What to run this quarter beside the fixes
Repairs and campaigns happen in the same quarter, not one after the other. The Suggested Plan puts them side by side: Where to invest on one side, What to fix on the other, both built from one reading of your own records. This article explains where each growth item comes from and what it needs to appear; for the ranked repairs on the other side of the screen, read What should you fix first?
Before you start
Plan required
Pro, Business, Enterprise, Partner or a Free Trial. On Starter the page says which plan unlocks it.
Permissions required
The plan reads only records already synced into your workspace. It writes nothing to your tool, so a read-only connection produces the whole thing.
- One connected tool with a finished sync. The plan is built from MarqIQ’s copy of your records, not from a live call.
- Nothing else. The four items in the second half of this article need no conversion statistics at all, which is the point of them.
Open the plan
Open Insights in the left-hand menu, then Suggested Plan.
Pick the tool you want the plan for along the top.
Read the population line under the headline before anything else.
It names how many records the plan was built over, and the period when you have set one. It is the same sentence the Analysis screens print, so the two cannot describe different populations.
Read the Where to invest card.
At most three items. A channel bet is ranked on how many customers it produced above or below what your overall rate would have produced on the same volume; a motion, which claims no rate, is ranked on the size of the audience it names. They sit beside What to fix on purpose.
Open an item for its evidence and its stop rule.
Each card carries What we found, What to do, Why it matters and Done when, then a strip of evidence where every value prints what it was measured on.
Click Rebuild after a sync if the figures look stale.
The plan is cached for five minutes on the server, and the time it was built is printed under the button.
Note
The same tab also carries five tiles, a scorecard, This week (at most three items, low effort and with a 30-day horizon, so something ships before the quarter plan is agreed), two short columns headed What sales should know and What RevOps should settle, and a six-month roadmap. Worth doing alongside, What we could not tell you, Definitions used here and the limits sit behind Show the working.
The four bets that rank a channel
These read your own conversions, so they need enough outcomes to be worth ranking. Each is a bounded scale-up with a stop rule rather than a reallocation on faith.
| Item | What it names | What it needs |
|---|---|---|
| Put more into a channel | The channel converting above your baseline, with its own rate, your overall rate, and the customers it produced above what the baseline would have produced | A channel with at least 30 records, at least 2% of the volume analysed, and a conservative lift of 1.5 times your baseline |
| Your best buyer, from the data | One winning value per who-they-are field — industry, job title, company size — combined into a profile to point targeting, outbound and content at | At least one such field beating the baseline at the conservative end, on at least 30 records |
| Stop feeding what does not convert | The channel carrying 10% or more of your volume and returning well under the baseline, with the shortfall in customers | At least 60 records in that channel and a conservative lift at or below 0.6 times the baseline |
| Revenue concentration | The channel carrying 40% or more of recorded revenue, to protect while a second engine is built | Amounts recorded against at least two values of a channel field |
All four sit behind one gate, and MarqIQ refuses to rank anything until it clears.
- At least 500 records synced. Below that nothing is ranked and the reason is printed.
- An outcome recorded at all, and at least 30 records that reached a decided outcome. A rate on fewer than 30 moves with a single deal.
- Lead source filled on at least half your records. Below that a channel ranking would describe the recorded minority and call it your pipeline.
- No lift is ever quoted from a point estimate. The conservative end of the 95% range is used throughout, so a segment has to be genuinely ahead rather than luckily ahead.
When the gate fails, the plan says so
The whole plan switches to a readiness footing: the heading over the repairs becomes Make the next quarter measurable, and every rate-based check is listed under What we could not tell you with the reason it did not run and what would unlock it. That sentence is the product working, not the product failing — and the four items below still appear, because none of them needs a rate.
The four motions that need no conversion statistics
A workspace whose data needs fixing has no channel ranking, and used to get a list of repairs and nothing to run. These four are grounded in what an analysis already found, name the audience and its size from your own records, and each one says which fix it runs alongside.
| Motion | Where the number comes from | What it needs to appear |
|---|---|---|
| Win back the people you can still email but have stopped talking to | Audience health: the people whose last open or click is more than a year old, plus the ones who never engaged at all | At least 100 people across those two groups |
| Open a second way in | Your tool’s own form records: the share of everyone a form brought in who came through a single form | At least 30 people brought in by a form, the top form carrying 50% or more. HubSpot and Marketo only |
| Run the next campaign tagged properly | Source truth: the people carrying no source at all | At least 200 people, with 25% or more carrying no source |
| Run something that produces the signals a score can read | The capability check: your tool supports lead scoring and this workspace has not set it up | At least 100 people, and scoring genuinely not set up on a tool whose scoring MarqIQ has verified |
The shape of what each one tells you to run is taken from your business model rather than from a list of tactics:
- Sells to businesses through a sales team, large deals: a live session on the problem you solve — a webinar or a small round-table, with the sales team on the call.
