Marketing Campaigns - live preview

Good morning, Elena

The autumn campaign sent to 18,400 contacts yesterday and 2 segments need a consent review.

Contacts, mailable42,180of 61,400 total

Sent, last 30 days94,2006 campaigns

Open rate31.4%up 2.1 pts

Unsubscribe rate0.42%watch above 0.5%

Needs attention

3 open

Autumn campaign - unsubscribe rate 0.9%Roughly double normal, sent to a cold segmentHigh opt-out

Segment 'Trial 2024' has no consent basis19,220 contacts excluded from sending until resolvedConsent

Sending domain reputation dippedBounce rate rose after the 18,400 sendDeliverability

Campaigns

What was sent, to whom, and what came back.

Recent campaigns

Campaign	Segment	Sent	Recipients	Open	Click	Opt-out
Autumn range launch	Active customers	31 Aug	18,400	31.4%	4.2%	0.90%
Webinar invite	Engaged, 90 days	24 Aug	6,120	42.1%	11.4%	0.21%
Monthly digest, August	All subscribers	01 Aug	41,800	28.6%	3.1%	0.34%
Re-engagement	Dormant, 12 months	18 Jul	12,400	9.2%	0.8%	1.80%
Product update	Active customers	04 Jul	18,200	34.8%	6.6%	0.18%

Re-engagement campaigns always cost opt-outs. The question is whether the ones who stay are worth it.

Segments

Who a campaign goes to, and whether you are allowed to send to them.

Saved segments

Segment	Definition	Contacts	Mailable	Consent basis	Status
Active customers	Purchased in last 12 months	18,400	18,400	Contract	Ready
Engaged, 90 days	Opened or clicked in 90 days	6,120	6,120	Consent	Ready
All subscribers	Opted in, not unsubscribed	41,800	41,800	Consent	Ready
Dormant, 12 months	No engagement in 12 months	12,400	11,180	Consent	Ready
Trial 2024	Signed up during 2024 trial	19,220	0	None recorded	Blocked
Event attendees 2023	Attended any 2023 event	3,140	1,860	Mixed	Partial

A segment with no lawful basis recorded cannot be sent to. The app will not let you argue with it.

Contacts

One record per person, with the consent history attached to it.

Recent contact activity

Contact	Source	Opted in	Basis	Last engaged	Status
Contact 88412	Website form	12 Aug 2026	Consent	31 Aug 2026	Mailable
Contact 88408	Event badge scan	18 Jun 2026	Consent	24 Aug 2026	Mailable
Contact 88396	Purchased	04 Feb 2026	Contract	31 Aug 2026	Mailable
Contact 88371	Imported list 2024	-	None	-	Suppressed
Contact 88355	Website form	22 Mar 2025	Consent	18 Jul 2026	Unsubscribed
Contact 88340	Website form	09 Jan 2026	Consent	-	Bounced, hard

An unsubscribe is permanent and global. Re-adding someone through an import does not undo it.

Performance

What campaigns produced, beyond opens.

Attributed pipeline$286,000last 90 days

Cost per lead$18.40down $2.10

Leads to opportunity14%of 412 leads

By campaign, last 90 days

Campaign	Recipients	Leads	Opportunities	Pipeline	Cost per lead	Verdict
Webinar invite	6,120	142	31	$186,000	$8.20	Strong
Product update	18,200	96	14	$62,000	$14.10	Good
Autumn range launch	18,400	88	6	$38,000	$22.40	Watch
Monthly digest	41,800	62	4	$0	$41.80	Poor
Re-engagement	12,400	24	1	$0	$96.20	Stop

The monthly digest has the largest audience and the worst return. That is usually a habit rather than a decision.

Lead scoring

What a score is made of, so sales can argue with it usefully.

Scoring model

Signal	Source	Points	Decay	Status
Opened an email	Campaign	+2	30 days	Active
Clicked a link	Campaign	+5	30 days	Active
Visited pricing page	Website	+15	14 days	Active
Requested a demo	Website form	+40	None	Active
Job title matches ICP	Contact record	+20	None	Active
Unsubscribed	Campaign	Reset to 0	-	Active
No engagement 90 days	System	-10	-	Review due

A score with no explanation is a score sales will ignore. Every point is attributable to a signal.

Consent and sending

Lawful basis per contact, and the sending setup that keeps mail delivered.

Consent bases in use

Basis	Definition	Contacts	Evidence held	Expiry	Status
Consent	Explicitly opted in	41,800	Timestamp, IP, form	None	Valid
Contract	Existing customer relationship	18,400	Order record	12 months after last order	Valid
Legitimate interest	Assessed and documented	1,860	LIA document	Annual review	Review due
None recorded	Imported without basis	19,220	None	-	Suppressed

Contacts with no recorded basis are suppressed rather than deleted, so the record of what happened survives.

