What is it really like to run a submission cycle while holding a full time job?

Running a query cycle around a full time job is mostly an exercise in protecting one thing: the ability to respond fast when something good happens. The querying itself is not the hard part. Sending ten letters takes an evening. What breaks people is the request that arrives on a Tuesday during quarter close, when the agent wants the full in a format you have not built yet.

So the whole cycle gets designed backward from that moment. Batches stay small so the letter can still be revised. Materials stay pre built so a full request is a send, not a project. And the tracking has to survive weeks where you do not open the file at all, because those weeks will happen.

What follows is an account of how one cycle actually ran, described in enough detail that you can copy the parts that fit your own week.

Batching queries in tens instead of sending them all at once

The argument for batching is not patience. It is that the query letter is a document you are still testing, and sending eighty copies of an untested document destroys your list.

A batch of ten goes out. Then nothing happens for a while, which is normal, and the temptation is to send twenty more to feel productive. Resist for about six weeks, which is roughly when the first pattern becomes readable.

The patterns worth reading:

• Zero responses of any kind after eight weeks on ten agents suggests the letter is not landing, or the comp titles are wrong for the category.
• Form rejections only, arriving fast, usually points at the pitch paragraph rather than the pages.
• Partial requests that convert to rejections after the sample points at the opening chapters, not the letter.
• Full requests that go quiet points at nothing you can fix. That is the business.

Each batch mixes tiers deliberately: two or three agents you would be thrilled by, five solid fits, two newer agents actively building a list. Sending only to the dream tier in batch one wastes the names you most want on your weakest version of the letter.

Keep reading: How long should I wait before nudging an agent who still has my full manuscript?

Where the writing hours actually sit in a working week

The honest accounting from one cycle looked like this. Weeknight evenings produced almost nothing usable after a full day of meetings. What produced work was two fixed blocks: Saturday morning from six to nine, and Sunday afternoon from two to four thirty. Five and a half hours a week, protected.

Querying work is different from drafting work and can go in the leftover slots, because it is administrative rather than creative. Researching agents, checking submission guidelines, confirming an agency is open, logging a response: all of that fits in twenty minute pieces on a weeknight, on a phone, in a waiting room.

The split that made a difference:

The single largest efficiency gain was refusing to research agents during the weekend blocks. Research feels like writing work. It is not, and it will eat three hours you cannot get back.

Handling a full request that lands on a deadline week

It happened on a Wednesday in a week with a client presentation on the Friday. The email asked for the full as a Word document, double spaced, with a one page synopsis.

The synopsis did not exist. That was the failure, and it cost four days of turnaround while the writer scraped together evening hours to write one under pressure, badly.

The correction for the rest of the cycle was a materials folder, finished before any further queries went out, containing:

1. The full manuscript as a DOCX, Times New Roman 12 point, double spaced, one inch margins, header with author name, title and page number.
2. The same manuscript as a clean PDF, because some agents ask for it.
3. A one page synopsis, single spaced, present tense, whole plot including the ending.
4. A two page synopsis, because a minority of agencies specify that length.
5. First fifty pages as a standalone file.
6. First three chapters as a standalone file.
7. A short author bio in two lengths, fifty words and one hundred words.

Building all seven took one Saturday. After that, a full request took eleven minutes: confirm which version of the manuscript is current, attach, write four polite sentences, send, log the date.

The other rule that came out of that week: reply to the request within twenty four hours even if you send the material later. A short note saying the material will be with them by Sunday is entirely acceptable and buys you the weekend.

Keep reading: Should I query literary agents or submit straight to small presses with this book?

Personalizing at scale without rewriting every letter

Personalization does not mean writing a new letter. It means one paragraph that proves you chose this agent on purpose, attached to a fixed body you have already tested.

The structure that held up across the cycle: a personalization sentence or two, then the hook paragraph, then the plot paragraph, then the comps and metadata line, then the bio, then the sign off. Only the first block and occasionally the comp titles change.

Good personalization sources, in descending order of usefulness: something specific the agent said in a recent interview or on their agency page about what they want; a client's book you have genuinely read and can name a reason for; a stated wishlist item your book actually matches. Weak personalization, which is worse than none, is naming a client book you have not read or complimenting the agency generically.

Practical limit: cap personalization research at ten minutes per agent. If you cannot find a specific hook in ten minutes, send the clean unpersonalized version. It is not a penalty. A tight letter with no personalization beats a padded one with invented enthusiasm.

Keeping response deadlines visible when weeks blur

This is where a job does the most damage. Not to the writing, to the memory. Six weeks after sending, you genuinely will not recall whether a given agency has a no response means no policy or a stated eight week window.

What needs recording at the moment of sending, not later:

• Agent name, agency, date sent, and which materials went with the query.
• Which version of the manuscript they hold, because you will keep revising.
• The agency's stated response time, copied from their guidelines that day, since guidelines change.
• The nudge date, calculated as stated response time plus two weeks.
• Whether they accept nudges at all. Some explicitly do not.

The nudge itself is three sentences: the date you queried, the title and category, and a polite request for a status update. It goes as a reply on the original thread so the material is attached below it.

One correction from experience: nudge fulls, not queries. A query with no reply past its stated window is almost always a pass. A full sitting unanswered at four months is worth a courteous note, especially if you have another offer or a nearby deadline.

See how DraftAndQuery handles this for independent authorship and publishing

Conferences, pitch events and the cost of taking days off

A regional writers conference costs real money and real leave. Count it properly before deciding.

A two day conference at a $450 registration, $260 for two nights of lodging, $180 for travel and $60 for meals comes to $950, plus a vacation day if it runs into Friday. Add a paid pitch session at $30 to $50 and you are near $1,000. Those are illustrative figures for a mid size regional event, not a survey.

What the money buys, realistically: eight to ten minutes in front of one or two agents, and a request rate on pitches that tends to be more generous than the slush pile, partly because agents in a pitch room are being kind. A requested full from a pitch still enters the same reading queue as everything else.

Online pitch events cost nothing but attention, and attention is the scarce resource on a workday. If you use them, schedule the day, set the pitches in advance, and check twice rather than refreshing all day.

What the writer would change about the last cycle

Three things, in order of how much time they would have saved.

Build the materials folder first. Before query one. The four day scramble was entirely avoidable and it happened at the worst possible moment in the year.

Freeze the manuscript during a batch. Revising between sends means agents hold different versions, and when a full request arrives you spend twenty minutes reconstructing which one someone read. Pick a version, send the batch, revise afterward if the feedback warrants it.

Log at the moment of sending. Every gap in the record came from a night when logging felt like it could wait. It could not, and three entries in that cycle were reconstructed from sent mail weeks later, one of them wrongly.

Making the next cycle cheaper in hours

None of the above requires more writing time. It requires that the administrative layer stop consuming the writing time, which is what happens when your record lives in a spreadsheet you only remember to open on Sundays.

That is the specific job DraftAndQuery does: each manuscript with its version history, every agent and press holding a copy, which version they hold, the response window they stated, and a follow up queue that surfaces the nudges due this week. Log two lines when you send. Open it when a request lands and know immediately which file to attach.