Flower Shop Case Study - Parizhanka
In 2017, our client was the florist shop ‘Parizhanka’ 💐. The company’s strengths include the premium quality of their bouquets, the aesthetic approach to their arrangements, and fast delivery. But the floristry business has its own unique characteristics as it’s full of great deals, and customers make decisions rather quickly. In order to be visible to the target audience and to be in demand, we needed an integrated approach to promotion and careful adjustment to advertising.
From the first advertising campaigns to comprehensive internet marketing – we’ll explain how we began with the project and what it’s become today.
- Client’s website – flowerfor.me
- Advertising account – Google Ads
- SEO optimization
- Targeted advertising on social networks (Facebook Ads, VK Target, My Target)
The client contacted us with the task of configuring Google Ads. After filling out the brief, we had a good idea of the target audience as well as the supply and demand in the floristry market. To reach more customers, after about 6 months of work, we added advertising on Yandex.Direct. With the growing competition in the florist market, we had to rethink the advertising approach and we decided to launch a comprehensive promotion, including SEO promotion and social media targeting.
Initially, we were faced with the task of getting as many conversions (leads) as possible at the price of 350 rubles from the Google Ads advertising system.
Doing this required:
- Setting up turnkey advertising campaigns on Google Ads.
- Setting up full conversion tracking through Google Analytics + Google Tag Manager.
- Configuring call conversion tracking using a link tracking system and Google Analytics.
In the process of scaling the work, the range of objectives expanded.
Additionally, new goals also presented themselves:
- Reconfiguring and permanently optimizing the Yandex.Direct advertising account to achieve a conversion cost of less than 350 rubles.
- Carrying out a full SEO audit of the site in order to start a full SEO promotion, to raise the site above the top 10 and secure consistently high positions 🔝.
- Setting up and constantly optimizing targeting on social media. Facebook Ads, VK Ads, MyTarget were selected as the most popular social media networks in the Russian Federation.
Plan of Action
One of the most important decisions involved setting conversions on the client’s site.
We decided to use any customer request (lead) as a conversion and ended up with 5 methods:
- the chatbox on the site (either started by the customer or with an accepted invitation to the chat from an operator)
- a call to a substitute number (dynamic call tracking)
- purchases through the site’s cart (visiting the thank you page after placing an order)
- one-click ordering on the site (the client leaves only their phone number and clicks ‘order’)
- a call through a search ad (snippet) without going over to the site itself.
For tracking we used:
- Callibri for tracking orders through the chat
- Callibri Call-tracking to record calls
- Thank you for your purchase pages through the shopping cart or the quick order button
- Google Ads: customized conversion in the search snippet for calls through an ad
We made a number of decisions to make advertising more profitable:
- We chose not to remarketing in Google Ads and retargeting in Yandex.Direct, since flowers are an FMCG (fast-moving consumer good). Practice and measurable results have shown us that there is no point in ‘catching up’ with a failed customer when they’ve already bought a bouquet. All performance marketing in the floristry market is based on the idea that flowers are an instant commodity.
- 70% of all conversions on the site were on mobile devices 📱. With that in mind, it became necessary to optimize advertising for mobile devices in order to make conversions more profitable.
- To attract customers to promo offers, we decided to connect targeting in social networks in conjunction with the classic method of maintaining accounts on social media (SMA).
Google Ads – Search Ads
Advertising campaigns in Google Ads were categorized by keyword type. Separate campaigns were created for each category of flowers (tulips, roses, sweet bouquets, etc.), as well as a general campaign with groups for various services: ‘delivery’, ‘ordering bouquets’, ‘buy flowers’ and so on. Ads that match the keywords, landing page, and the group of advertised products as closely as possible were launched into the display. And all ads lead to their respective landing pages.
With all that, if a person wants to order 101 roses, buy a bouquet in the online store, order a delivery of roses or other flowers, they only need to complete 3 steps: enter a request, go to the webpage they like, and place an order. The most effective ads were those with promotional offers, for example – 33 roses for 999 rubles.
thanks to optimization and a proprietary bidding strategy, in the first few weeks the ad campaigns brought in the required number of conversions and trained enough to incorporate the smart Target Cost Per Conversion strategy, with which we began to receive conversions worth an average of 280 rubles.
Also, the first leads gave us a better understanding of the audience for subsequent optimization.