Why consent is a record, not a checkbox

The hard question is never whether someone is on your list, it is whether you can show how they got there. Consent here is a dated, evidenced record attached to the contact, not a flag that can be flipped by an import.

Timestamp, source and form content stored at opt-in
An import cannot set a consent basis - it must be evidenced
Unsubscribes are permanent, global and survive re-import
Contacts without a basis are suppressed automatically
A contact can be exported or erased on request, with the action logged
Suppression lists are never overwritten by a bulk update

Audit log

Every send, suppression and consent change. Retained for seven years.

Events

When	Actor	Action	Target	Source
01 Sep 08:40	System	Blocked send, segment has no lawful basis	Trial 2024	scheduler
31 Aug 10:00	Elena Fischer	Sent campaign to 18,400 contacts	Autumn range launch	web
31 Aug 10:04	System	Suppressed 220 hard bounces	Autumn range launch	scheduler
30 Aug 15:20	Lena Brandt	Recorded legitimate interest assessment	Event attendees 2023	web
28 Aug 09:10	System	Processed 14 erasure requests	Contacts	scheduler

Autumn renewal

One campaign, what it is doing and what it is costing.

LiveSending heldCMP-0412 · Autumn renewal · email and post · 18,400 contacts

Autumn renewal

Held after the first batch. Open rate is the best this year and unsubscribes are four times the normal rate, which is the same audience telling you two different things.

Sent4,200of 18,400 · batch 1 of 5

Opened42.8%best this year

Clicked6.1%above the 4.2% average

Unsubscribed1.9%normal is 0.4%

What the numbers say

Subject lineTested against two others · clear winnerWorking

OfferRenewal discount · 6.1% clickedWorking

AudienceIncludes 3,100 who renewed last monthThe problem

FrequencyThird email to this list in 11 daysThe problem

People who already renewed are being asked to renew, for the third time in eleven days. They open it because the subject line is good and then leave.

Who unsubscribed

Renewed in the last 30 days62 of 80Excludable

Contacted 3+ times this month71 of 80Excludable

Never opened before today4 of 80

Long-standing customers58 of 80Expensive

Fifty-eight of the eighty had been on the list for over three years. The next batch is 14,200 people and the same two exclusions would remove 5,900 of them.

Batches

Batch 14,200 sent · 6 September 09:00Sent

Batch 24,200 · scheduled 8 SeptemberHeld

Batch 34,200 · scheduled 10 SeptemberHeld

Batch 43,400 · scheduled 12 SeptemberHeld

Batch 52,400 post · scheduled 15 SeptemberHeld

Batching exists so that a mistake costs 4,200 contacts rather than 18,400. This is the case it was built for.

New campaign

Who it goes to, on what channel, and how fast.

The campaign

NameChannelGoal

SegmentBatching

What will be excluded automatically

No consent for this channel2,140Always

Contacted 3+ times this month5,900On

Already did the thing being asked3,100On

Unsubscribed1,880Always

The middle two are switchable and were off for the autumn renewal. Turning them on would have removed 5,900 of the 18,400 and most of the unsubscribes with them.

Frequency cap

Maximum contacts per person1 a month2 a month3 a monthNo cap

A cap is the cheapest control here. It costs some reach on every campaign and saves the list, which only has to be lost once.

Nothing sends until a second person approves the first batch.

Content

What is actually being sent, and which version won.

Version C liveTested on 1,200Autumn renewal · email · three subject lines, one body

Subject lines

C live

Version	Subject line	Sent	Opened	Clicked	Outcome
A	Your renewal is due	400	24.1%	3.8%	Lost
B	Renew before 30 September	400	31.6%	5.2%	Lost
C	What changed in your plan this year	400	42.8%	6.1%	Live

Version C wins on both measures and still produced the unsubscribes. A good subject line sent to the wrong list gets more people to read the wrong message.

The body

OpeningWhat changed in their plan, personalised

OfferRenewal discount, expires 30 September

ProofTwo customer lines, no logos

AskOne button, one link to the plan page

Checks before sending

Every link resolves and carries the campaign tag
Unsubscribe is one click and works without signing in
Plain text version reads sensibly on its own
Sender address accepts replies from a human
Nothing in it claims a saving that is not in the pricing table

Attribution

What the spend produced, under more than one rule.

