Traffic mix and seasonality: compare like with like

Maya

Version 5.0.0 went out on 6 July, and week-2 retention of new users has slid from 37.0% in June to 30.3% in July. The team wants the old home screen back.

You

Before we blame the screen: what else was different in July?

Alex Tutor

Good first question. The metric is identical in both months: new users, days 8–14, every cohort complete. What can differ is the crowd inside it. And July is when Halves users go on holiday together.

Here are the weekly cohorts, split by channel and by the type of group each person joined first. Start with the season: how much of July arrived through trip groups?

Alex Tutor

Dataset · 78 rows

Weekly sign-up cohorts, June and July

New Halves users, weeks of 8 June to 27 July 2026. Version 5.0.0 was released on 6 July. Every cohort has its full days 8–14.

Download CSV

Columns
signup_week: Monday of the sign-up week
app_version: Version installed: 4.9.0 in June, 5.0.0 in July
channel: invite, app_store_search, paid_social or partner_promo
first_group_type: Type of the first group the user joined: flat, couple or trip
new_users: New users in this cell
active_days_8_14: Of them, opened the app on at least one of days 8–14

Data

signup_week	app_version	channel	first_group_type	new_users	active_days_8_14
2026-06-08	4.9.0	invite	flat	106	44
2026-06-08	4.9.0	invite	couple	64	31
2026-06-08	4.9.0	invite	trip	43	7
2026-06-08	4.9.0	app_store_search	flat	79	28
2026-06-08	4.9.0	app_store_search	couple	52	16

Show 73 more rows

2026-06-08	4.9.0	app_store_search	trip	32	5
2026-06-08	4.9.0	paid_social	flat	61	12
2026-06-08	4.9.0	paid_social	couple	37	15
2026-06-08	4.9.0	paid_social	trip	26	2
2026-06-15	4.9.0	invite	flat	106	44
2026-06-15	4.9.0	invite	couple	69	29
2026-06-15	4.9.0	invite	trip	46	5
2026-06-15	4.9.0	app_store_search	flat	86	39
2026-06-15	4.9.0	app_store_search	couple	49	23
2026-06-15	4.9.0	app_store_search	trip	34	9
2026-06-15	4.9.0	paid_social	flat	65	26
2026-06-15	4.9.0	paid_social	couple	38	11
2026-06-15	4.9.0	paid_social	trip	25	5
2026-06-22	4.9.0	invite	flat	105	57
2026-06-22	4.9.0	invite	couple	66	28
2026-06-22	4.9.0	invite	trip	45	9
2026-06-22	4.9.0	app_store_search	flat	82	37
2026-06-22	4.9.0	app_store_search	couple	52	20
2026-06-22	4.9.0	app_store_search	trip	32	6
2026-06-22	4.9.0	paid_social	flat	62	20
2026-06-22	4.9.0	paid_social	couple	37	12
2026-06-22	4.9.0	paid_social	trip	24	3
2026-06-29	4.9.0	invite	flat	106	60
2026-06-29	4.9.0	invite	couple	66	29
2026-06-29	4.9.0	invite	trip	42	8
2026-06-29	4.9.0	app_store_search	flat	85	46
2026-06-29	4.9.0	app_store_search	couple	50	21
2026-06-29	4.9.0	app_store_search	trip	33	5
2026-06-29	4.9.0	paid_social	flat	65	23
2026-06-29	4.9.0	paid_social	couple	38	14
2026-06-29	4.9.0	paid_social	trip	25	3
2026-07-06	5.0.0	invite	flat	89	45
2026-07-06	5.0.0	invite	couple	60	27
2026-07-06	5.0.0	invite	trip	100	18
2026-07-06	5.0.0	app_store_search	flat	65	28
2026-07-06	5.0.0	app_store_search	couple	40	17
2026-07-06	5.0.0	app_store_search	trip	71	16
2026-07-06	5.0.0	paid_social	flat	47	17
2026-07-06	5.0.0	paid_social	couple	29	9
2026-07-06	5.0.0	paid_social	trip	52	10
2026-07-13	5.0.0	invite	flat	84	38
2026-07-13	5.0.0	invite	couple	56	24
2026-07-13	5.0.0	invite	trip	93	29
2026-07-13	5.0.0	app_store_search	flat	60	26
2026-07-13	5.0.0	app_store_search	couple	43	17
2026-07-13	5.0.0	app_store_search	trip	68	11
2026-07-13	5.0.0	paid_social	flat	43	15
2026-07-13	5.0.0	paid_social	couple	29	12
2026-07-13	5.0.0	paid_social	trip	52	8
2026-07-13	5.0.0	partner_promo	flat	198	48
2026-07-13	5.0.0	partner_promo	couple	81	15
2026-07-13	5.0.0	partner_promo	trip	114	13
2026-07-20	5.0.0	invite	flat	88	53
2026-07-20	5.0.0	invite	couple	60	25
2026-07-20	5.0.0	invite	trip	92	17
2026-07-20	5.0.0	app_store_search	flat	62	25
2026-07-20	5.0.0	app_store_search	couple	40	15
2026-07-20	5.0.0	app_store_search	trip	68	11
2026-07-20	5.0.0	paid_social	flat	43	17
2026-07-20	5.0.0	paid_social	couple	31	17
2026-07-20	5.0.0	paid_social	trip	52	9
2026-07-20	5.0.0	partner_promo	flat	204	52
2026-07-20	5.0.0	partner_promo	couple	76	16
2026-07-20	5.0.0	partner_promo	trip	115	9
2026-07-27	5.0.0	invite	flat	87	41
2026-07-27	5.0.0	invite	couple	60	31
2026-07-27	5.0.0	invite	trip	98	19
2026-07-27	5.0.0	app_store_search	flat	62	25
2026-07-27	5.0.0	app_store_search	couple	43	21
2026-07-27	5.0.0	app_store_search	trip	72	13
2026-07-27	5.0.0	paid_social	flat	46	13
2026-07-27	5.0.0	paid_social	couple	31	13
2026-07-27	5.0.0	paid_social	trip	48	9

