How we got burned before we reached $26,000. Case targeting for an online clothing store
Hi, I’m Shukhrat, founder of Hope group social media promotion agency. In this article I will tell you how we got burned scaling targeting on Instagram for a clothing store @aim_clo , but still found a way and techniques to reach $26000 per month from targeting ads.
First of all, it should be noted that this success is the result of joint work of our team and the owner of the business, and thanks to the fact that the brand has established work with bloggers.
Period of work: May-July 2020
Our baseline: In April, we were able to reach $15,000 per month in turnover from Target
New goal: Increase turnover by multiples, using targeting ads on Instagram.
Well, what?! The month of May — we failed.
The turnover from targeted advertising 17000 dollars a month. The cost per sale is $14.
What did we run?
20% — retargeting,
80% — new audiences (we excluded audiences who had already made purchases and those who had visited the site).
We added more interests. Audiences that came in well were:
- photo-related (food photos, photo art, etc.)
- lookalike audiences also brought in good results, viz.
- lookalike by direct messages
- Lookalike on saved Instagram posts in 60 days
The narrow interests went badly:
- hoodies, sweaters, sweaters.
- T-shirts, blouses, dresses
Two sends worked:
- Discounts of up to 30% on spring collection
- Emotional message
Our failure is the month of June. What we recommend you learn:
1) Not all lookalike works well. Training and replenishing the lookalike audience happens automatically every 7 days. But in our case it worked for the worst — sales started to get expensive and decrease in quantity.
2) By creating companies on conversion and selecting a “Buy” goal, it takes longer to train these companies because the frequency of “Buy” actions is less than other actions and takes more time.Our companies on new audiences took longer to train and more budget was spent before the campaign trained well.
3) Launch the summer collection early, not in the summer) We didn’t launch the summer collection until late May/early June when it was already +20–30 and many people had already bought their clothes. We couldn’t influence that — the brand had late photos and the collection itself out.
Results of June:
- turnover of $13,000 per month
- The cost of one sale is $17
$26,000 in turnover in a month through Target in July
What did we do to achieve this result?
1) We analyzed the postings from bloggers, which the brand carried out for 3–4 months. From the topics of the blogs made a table of interests that we can test. By testing the interests, we got a number of successful audiences that we didn’t expect ourselves:
2) We launched a lookalike on conversion value. This audience includes information on purchases made on the site over the past 2 months, takes into account the average check, maximum check and minimum check, and builds an audience based on the person’s ability to make purchases for that amount, which is included in these criteria. It also takes into account their geo, mail, phone, interests and behavior. One of the most accurate audiences Facebook builds through artificial intelligence.
3) Working with the targeting, we determined the following indicators:
- conversion to purchases from visits to the site is 3.6%.
- We need a minimum of 4-touch through the targeting for a user to visit the site
With similar statistics, you can shape your strategy and advertising settings in detail.
4) We’ve launched a number of video formats with feedback from bloggers and users. Here’s one example:
These works helped to achieve the following result:
Average cost of 1 sale $14
Average check $ 90
Sales for the month 348 pcs.
You should understand that scaling is a constant test and search for working hypotheses, tools and methods to expand sales. And having done the same thing that we have done, you may not get this result, because your project requires other methods and hypotheses.