Spend this quarter$184,000across 22 campaigns

Attributed revenue$1.24mlast touch

Under first touch$984,000$256k lower

Cost per acquisition$412target is $500

By rule

3 rules shown

Campaign	Spend	Last touch	First touch	Even split	Outcome
Autumn renewal	$18,400	$284,000	$96,000	$188,000	Depends
Trial nurture	$24,000	$96,000	$312,000	$204,000	Depends
Search, brand	$41,000	$412,000	$88,000	$248,000	Overstated
Conference stand	$62,000	$41,000	$284,000	$162,000	Understated
Winback post	$38,600	$188,000	$204,000	$196,000	Stable

The same five campaigns, three rules, entirely different rankings. Brand search looks like the best line under last touch and the worst under first.

Only one row holds still whichever rule you pick. The rest of this table is an argument about method, and showing all three is more honest than picking one and calling it the number.

Calendar

What is going out, to whom, and how often they hear from you.

Campaigns this month93 to the same list

Contacts hit 3+ times5,90032% of the list

Frequency capOffno limit is set

Quiet contacts4,120nothing for 6 months

Schedule

5,900 over-contacted

Date	Campaign	Channel	Audience	Size	Overlap
6 Sep	Autumn renewal	Email	Renewals due	18,400	3,100 renewed already
2 Sep	Product update	Email	All customers	24,800	18,400 overlap
28 Aug	Webinar invite	Email	All customers	24,800	18,400 overlap
19 Aug	Winback	Post	Lapsed 6-12 months	4,200	None
12 Aug	Trial nurture	Email	Trials	3,100	None

Three of the top five went to overlapping lists inside eleven days. The overlap column is the only place that was visible before the unsubscribes arrived.

Four thousand contacts have heard nothing in six months while five thousand heard three things in eleven days. Both numbers come from having no cap.

Reports: activity

Volume and throughput, by owner and by channel.

Volume over time

12 weeks

Emails sent per week. The three peaks are three campaigns to overlapping lists inside a fortnight, which is what the calendar screen was built to prevent.

Contacts reached24,800across 22 campaigns

Opened this month22against 18 closed

Median age11 daystarget is,

Owned by nobody3oldest is 9 days

By channel

This quarter

Email18,400

In product8,200

Post4,200

SMS2,100

Events900

By marketer

This quarter

Marketer	Open	Closed	Median days	Overdue
Marta Reyes	6	9	8	2
Joran Bakke	5	7	11	4
Celine Aubert	4	6	9	3
Tobias Lund	4	4	14	6
Unassigned	3	0	-	9

Median days is brief to first send. The three unassigned campaigns are all scheduled and none has an owner to stop them.

Reports: performance

Against target, with the gap named.

Against target42.8%open rate on the best campaign

Best marketerAutumn renewal42.8% opened

WorstAutumn renewal1.9% unsubscribed

Trend+11 ptson last quarter

Against target by channel

100 is the target

Autumn renewal42.8

Trial nurture33.9

Product update28.1

Webinar invite21.6

Winback18.8

The best and the worst row are the same campaign. Opens went up and the list went down, and a single headline rate hides that completely.

What moved

Rule-derived

Open rate up 11 pointsThe tested subject line beat both alternativesUp

Unsubscribes at 1.9%Four times normal, on the same campaignDown

5,900 contacts hit three times in eleven daysNo frequency cap is setWatch

Threshold rules over the figures above, not a model. Each line names the number it came from so you can disagree with it.

Reports: exceptions

What breached, what is late and what nobody owns.

Breached80unsubscribes on one campaign

At risk5,900inside 48 hours

Unassigned3no owner at all

Reopened0closed then raised again

Exceptions

80 open

Campaign	Problem	Contacts	Cost	Preventable
Autumn renewal	Sent to 3,100 who already renewed	3,100	80 unsubscribes	Yes, an exclusion
Autumn renewal	Third email in 11 days	5,900	Same 80	Yes, a cap
Product update	Full list, no segmentation	24,800	31 unsubscribes	Yes, a segment
Webinar invite	Full list, no segmentation	24,800	24 unsubscribes	Yes, a segment
Winback post	4,120 contacts never included	4,120	Reach not used	Yes, a segment

Every row is preventable by a control this system already has and nobody turned on. The last one is the opposite failure: people who hear nothing at all.

Eighty unsubscribes, fifty-eight of them customers of three years or more. Two switches on the new-campaign screen would have removed 5,900 people from the send and almost all of the damage with them.

People

Who has an account, and what they own.

