Marketing Planning Cadence and OKRs: A Simple System for Startups

  1. Blog
  2. Measurement & Budget

By Saroj Jha · October 5, 2026 · 11 min read

Plan marketing on a fixed rhythm: a yearly goal, a quarterly plan with two or three OKRs, and a short weekly check of the numbers. Keep the OKRs few, shared with sales and tied to results, not activity. The five steps below show how, with one example company.

In three numbers

  • 60–70% the “sweet spot” for an OKR grade in Google's guide; scoring higher may mean the goals are not set high enough[1]
  • 360 of 452 Amazon's goals for 2010 that would directly affect customer experience; net income and operating profit were not used once[2]
  • 138 studies in a meta-analysis that found monitoring progress improved goal attainment, more so when progress was reported or made public[3]

Most startup marketing plans are either a long document nobody opens after week two or a list of activities. A cadence fixes both: the plan stays short, and it gets checked often.

The research on goals is unusually strong. Summarizing 35 years of studies, Locke and Latham wrote that "specific, difficult goals consistently led to higher performance than urging people to do their best"[4]. A meta-analysis of goals set for groups found that "specific difficult goals yield considerably higher group performance compared with nonspecific goals"[5]. OKRs (objectives and key results) are one practical way to write such goals down.

Planning is also about choosing. In Gartner's 2026 survey of 401 marketing leaders, mostly at companies with more than $1 billion in revenue, "56% of CMOs say their marketing organization lacks the budget required to deliver their 2026 strategy"[6]. A startup has far less. In my view, a good plan is mostly a list of what you will not do this quarter.

Running case: Example Co. Fictional

Example Co. is a seed-stage B2B SaaS startup whose software automates the month-end close for finance teams at companies with 50 to 500 employees. Its message is "Close the books in days, not weeks, without adding headcount." The team is eight people, including two founders and one account executive. The price is about $1,500 a month (about $18,000 a year), the typical sales cycle is about 45 days, and the goal is 30 new customers in the next 12 months. Early channels are founder-led outreach, LinkedIn, and accounting communities and partners. In this post, Example Co. sets its first quarterly marketing plan and OKRs. It uses the funnel rates from the Lead Management Playbook. Example Co. and all its numbers are illustrative.

Step 1: Set four planning clocks

Each clock answers a different question. The year sets the goal and the budget. The quarter decides what to bet on. The month checks where the money goes. The week checks whether the numbers are moving.

Google's OKR guide describes the same layering at a much larger scale: "Google sets annual and quarterly OKRs and holds company-wide meetings quarterly to share and grade OKRs"[1]. In my view, for a startup the quarter is the main unit of planning. It is long enough to run a real test of a channel and short enough to change course.

Example Co., step 1 Fictional

The CEO writes a one-page annual plan: 30 new customers, the budget, the target segment and three channels. Each quarter's plan and OKRs are set in the last two weeks of the quarter before. The founders review spend by channel on the first Monday of each month, and the CEO and account executive check the numbers every Monday. All fictional.

Figure 1 Four clocks cut the same year four ways; the quarter is the main unit of planning
  • Yearly Once a year One year: the goal and the budget
  • Quarterly 4 a year: the main unit Q1 Q2 Q3 Q4
  • Monthly 12 a year
  • Weekly 52 a year
  • Clock The question it answers What it sets or checks Example Co. (fictional)
  • Yearly What must be true in 12 months? The goal and the budget A one-page plan: 30 new customers, the budget, the target segment and three channels
  • Quarterly What will you bet on this quarter? The quarter's plan and two or three OKRs Two objectives with three key results each, set in the last two weeks of the quarter before
  • Monthly Is the money going where it works? Where the money goes The founders review spend by channel on the first Monday of each month
  • Weekly Are the numbers moving? Five numbers in 30 minutes, and one to three actions Every Monday, the CEO and the account executive, in one shared tracker

The clocks and what each checks are from Step 1; the bars show 1, 4, 12 and 52 meetings a year, drawn to scale. Example Co.'s versions are fictional.

Do this

Put all four meetings in the calendar now, each with an owner and a fixed length.

Free template

Write the one-page annual plan

The free Startup Marketing Plan template puts the yearly goal, the budget, the target segment and the channels on one page, with a 90-day roadmap for the first quarter.