Analytics showed us that about 60% of customers make their order by phone 📞 using the number on the site or from the search snippet. In most cases, the purchase is instant and is made either spontaneously or at the last minute.
Google Ads – Display Network
We Configured advertising in GDN for look-alike audiences (a customer-like audience created by Google’s machine learning), which brought in an additional 1-2 conversions per day, but the cost of these conversions was quite high. There were several reasons for this:
- In the CIS market, it is necessary to work very carefully with GDN because of the large number of low-quality sites.
- As mentioned before, flowers are a quick demand product, therefore targeting at a sufficiently high cost per click does not allow campaigns on GDN to be profitable.
- GEO is too narrow to work with strategies based on Google machine learning. Since the client only operates in the city of Kazan, it didn’t make sense to target the entire Russian Federation.
We decided to discontinue advertising campaigns in GDN after testing it for 3-5 weeks.
Google Ads – Product Advertising
Setting up Google Shopping Product Ads required developing a product feed. A programmer was brought in to create it. We compiled a detailed TK with all possible parameters and received an updated feed from the programmer, which was uploaded through the Google Merchant Center. We linked Google Merchant Center with Google Ads and set up a Google Shopping ad campaign.
- Launched the ad campaign with one general goal per click.
- After 2 weeks, we received enough clicks and data that we used for further adjustments.
- After another month of work on the campaign, we divided all products into 10 groups depending on the price per conversion. We set the maximum CPC on the most converting products, a few rubles less on the next category down, and so on.
We are still working on optimizing product ads today. We exclude non-targeted traffic and sometimes change groups for products when they become more purchased, and adjust bids and audiences. Thanks to the right approach to product advertising, Google Shopping managed to achieve a price per conversion of 299 rubles (as of September 2019).
Yandex.Direct – Search Advertising
At first, the price of conversion on Yandex.Direct exceeded the limit set by the client.
In turn, We used broad keyword matching, exact query matching (which led to the expansion of the core to 4000 keywords), and left only the keys that generated the most conversion.
As such, we got the following results:
- YAN – conversions from 800 to 1100 roubles.
- Search advertising – conversions from 500 to 800 roubles.
The most likely reason for such high rates is too much competition in the floristry market on Yandex.Direct.
In 2020, we tested Yandex automatic strategies, trained an advertising campaign, and we succeeded! We began to receive conversions for 350-400 rubles 💪. The client was completely satisfied with this cost since by this time, the flower business had changed a bit, with prices for buying flowers from suppliers increasing, new competitors appearing and the average cost of a bouquet had to be raised.
Keeping the conversion price of 350 rubles or less would lead to fewer conversions. The way to increase the number of conversions, and, accordingly, their cost, looked more optimal. However, all this is still within the price of 400 rubles per conversion.
As such, since 2020, the new maximum conversion price for Google Ads and Yandex.Direct is 400 rubles.
Yandex.Direct – YAN (Yandex Advertising Network)
We attempted to launch advertising in YAN with the client several times. We were constantly changing creatives, working with keywords, and filtering negative keywords. However, all this only helped us to achieve a result of 700-800 rubles per conversion. For the client, this cost of conversion was unacceptable.
In 2020 we tried to configure YAN through Yandex automatic strategies. The 4-week test showed a conversion cost of 650 rubles, which was still too expensive for a GEO in Kazan and for the floristry sector. However, this result was quite expected. With a high enough cost per click on display ads, the results can in no way be considered equal to search ads.
In the end, we decided to discontinue working with YAN.
Yandex.Direct – MKB or Banner in Search
This format worked well. Since it is not so popular among advertisers, we expected an adequate CPC and were proven right.
Media-contextual banners generate conversions for an average of 400 rubles. In addition, we can place a picture of the bouquet and the price of the product on the banner, which gives us an advantage over conventional search advertising.
However, after several months of work, the cost per click had risen. Many competitors had appeared in this ad format, and the conversion price has gone up to an average of 700 rubles so we decided to discontinue working with it. In addition, at the time of writing the case, automatic Yandex.Direct strategies for this ad format were not yet available.
To set up Yandex.Market product advertising, we also had to bring in a programmer. Based on the technical specification for Google Shopping, we made a product feed for Yandex.Market and then made the necessary edits. After we launched the ad, we were able to achieve 5-7 conversions per week at any given rate. However, the conversion price was about 250 rubles, which worked out great.
