BlvkWare.

Quote follow-up automation

Most small businesses lose more work to quotes nobody chased than to quotes that were rejected. This page explains how automated quote follow-up actually works, how to work out whether it pays for your business, and the one constraint that decides whether you can be live this week or next month.

The problem is not the quote, it is the silence after it

A quote gets sent on Tuesday. The customer reads it, thinks "I need to talk to my wife about this", and puts it aside. By Friday it is buried. Nobody rejected anything. The job simply evaporated.

Meanwhile the business owner is on a roof, under a sink, or driving between calls. Chasing quotes is the first thing that falls off the list when work gets busy, which means it fails hardest exactly when there is most to lose.

This is a scheduling problem wearing the costume of a sales problem. The fix is not better salesmanship. It is that the follow-up should not depend on a human remembering it.

Work out whether it pays, before you buy anything

There is no shortage of vendors quoting a recovery percentage. Ignore them: the number depends entirely on your trade, your close rate and your job values, and no vendor knows those. Do the arithmetic instead. It takes two minutes and it is checkable:

That is a ceiling, not a forecast, automated follow-up will not close every quote a human would have. But it tells you the shape of the decision. If a plumbing business sends 40 quotes a month, chases maybe 15, closes a third of what it chases, and averages $600 a job, the 25 unchased quotes represent roughly $5,000 of ceiling per month. Against a $79 kit plus what it costs to build and run, that is not a close call.

Run the same numbers with 6 quotes a month at $200 each and the answer flips. It genuinely is not worth doing. Anyone who tells you otherwise is selling.

What an automated sequence actually does

The mechanism is simpler than the marketing around it suggests.

The important part is not the copy. It is that the sequence stops the moment the customer answers. A follow-up that keeps arriving after someone has replied is the difference between a service and spam, and it is the single most common way these systems are implemented badly.

One-click replies, not "please reply to this email"

Asking a customer to write a reply is asking for effort they will not spend. A better pattern is four buttons, each a signed link the recipient taps once:

• I'm interested: the business gets alerted, chasing stops
• Not right now: sequence stops, the quote stays open for later
• Already booked it: marked won, no further contact
• Not going ahead: marked lost, and the reason is recorded

No account, no login, no app. Practically, this also means the business does not need software that reads its inbox, which is a far heavier thing to build and maintain than most people expect.

Why email, and why that is not a compromise

Text messages feel more immediate, so the instinct is to chase by SMS. There are two reasons not to.

The first is context. The quote was sent by email, with the pricing in it. A follow-up in that same thread lands where the customer can actually re-read what they are deciding on. An SMS makes them go and find it.

The second is the constraint that catches almost everyone. In the United States, sending application-generated SMS to consumers requires A2P 10DLC registration: brand registration, then campaign registration, then carrier vetting. AT&T's manual review alone commonly runs two to four weeks, and full approval across carriers realistically takes three to six weeks. It costs roughly $44 for the brand plus $15 per campaign.

Staying on the right side of the rules

Automated follow-up to people who asked you for a quote is legitimate commercial email, but it is still commercial email:

• Only ever contact people who requested a quote from you. No purchased lists, ever. This is the line between follow-up and spam, and it is not a grey area.
• Honour opt-outs permanently and immediately. If someone unsubscribes, they must drop out of any sequence they are currently in: not merely be skipped next time.
• Identify the business clearly in every message, with a real reply address someone reads.
• Make declining easy. Beyond compliance, a clean "no" is genuinely worth more to you than a fourth unanswered email.

How the quotes get into the system

This is the part that decides whether the system survives contact with a real week. Three workable patterns, in order of how much they ask of you:

1. Manual entry. Twenty seconds per quote: name, email, service, amount. Unglamorous, and completely reliable.
2. Push from your existing software. If you quote through job-management software, it posts each new quote to the follow-up system automatically.
3. Forwarding. You send quotes as normal and copy an address that files them.

Start with the first. A system that requires an integration to be finished before it does anything tends never to start. Twenty seconds a quote is cheap next to a job you would otherwise have lost.

What it will not do

Worth being blunt, because the failure cases are predictable:

• It will not fix a quote that is too expensive, unclear, or slow to arrive. Follow-up amplifies whatever you are already sending.
• It will not sell for you. When a customer says they are interested, a human still has to pick that up.
• It will not work on a bad email address. Whatever captures the address at quoting time is now load-bearing.
• It cannot recover a quote you never recorded.

Build it yourself, from scratch or from a kit

If you want to build this: you need somewhere to store quotes, a scheduler that wakes up daily and finds what is due, an email sender with a verified sending domain (SPF, DKIM and DMARC all passing, or you will land in spam), signed one-click reply URLs, and an opt-out list that is genuinely honoured. It is a weekend for an experienced developer and considerably longer for anyone else, mostly because deliverability, not logic, is where the time goes.

Or start from the design rather than a blank page. The Follow-Up Agent kit is this job specified completely: the sequence and its cadence, the rules for stopping it, the opt-out check before every contact, the records it keeps, its tools in OpenAI, Anthropic and MCP formats, and an acceptance test, all generated from your own answers. It is $79, one-off, and downloads the moment you pay. It does not do the deliverability work above, which is yours whoever builds it. It removes the design. The agent it specifies starts at Draft, where it prepares every message and you send it, and you raise it to acting on its own when you have seen enough.