Get the template

Step 2: Work back from the yearly goal to this quarter's numbers

A yearly goal is too big to act on. Break it into the numbers that produce it, quarter by quarter. For a sales-led B2B company, that is a short chain: customers come from opportunities, opportunities from qualified leads, qualified leads from marketing-qualified leads (MQLs). Multiply back through your conversion rates, using last quarter's real rates once you have them.

Then mark which links marketing controls and which it shares with sales. That tells you where marketing's key results should sit.

Example Co., step 2 Fictional

Thirty customers a year is 7.5 a quarter. With assumed rates of 25% from opportunity to customer, 71% from sales-qualified lead to opportunity, 47% from sales-accepted to sales-qualified, and 83% from MQL to sales-accepted, Example Co. needs about 30 opportunities, 42 sales-qualified leads, 90 sales-accepted leads and 108 MQLs this quarter. That is about 36 MQLs a month, the same number the Lead Management Playbook uses. All fictional.

Figure 2 Work back from 30 customers a year to 108 MQLs a quarter

Start at the customer target and divide by each conversion rate

7.5 New customers this quarter Commitment 30 a year, split evenly; tracked on its own, not graded ÷ 25% win rate 30 Opportunities Key result 1.2 Owned by marketing and sales together ÷ 71% become opportunities 42 Sales-qualified leads Not a key result ÷ 47% become sales-qualified 90 Sales-accepted leads Checked every Monday At least 80% of MQLs accepted: the Step 4 guardrail ÷ 83% accepted by sales 108 Marketing-qualified leads (MQLs) Key result 1.1 About 36 a month Show the numbers as a table
Stage This quarter Rate to the stage above Role
Marketing-qualified leads 108 83% accepted by sales Key result 1.1
Sales-accepted leads 47% become sales-qualified Checked weekly; guardrail at 80% accepted
Sales-qualified leads 71% become opportunities Not a key result
Opportunities 25% win rate Key result 1.2
New customers 7.5 Commitment, tracked on its own

Example Co.'s first quarter. Worked back and rounded: 7.5 ÷ 25% = 30; 30 ÷ 71% ≈ 42; 42 ÷ 47% ≈ 90; 90 ÷ 83% ≈ 108. Assumed rates, as in the Lead Management Playbook; all fictional. Bars drawn to scale.

Do this

Write your own chain from the annual goal down to the top of your funnel, and mark who owns each link.

Step 3: Write two or three OKRs, shared with sales

Google's guide sets the format. "Objectives are ambitious and may feel somewhat uncomfortable." "Key results are measurable and should be easy to grade with a number (Google uses a scale of 0 – 1.0)." It suggests three to five objectives with about three key results each, and puts the "sweet spot" for a grade at 60% to 70%: scoring higher may mean the goals were not set high enough[1]. For a marketing team of one or two people, I would start with two objectives.

Because a good OKR grade is 0.6 to 0.7, keep the number you promised your board out of the stretch OKRs. In my view, the customer target is a commitment, tracked on its own; the OKRs are the stretch bets that drive it.

Share the objectives with sales. In the group-goal meta-analysis, "egocentric" individual goals, aimed at maximizing individual performance, had a strongly negative effect on group performance (d = –1.75, from six effect sizes), while "groupcentric" goals, aimed at each person's contribution to the group, had a positive one (d = 1.20, from four). The authors conclude: "Individual goals can also promote group performance but should be used with caution in interdependent groups"[5]. Marketing and sales are interdependent. An MQL target that only marketing owns is an egocentric goal.

Figure 3 Goals aimed at the group raised group performance; goals aimed only at one's own performance lowered it
  • Specific, difficult group goals Compared with nonspecific goals; 23 effect sizes d = 0.80 Raised group performance
  • “Groupcentric” individual goals Aimed at each person's contribution to the group; 4 effect sizes d = 1.20 Raised group performance
  • “Egocentric” individual goals Aimed at each person's own performance; 6 effect sizes d = −1.75 Lowered group performance
  • −2 −1 0 +1 +2

d: the difference in group performance, in standard deviations. The two individual-goal estimates rest on few effect sizes, with intervals of ±0.60 and ±1.03.

