can be automated. But for custom jobs, heritage renovations, special requests? That needs the expert who corrects the AI suggestions. And let us be honest: the custom jobs are often the most profitable. Exactly where AI is weakest, margins are highest.

Data quality is everything. An AI is only as good as the data it works with. If your material prices are from 2024, your line items are not standardised, and your photos are blurry, then even the best AI will not deliver good results. Garbage in, garbage out. That applies in the trades just as much as it did at Delivery Hero, where I saw the same principle play out across millions of data points.

Be wary of "AI badges" on software. Right now many vendors are jumping on the AI bandwagon. A chatbot in the help section is not AI quoting. Ask specifically: what does the AI in your software do? What data was it trained on? What is the error rate? If the salesperson cannot give you a clear answer, be sceptical.

What You Can Try Next Week

I am a fan of small, concrete steps. Here are three things you can do next week without needing budget approval:

Step 1: Measure your current quoting time. Time your next three quotes. Note how long each step takes, from enquiry to sending. Most businesses have no idea how much time actually goes into quotes. And without that number, you cannot make a meaningful before-and-after comparison.

Step 2: Test image recognition. At your next site visit, photograph systematically: room, wall, connections, existing materials. Upload a photo to ChatGPT or Google Gemini and ask: "What materials do you recognise in this photo? Create a list with estimated quantities." The result will not be perfect. But it gives you a feel for what is possible today.

Step 3: Identify your standard job. Which job type makes up the bulk of your quotes? That is the candidate for your first AI use case. The more standardised the job, the better the automation works.

By the way: you do not have to do this alone. We help trade businesses in the Eifel region and the Rhineland digitise exactly these processes. Not with massive IT projects, but with solutions that fit into your daily workflow. Get in touch.

Start Now, Lead in 6 Months

Most trade businesses in the region are still watching AI from the sidelines. Understandable. There is always enough to do on-site. But that is precisely where the opportunity lies: if you start engaging with AI-powered quoting now, you will have a real edge in six months.

Let us do the maths. If you currently spend 15 hours a week on quotes and save 10 of those, that is 40 hours a month. At an internal rate of 60 euros per hour, that is 2,400 euros a month you can either pocket or reinvest in additional jobs. Over a year, that is nearly 29,000 euros. Even if the saving is only half that, getting started pays off.

Faster quotes also mean something else: the customer gets your quote first. Before the competition. And whoever quotes first statistically wins the job more often. That is no secret; every tradesperson knows it. But with manual quoting, "fast" is hard to pull off when you are on-site at the same time.

This is no longer a future topic. The tools are here. The only question is whether you use them.

If you want to know where AI can make the biggest difference in your business, reach out to us. We will look at your processes and tell you honestly what is worth it and what is not.