In addition, while working with Yandex.Market, we received warnings and bans, for both legitimate and far-fetched reasons, so we stopped this ad format and made edits either to the site or to the product feed.
As a result, at the end of 2019 placement in Yandex.Market had to be discontinued since the amount of effort spent to stay on the marketplace exceeded the benefits.
When we started working with a client, their store was already around the 10th position on Google search results, and in the 30s and 40s on Yandex, but they couldn’t rise any higher. Working with their previous contractor brought Parizhanka some results, but at some point, growth stopped.
Work on SEO was carried out by:
- Analyzing the site 🔎 and identified problems that prevented the site from growing dynamically.
- Putting together a new semantic core, which showed further directions for development.
- Relying on site speed and user-friendliness. To do this, we carried out work on technical optimization (they redesigned the structure, converted some of the “heavy” graphics into the webp format, applied adaptive images for different screens, and optimized the server, which was especially difficult to implement due to the closeness of the Bitrix system in terms of resource use).
- Updating content on the site (added new descriptions of categories and products, starting a blog).
- There was a lot more extra work we would talk about for hours…
Work on website SEO promotion has been underway for over a year. However, at the current stage, we have already managed to reach stable positions of 1-2 on Google and 5-10 on Yandex.
Targeting on Social Media
Social media plays a special role in the florist business. After all, it is here that someone chooses what they want with their eyes 😍. We’re working with an SMI specialist who maintains accounts far Parizhanka on Facebook, Instagram, VKontakte, and Odnoklassniki.
Targeted advertising was configured by us on all sites except for Odnoklassniki (this network was abandoned after 3 weeks of testing and due to low popularity, weak target settings, and lack of full-fledged analytics).
We implement video and graphic banners as creative content.
The most effective format for working on Facebook, Instagram, VK is a carousel of 3 banners.
Advertising was carried out on promotional pages and was displayed to interested audiences, which were selected as follows:
- We set up ads for ready-made audiences in Facebook Ads and VK Ads.
- Using Facebook Ads and VK Ads pixels, we put together an audience that made conversions and created a look-alike audience based on it.
- We added all contacts of existing clients from the client’s CRM to Facebook Ads. Based on this data, Facebook also created a full-fledged look-alike audience for targeting.
Also, in order to get a more complete look-alike audience, we configured the transmission of call conversion actions to Facebook Ads by using Callibri.
The result of this continuous long-term work was conversions of up to 300 rubles in VK Ads and Facebook Ads. We left these as one of the main advertising sources, due to their flexibility and low cost per conversion.
Currently, the work on comprehensive promotion continues. The average cost of conversion is:
- On Google Ads – 300-350 rubles.
- On Facebook Ads and VK Ads – 250-300 rubles.
Complex Internet marketing generates 35-55 conversions every day. This number depends on holidays, the season, events, and other local factors.
Working on SEO helped us to more than quadruple search traffic from Yandex and Google. From the 10th position in Google, we have moved up and remain at the top 1 and top 2 spots 🥇 . We are in the top 5-10 Yandex by target keywords. We have been holding such positions over the past year and a half. The overall site visibility is over 70%, and the share of free search traffic to the site is 25%.
Targeted advertising has proven to be an effective and inexpensive tool for getting additional conversions. And even if there were not as many of them as there were from direct search traffic, they cost about the same if you take into account the work of an SMM specialist.
Work on the Parizhanka project is a great example of scalability and complex advertising 🌱 🌷. After getting the first positive results, cooperation with the client didn’t stop there as we began to develop new directions for promotion. Simultaneous work on contextual advertising, SEO, and SMA let our client be more visible to buyers, be the first to offer them the best bouquets, and respond to market changes quickly and in real-time.
LLC “Parizhanka” thanks the advertising agency Web-Sci for our fruitful cooperation! We entrusted Internet advertising to the agency a few months ago and during this time we were convinced that our choice was correct.
Today, Web-Sci advertises our products in Google AdWords. A high-quality technical support, the opportunity to participate in private testing of new advertising products, high professional qualifications of employees are important characteristics of the Web-Sci agency for us.
We express our gratitude to the account manager Tula Ruslan for his professionalism, high qualification, responsibility and personal qualities. The agency uses the latest innovations in performance marketing to achieve the best results.