Example Co.: both objectives belong to marketing and sales together, so neither team chases an MQL target alone. Fictional

Show the numbers as a table
Goal Effect on group performance (d) Effect sizes
Specific, difficult group goals vs nonspecific goals 0.80 (±0.35)
Groupcentric individual goals 1.20 (±1.03)
Egocentric individual goals −1.75 (±0.60)

Kleingeld, van Mierlo and Arends (2011), Journal of Applied Psychology: a meta-analysis of goal setting and group performance. Effect sizes and intervals as reported in the abstract. Drawn to scale.

Example Co., step 3 Fictional

Objective 1: "Fill the pipeline with the right finance teams", with key results of 108 MQLs from companies with 50 to 500 employees, 30 opportunities each with the CFO met, and four accounting-firm partners sending leads. Objective 2: "Turn first demos into signed contracts, faster", with a win rate of 25% or more, a median sales cycle of 45 days or less, and no deal waiting more than five days in security review. Both objectives belong to marketing and sales together. All fictional.

Figure 4 Example Co.'s Q1 OKRs: two objectives shared with sales, three key results each

Q1 marketing OKRs Shared with sales

Objective 1 Fill the pipeline with the right finance teams

Key result Target

  1. 1.1 MQLs from companies with 50 to 500 employees 108
  2. 1.2 Opportunities, each with the CFO met
  3. 1.3 Accounting-firm partners sending leads

Objective 2 Turn first demos into signed contracts, faster

Key result Target

  1. 2.1 Win rate 25% or more
  2. 2.2 Median sales cycle 45 days or less
  3. 2.3 Longest wait for any deal in security review 5 days or less

Commitment 7.5 new customers this quarter (30 a year): promised to the board and tracked on its own, not graded as a stretch OKR

At quarter end, each key result is graded from 0 to 1.0

0 0.6 0.7 1.0
  • Below 0.6: the team may not be achieving all it could
  • 0.6 to 0.7: the sweet spot
  • Above 0.7: the goals may not be set high enough

Example Co.'s objectives and targets are fictional. The 0 to 1.0 scale, the 0.6 to 0.7 sweet spot and what higher and lower grades may mean are from Google's OKR guide on re:Work.

Do this

Write no more than three objectives, three key results each, and agree them with whoever owns sales before the quarter starts.

Step 4: Pair every target with a guardrail

Goals have side effects. Ordóñez and colleagues describe "goal setting as a prescription-strength medication that requires careful dosing, consideration of harmful side effects, and close supervision." The side effects they list include "a narrow focus that neglects nongoal areas, distorted risk preferences, a rise in unethical behavior, inhibited learning, corrosion of organizational culture, and reduced intrinsic motivation"[7].

In marketing, the classic side effect is a lead target met by loosening the lead definition. The fix is a guardrail: a health metric that is not a goal but must not get worse. Pair each volume key result with a quality check.

If you sell an AI product

In my view, an AI product needs two extra guardrails. The first is quality: a usage or activation target can be met while the answers get worse, so track an accuracy or error measure your buyers care about next to it. The second is cost. AI features can carry a running cost each time they are used, so a key result that drives usage can also lower gross margin.

Duolingo's shareholder letter for the fourth quarter of 2024 shows that trade-off. It says gross margin "decreased by approximately 120 basis points year over year to 71.9%" "due to lower subscription margins from increased generative AI costs due to the increased adoption of Duolingo Max", which it calls "our highest subscription tier". The margin sentence also names a second cause, lower advertising gross margin. It adds: "While we expect to continue growing margins this year, we will do so at a more measured pace due to the investments we are making into AI and the variable costs associated with Video Call"[8]. This is what the letter says, not how the company decided internally. In my view, AI categories also move fast: a new model release can change what buyers expect within a quarter. So I would look at AI-related key results at the monthly check, not only at quarter end.

Figure 5 Pair every target with a guardrail: a health metric that must not get worse
Target The side effect to catch Guardrail

Example Co. (fictional)

Target Key result 1.1: 108 MQLs Side effect Loosening the MQL definition Guardrail At least 80% of MQLs accepted by sales Target Key result 2.1: a win rate of 25% or more Side effect Discounting to win deals Guardrail Average contract stays near $18,000 a year

If you sell an AI product

Target A usage or activation target Side effect Answers get worse as usage grows Guardrail An accuracy or error measure your buyers care about Target A key result that drives usage Side effect Each use has a running cost, so gross margin falls Guardrail Gross margin, checked monthly 71.9% Duolingo's gross margin, fourth quarter of 2024: down about 120 basis points year over year. The letter names higher generative AI costs as adoption of Duolingo Max grew, and lower advertising gross margin.