Alex Tutor

Question 1

What share of July (5.0.0) new users had trip as their first group type? Percent, two decimals. (In June it was 20.02%.)

Your answer

Units: %

Alex Tutor

Almost double June. A trip group comes with an end date: the holiday is over, the last dinner is split, everyone settles up. Days 8–14 often fall after the flight home, so a trip user who stays away may be perfectly satisfied.

You

So even if 5.0.0 changed nothing, more trips would drag the July average down.

Alex Tutor

Yes. The traffic mix moved, and a mix shift moves the total while no segment changes. In June trip users kept 16.5% in days 8–14 and flats 43.3%. Trade flats for trips and the average sinks by itself.

Theo

There’s a second thing, and it’s mine. For two weeks from 13 July a bank’s student card paid holders a $5 bonus for signing up to Halves and creating a group. Those are the partner_promo rows.

Alex Tutor

Then those are incentivized sign-ups: people who joined for the reward. Some really had rent to split; many made a group only to meet the condition and never added a bill. How big was the wave?

Alex Tutor

Question 2

What share of the new users who signed up on 5.0.0 (July) came through partner_promo? Percent, two decimals.

Your answer

Units: %

Alex Tutor

Over a quarter of July, and only 19.4% of them were active in days 8–14, below every regular channel in June (29–41%). June had no such channel, so they have no counterpart: set those rows aside first. July without them: 34.24%, still under June’s 36.99%.

Maya

Still lower than June, though.

Alex Tutor

Because the extra trips are still in there. Step two: compare inside each group type, trips against trips, flats against flats. Try flats.

Alex Tutor

Question 3

Without the partner_promo rows, what share of July flat users was active in days 8–14? Percent, two decimals. (June flats: 43.25%.)

Your answer

Units: %

Alex Tutor

Flats, couples and trips each did slightly better in July. For one summary number, keep June’s mix and swap in July’s rates: weight each channel × group type cell of July by that cell’s share of June.

Alex Tutor

r^{\text{July}}_{\text{adj}} = \sum_{c} w^{\text{June}}_{c}\, r^{\text{July}}_{c}

July’s retention at June’s mix: w_c is cell c’s share of June’s new users, r_c is July’s retention in that cell, promo rows left out. Here 38.89%, against 36.99% in June.

You

So at June’s traffic, July would be 1.9 points better, not 6.7 worse.

Maya

Then 5.0.0 improved retention. Great, I’ll say that.

Alex Tutor

Slow down. It adjusts only for what we split by. July also brought holidays and heat, and some cells hold little more than a hundred users. It supports “no sign 5.0.0 hurt retention”, not “5.0.0 added 1.9 points”.

Alex Tutor

Question 4

Maya has to answer the team today. Which statement can she defend with this table?

Choose one answerThe new home screen cost 6.7 points of week-2 retention; roll it backThe new home screen added 1.9 points of week-2 retentionJuly’s drop comes from who signed up (more trips, the bank promo); compared like for like, nothing points to harm from 5.0.0Without an A/B test the table says nothing about the release

You

So a mix-adjusted number is a fair comparison?

Alex Tutor

Fairer, not fair. It removes the differences you measured: channel and group type. Anything you didn't split by, like the weather or the school holidays, stays inside.

Theo

The bank wants to rerun the offer when the university term starts. What should I do differently?

Alex Tutor

Give the offer its own sign-up code, so every bonus user carries partner_promo from day one. Judge the deal by users still active in week 2, not by sign-ups: 788 joined, 153 stayed. And agree the dates with Maya, so a release and a promo don’t start in the same fortnight.

Alex Tutor

Question 5

July’s rate re-weighted to June’s mix of channel × group type is 38.89%. Which of these can still be hiding inside that number? Select all that apply.

Choose all that applySchool holidays and summer weatherChance: each July cell holds only about 120–350 usersA rival app’s summer campaign that pulled some users awayThe larger share of trip groups in JulyThe incentivized sign-ups from the bank promo

Maya

Fine, no rollback. But I need to explain it to the team without a lecture.

Alex Tutor

Then keep it to three moves: who came in July and how they differ, what the like-for-like numbers show, and what we can and can’t say.

You

With a number behind each move, so nobody has to take our word for it.

Alex Tutor

Question 6Short answer · AI-checked task

Maya asks in the team channel: “Do we roll back 5.0.0?” Reply in 3–5 sentences: give the answer first, then the numbers behind it, then what the data can’t tell us.

Write your answer and get a score with feedback from our AI reviewer.

Log in to get AI feedback

Alex Tutor

That’s the lesson. You answered every task — nicely done.

Next lesson: Users who do X retain better