- Sells to businesses through a sales team, smaller deals: a webinar or a short live demo series, with a follow-up sequence for everyone who attends or registers.
- Sells to businesses through the product: a free tool, calculator or assessment that gives a result worth having.
- Sells to consumers: an offer with a deadline, sent as a short series rather than one email.
- Sells services: a short piece of real work given away — an audit, a teardown, a first consultation.
- Where nobody has told MarqIQ which of those you are: one campaign with a single clear offer, sent as a short series. Fill in the business details the tab asks for to get the shaped version.
The win-back divides by the same number as the analysis
The share it quotes is of reachable people who are not customers, which is the denominator Audience health uses for the same group. The two screens cannot print different percentages for one set of people, and the card prints both the count and the denominator.
How the two halves are kept from arguing
- Growth items and repairs are ranked separately. Growth is ranked on customers above or below baseline; repairs are ranked on severity first, then on how much of the database they touch, then toward the cheaper action.
- Only three growth items are shown. The rest are listed under What we could not tell you as real opportunities held back so the three above get funded first.
- Where an analysis and a play describe the same problem, the play is folded into the analysis finding and the fold is recorded with its reason. The analysis reads your tool far more closely than a play can.
- Where a growth bet and a channel test name the same segment, the test is dropped rather than said twice, and the drop is listed with its reason.
- An item that carries a Found by the ... analysis line links back to the analysis holding its evidence, so no plan item is a claim you have to take on trust.
Take it to a meeting
Read Definitions used here first, under Show the working.
Agree those before the meeting, or the meeting will be about those.
Take the top bet and the top fix together, not one without the other.
They are printed side by side because they are meant to run in the same quarter.
Quote the denominator with every figure.
Each evidence value already prints what it was measured on. Keep that half of the sentence when you copy it.
Agree the stop rule as well as the scale-up.
The card gives one: scale while the segment holds at or above twice the baseline, and stop the moment it drops toward baseline. That is the channel’s ceiling, and finding it is the answer, not a failure.
Read What we could not tell you out loud.
It is rendered rather than omitted on purpose. A missing section reads as “we checked and it was fine”, which is worse than an absent one.
Take care
A segment converting well is correlational evidence. It is not proof that funding it harder produces more, which is exactly why the recommendation is a bounded scale-up with a readout on the segment’s own rate under load, and never a reallocation on faith. The plan carries no benchmark and no industry comparison anywhere.
If it did not work
The Where to invest card is missing and a single paragraph stands in its place
Why: No channel cleared the test, so instead of an empty card MarqIQ prints the reason and what would change it.
Fix: Read that paragraph. It names the bar — at least 30 records, 2% of volume, and a conservative lift of 1.5 times your baseline — and says whether the shortfall is outcomes or channel performance.
The plan says it cannot rank anything
Why: The outcome gate did not clear: fewer than 500 records synced, no outcomes recorded, fewer than 30 decided outcomes, or lead source filled on under half your records.
Fix: The reason on screen names which one. Fix that first — it is usually the source field — and the four ranked bets appear on the next rebuild. The four motions are there in the meantime.
You expected a win-back item and there is none
Why: Fewer than 100 people are quiet or never engaged, or Audience health cannot judge engagement at all because no field records when somebody last opened, clicked or replied.
Fix: Open Audience health for the same tool and read its headline. It says which of the two you have.
The second-way-in item never appears
Why: It reads your tool’s own form records, which MarqIQ has for HubSpot and Marketo only, and it needs at least 30 people brought in by a form with the top form carrying half of them.
Fix: On another tool, use Form fills for what MarqIQ’s own script captured instead. The motion itself is worth running regardless; the card only appears when MarqIQ can put your numbers behind it.
A growth item has no owner tag where the repairs do
Why: Ownership is a policy table covering the ranked plays. The four motions are not in it, so no owner pill is drawn on them.
Fix: Assign one yourself. Each motion still carries a Done when line, which is the part that makes it schedulable.
The plan has not changed since your last sync
Why: It is cached for five minutes on the server.
Fix: Click Rebuild. The time the plan was built is printed under the button, so you can tell a cache from a plan that genuinely has not moved.
The plan says nothing here needs changing this quarter
Why: No check cleared the bar for a recommendation.
Fix: Treat it as a real result and read What we could not tell you underneath it, which lists what was not computable. An empty plan and a clean one look different there.
Next steps
Related articles
Was this article helpful?