People

18 accounts

Name	Role	Brand	Last active	Status
Hanne Iversen	Administrator	Marketing	6 minutes ago	Active
Marta Reyes	Campaign manager	Lifecycle	35 minutes ago	Active
Joran Bakke	Marketer	Acquisition	1 hour ago	Active
Celine Aubert	Marketer	Retention	3 hours ago	Active
Tobias Lund	Marketer	Events	Yesterday	Active
Priya Ellison	Data protection	Legal	2 days ago	Active

Deactivating somebody prompts you to reassign what they own rather than doing it silently, because an unowned record is how a register rots.

Roles

What each role is scoped to and what it may approve.

Roles

5 roles

Role	Type	People	Scope	Can approve to	Status
Administrator	Built in	1	Everything	No limit	Active
Campaign manager	Built in	2	All campaigns	Approves sends	Active
Marketer	Built in	9	Own campaigns	Cannot send	Active
Data protection	Custom	2	Consent and suppression	Can stop any send	Active
Read only	Custom	4	Read all	-	Active

A role carries a scope and an approval ceiling. Both are enforced on the server; hiding a button is not access control.

Recent changes

Data protection given a stop on any live sendHanne Iversen · 3 JuneLogged

Marketers can no longer release their own first batchHanne Iversen · 3 JuneLogged

Role changes sit in the same audit log as record changes, because both change what somebody can do.

Why custom roles exist

A marketer builds and tests but cannot be the one who presses send
A campaign manager approves the first batch and can hold the rest
Data protection can stop a live campaign without asking anybody
Nobody can edit a suppression list, only add to it

Permissions

Every capability against every role, as enforced.

What each role can do

10 capabilities

Capability	Administrator	Manager	Marketer	Data protection	Read only
Build a campaign	Yes	Yes	Yes	No	No
Build a segment	Yes	Yes	Yes	Yes	No
Send a test	Yes	Yes	Yes	No	No
Release the first batch	Yes	Yes	No	No	No
Hold a live campaign	Yes	Yes	Own	Yes	No
Stop a live campaign	Yes	Yes	No	Yes	No
Set the frequency cap	Yes	Yes	No	Yes	No
Add to suppression	Yes	Yes	Yes	Yes	No
Remove from suppression	No	No	No	No	No
Export contact data	Yes	No	No	Yes	No

Enforced per query rather than by hiding controls. A scoped "yes" means the row exists but is limited, which is why some actions need a second person.

Nothing removes a contact from suppression, for anybody. A person who asked to be left alone stays left alone, and the only way back is for them to opt in again.

Notifications

What the system sends, to whom, and when.

What is sent

Unsubscribe rate above 1%Holds the campaign and tells the managerOn

Segment overlaps a recent sendAt build time, before schedulingOn

First batch awaiting approvalTo the campaign managerOn

Campaign scheduled with no ownerTo the manager, dailyOn

Consent record changedTo data protectionOn

Every message names the record it is about, so a reply lands somewhere.

Quiet hours

No routine notifications20:00 to 07:00, local to the personOn

Urgent overridesA live campaign held automatically, at any hourOn

Weekly digestMonday 08:00On

A notification that waits until morning is one that did not work, so the override is explicit rather than a setting nobody finds.

Notification rules are yours to set. What is here is a starting point drawn from how the queue actually behaves.

Integrations

Connected systems and their sync history.

Connected

4 connected

Email platformSending, opens, clicks and bouncesConnected

CRMContacts, consent and what they boughtConnected

Web analyticsSessions and conversions by campaign tagConnected

Print partnerPostal batches and returnsConnected

Data warehouseAttribution across all touchesNot connected

What breaks without them

Without the CRM the exclusion for people who already renewed cannot run, which is exactly the control the autumn campaign needed
Without analytics attribution collapses to last touch and stays there
Without the warehouse only one rule can be shown rather than three
Sending still works; the guardrails are what go missing first

Sync history

Last 24 hours

When	Source	Records	Result
Continuous	Email platform	4,200 events	Live
06:00 today	CRM	24,800 contacts	3,100 renewal flags updated
06:00 today	Web analytics	18,400 sessions	Current
2 Sep 18:00	Print partner	4,200 posted	62 returned

A sync that half-succeeded is listed as such rather than as a failure, because the records that did land still changed something.

Import and export

Bulk load and extract, with per-row validation.

Import

WhatIf a row already exists

Validated before anything is written. A file with one bad row imports the rest and tells you which line failed and why.

Export

WhatFormat

RangeEverythingThis yearFiltered view

Recent imports and exports

Last 30 days

When	What	Rows	Result	By
5 Sep 10:20	Import	1,840	1,802 loaded, 38 suppressed	Marta Reyes
1 Sep 17:00	Export	24,800	Complete	Priya Ellison
24 Aug 09:40	Import	620	Complete	Joran Bakke

Every import is kept with its source file, so a bad load can be traced rather than argued about.