Example Co.'s guardrails are from Step 4 and fictional; the AI guardrails are my judgement. Duolingo: its shareholder letter for the fourth quarter of 2024, Exhibit 99.2 to Form 8-K. What the letter says, not how the company decided.

What Amazon's shareholder letter shows

Amazon's 2009 letter to shareholders describes how the company set goals for 2010. "Our annual goal setting process begins in the fall, and concludes early in the new year after we've completed our peak holiday quarter." "For 2010, we have 452 detailed goals with owners, deliverables, and targeted completion dates." "We review the status of each of these goals several times per year among our senior leadership team and add, remove, and modify goals as we proceed." Then come the statistics: "360 of the 452 goals will have a direct impact on customer experience. The word revenue is used eight times and free cash flow is used only four times. In the 452 goals, the terms net income, gross profit or margin, and operating profit are not used once." The letter explains the choice: "focusing our energy on the controllable inputs to our business is the most effective way to maximize financial outputs over time"[2].

This is what the letter says, not an account of how Amazon made its decisions internally, and it describes 2010. In my view, the lesson for a startup is the split: most goals sit on inputs the team can control, while the financial outcome is watched, not chased directly.

Figure 6 360 of Amazon's 452 goals for 2010 would directly affect customer experience; profit terms were not used once

The 452 detailed goals for 2010, each with an owner, deliverables and a target completion date

Direct impact on customer experience: 360 Other goals: 92

How often the 452 goals use financial terms

  • Revenue 8 times
  • Free cash flow 4 times
  • Net income, gross profit or margin, operating profit not once

The yearly cycle the letter describes

  1. In the fall Goal setting begins
  2. Early in the new year It ends, after the peak holiday quarter
  3. Several times a year Senior leaders review each goal and add, remove or modify goals
Show the numbers as a table
Measure Value
Detailed goals for 2010 452
Goals with a direct impact on customer experience 360
Uses of “revenue”
Uses of “free cash flow”
Uses of net income, gross profit or margin, and operating profit

Amazon.com, Inc., 2009 letter to shareholders, Exhibit 99.1 to Form 8-K, filed April 14, 2010. What the letter says about its 2010 goals, not how decisions were made. Bars drawn to scale.

Example Co., step 4 Fictional

Example Co. adds two guardrails. At least 80% of MQLs must be accepted by sales, so the MQL key result cannot be met by loosening the definition. And the average contract must stay near $18,000 a year, so the win-rate key result cannot be met with discounts. Its key results are inputs it can influence (MQLs, opportunities, time in security review); the 7.5 customers is the output it watches. All fictional.

Do this

For each key result, write the one way it could be gamed, and add a guardrail that would catch it.

Step 5: Check weekly, then grade and reset every quarter

A plan works only if someone looks at it. A meta-analysis of 138 studies with 19,951 participants found that interventions that got people to monitor their progress improved goal attainment (d+ = 0.40), and that the effect was larger when progress was reported or made public, and when it was physically recorded[3]. Locke and Latham likewise report that goals work best when people get feedback on their progress[4].

Two meetings are enough: a short weekly check of a few numbers, written where everyone can see them, and a quarterly review that grades each key result and sets the next quarter. Google's guide adds that "OKRs are public so that everyone in the organization can see what others are working on", and that "OKRs are not synonymous with employee evaluations"[1]. Grade the work, not the people.

Example Co., step 5 Fictional

Every Monday, the CEO and the account executive review five numbers against plan (MQLs, sales-accepted leads, opportunities, win rate and sales cycle), mark each key result on track, at risk or off track, and agree one to three actions. At quarter end, they grade each key result. If the quarter ends with 95 MQLs against 108, key result 1.1 grades about 0.9. Then they set the next quarter's OKRs and send the board update. All fictional.

