The August Operations Checklist: What Should Be Running Before the Final 100 Days

August 1 is the starting line for the final sprint. The campaigns that win in November are the ones that show up to that starting line with the operation already running, not the ones that use August to figure out what they're doing.

Seven items. If any of these aren't checked before August 1, that's your configuration gap.

1. Voice Model: Current and Calibrated

Status to check: Does the candidate's AI voice model reflect how they're communicating right now (in July 2026), or how they were communicating when you trained it?

Voice models drift. The candidate who was running a primary campaign in January had a different stump speech, different signature phrases, and different donor relationship language than the candidate running a general election campaign in July. If your voice model hasn't been updated since the primary, your AI-generated communications are running on stale training.

What recalibration looks like: new speech samples, updated email examples from the last 60 days, a review of the current stump to identify phrasing that's become standard. Takes a day or two. Worth it before the sprint.

Checked means: Finance director reviewed a batch of recent AI drafts, confirmed they sound like the candidate today, not the candidate six months ago.

2. Approval Workflow: On a Schedule

Status to check: Is the approval workflow running daily on a predictable schedule, or is it running when someone has bandwidth?

This is the most common configuration failure in deployed AI campaigns. The tool is technically working. But the approval step (the human review before anything sends) runs on an informal schedule that bends to whatever else is happening. When the candidate is traveling or the finance director is stretched, the workflow slows. Emails go out late. Follow-ups miss their 24-hour window.

An approval workflow that runs "when someone gets to it" is not an approval workflow. It's a bottleneck.

Checked means: There's a daily approval window (specific time, specific person) that runs regardless of what else is on the calendar. Someone other than the candidate is empowered to approve standard donor communications.

3. Fundraising Email Program: Frequency and Segmentation Active

Status to check: Is the email program running at 2x per week minimum? Is your list segmented (house file, major donors, prospects, event attendees) with different messaging for each segment?

One email per week is the floor for a campaign that's present. Two per week is the floor for a campaign that's competing. Less than that and your list loses warmth between sends; you're essentially starting the relationship from scratch every time you mail.

Segmentation matters in the sprint. Major donors in cultivation sequences shouldn't be getting the same email as first-time small donors. Event attendees have a different relationship with the campaign than cold prospects. If you're sending one email to everyone on the list, you're leaving response rate on the table.

Checked means: Email frequency is 2x per week or above. At minimum two segments are active (major donors vs. everyone else). Cadence is steady, not a spike before FEC deadlines followed by silence.

4. Donor Follow-Up: Triggering Within 24 Hours

Status to check: After a call session ends, when does the first follow-up email go to donors from that session?

24 hours is the target. The follow-up that arrives the morning after a call lands while the conversation is still active. The one that arrives three days later lands after the donor has moved on.

If your follow-up velocity is 48 hours or longer, the problem is almost always the approval workflow (item 2 above). Fix the approval step and the velocity follows.

Checked means: Review the last three call sessions. Calculate actual time from session end to first follow-up send. Target: under 24 hours for 90% of contacts.

5. Major Donor Cultivation Sequences: Active Through September

Status to check: Do you have cultivation sequences running for your top 50-100 major donor prospects, with planned touches through September and ask timing scheduled for late September or October?

Major donors in October are in conversations that started in June. The check that comes in September comes from a relationship that was tended in July. If you're not running cultivation sequences now, the September fundraising results are already determined.

What an active sequence looks like: a personal research pull on each prospect (what they care about, who they know in the network), a summer touch (call, event, personal email, depending on relationship warmth), a cultivation email in August that references something specific, a formal ask conversation scheduled for late September.

Checked means: A list of major donor prospects exists. Each has an active next action. The sequence has dates attached, not just intentions.

6. News Monitoring: Updated for the General Election Race

Status to check: Does your news monitoring configuration reflect the current race (your actual opponent, current issue landscape, relevant local media), or is it still running the primary configuration?

The monitoring setup you built for the primary was tracking the wrong opponents and possibly the wrong issues. The general election race is different. If your daily digest is still surfacing primary-relevant content with primary-relevant framing, your content team is doing extra filtering work that the system should be doing for them.

Checked means: Source list reviewed in the last 30 days. Opponent name(s) updated. Issue keywords reflect current race dynamics. A recent digest reviewed to confirm relevance.

7. Rapid Response Trigger: Configured

Status to check: If something breaks fast (an opponent attack, a news development, a local controversy), does your operation have a rapid response path that doesn't require someone to build the process under pressure?

The campaigns that respond within hours to breaking news do it because they had a process before the news broke. The ones that respond the next day are figuring out process while the story is moving.

What a rapid response trigger looks like: a designated decision-maker who can initiate a send, a template structure the AI can adapt from quickly, an expedited approval path (one person, not three), and a distribution setup that can execute in hours rather than days.

Checked means: Someone on the team can describe, without looking it up, how a rapid response email gets from "we need to respond to this" to "that email is sending" in under four hours.

What August Looks Like With This in Place

The campaigns that show up to August 1 with all seven items checked are in execution mode for the entire sprint. Higher email frequency, consistent follow-up velocity, major donor relationships already in motion, news monitoring generating content on a daily feed.

The campaigns that don't have it checked are doing two things at once: running the operation and building the operation. That's a tax on every week of the sprint.

The checklist is short. The window to close the gaps is now.

Eric Linder is a former California State Assemblyman (2012-2016) and founder of AutomatedTeams, an AI operations consultancy for political campaigns and advocacy organizations.