Figure 7 Check five numbers every Monday; grade every key result at quarter end

Weekly check 30 minutes, every Monday

  1. Five numbers against plan MQLs, sales-accepted leads, opportunities, win rate and sales cycle
  2. Mark each key result On track, at risk or off track
  3. Agree one to three actions The CEO and the account executive
  4. Write it down In the one shared tracker everyone can see

Quarterly review At quarter end

  1. Grade each key result From 0 to 1.0
  2. Set next quarter's OKRs With sales, as in Step 3
  3. Send the board update The grades, the funnel and the guardrails

Grading key result 1.1 at quarter end

95 MQLs in the quarter ÷ 108 the key result's target = ≈ 0.9 the grade for key result 1.1

Why check every week: across 138 studies with 19,951 participants, getting people to monitor their progress improved goal attainment (d+ = 0.40), more so when progress was reported or made public, or physically recorded.

Show the numbers as a table
Key result 1.1 Value
Target 108 MQLs
Result 95 MQLs
Grade 0.88, about 0.9

Example Co.'s meetings and numbers are fictional; 95 ÷ 108 = 0.88. Evidence: Harkin and others (2016), a meta-analysis in Psychological Bulletin.

Do this

Set up one shared tracker with your key results and guardrails, and update it in the same 30-minute slot every week.

What changes by stage?

  • Pre-seed — One goal for the quarter (for example, a number of customer conversations or paying pilots) and a weekly check. OKRs can wait.
  • Seed — The four clocks (Step 1), a funnel chain from the annual goal (Step 2), and two objectives with three key results each, shared with sales (Step 3).
  • Series A–B — OKRs for each marketing team that roll up to company objectives, guardrails in the dashboard, and a quarterly review that feeds the board pack.

What mistakes should I avoid?

  • Too many OKRs. Google's guide warns that more objectives "can lead to over-extended teams"; a small team needs fewer still.
  • Activity as a key result. "Publish 12 posts" is a task. Write what the posts should change.
  • Marketing-only targets. Shared goals beat goals each team chases alone.
  • No guardrails. Every volume target can be gamed.
  • Mixing commitments and stretch goals. Track the board number separately from OKRs graded at 0.6 to 0.7.

Related on AAJ

Articles

Playbooks

Reports

Templates and tools

References

The superscript numbers in the article match this list.

  1. Google re:Work. "Guide: Set goals with OKRs." Official guide, originally published 2016. rework.withgoogle.com
  2. Amazon.com, Inc. (2010). 2009 letter to shareholders, Exhibit 99.1 to Form 8-K, filed April 14, 2010. US Securities and Exchange Commission. sec.gov
  3. Harkin, B., Webb, T. L., Chang, B. P. I., Prestwich, A., Conner, M., Kellar, I., Benn, Y. and Sheeran, P. (2016). "Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence." Psychological Bulletin, 142(2), 198–229. doi:10.1037/bul0000025
  4. Locke, E. A. and Latham, G. P. (2002). "Building a practically useful theory of goal setting and task motivation: A 35-year odyssey." American Psychologist, 57(9), 705–717. doi:10.1037/0003-066X.57.9.705
  5. Kleingeld, A., van Mierlo, H. and Arends, L. (2011). "The effect of goal setting on group performance: A meta-analysis." Journal of Applied Psychology, 96(6), 1289–1304. doi:10.1037/a0024315
  6. Gartner (2026). "Gartner 2026 CMO Spend Survey Finds CMOs Allocate 15.3% of Marketing Budgets to AI, But Only 30% Are Ready to Scale AI Capabilities." Press release, May 11, 2026; survey of 401 CMOs and other marketing leaders in North America, the United Kingdom and Europe, January–March 2026. gartner.com
  7. Ordóñez, L. D., Schweitzer, M. E., Galinsky, A. D. and Bazerman, M. H. (2009). "Goals Gone Wild: The Systematic Side Effects of Overprescribing Goal Setting." Academy of Management Perspectives, 23(1), 6–16. doi:10.5465/amp.2009.37007999
  8. Duolingo, Inc. (2025). Shareholder letter, fourth quarter and fiscal year 2024 (quarter ended December 31, 2024). Exhibit 99.2 to Form 8-K, filed February 27, 2025. US Securities and Exchange Commission. sec.gov

More in Go-to-Market & Growth Planning

Part of the Go-to-Market & Growth Planning hub - see all 27 resources on this topic.

Also useful in Analytics, Experiments & Budget.