The client is a developer of luxury real estate in St. Petersburg. Serenity Agency had to bring appeals for four objects: a residential complex, a club house, an apartment complex and a cottage village.
According to the contract, we cannot talk about the number of applications submitted.
The whole work was divided into 4 large parallel stages. To start bringing applications faster, we immediately launched contextual and targeted advertising. In parallel, they began to analyze them and develop a detailed promotion strategy.
We talk about all the stages in detail.
1. Content strategy
At the initial stage, we worked out a promotion strategy in social networks: using content and targeted advertising.
Analyzed site performance
We collected data from Yandex.Metrica for the year for all objects and found out that the promotion strategy at that time consisted in attracting the maximum number of initial requests, but not always targeted ones.
Studied the developer’s Facebook and Instagram
The accounts did not have a uniform design style, and the posts were not promoted.
Studied the market
We monitored the business media of Russia and St. Petersburg. We assessed how popular real estate is as a type of investment. We looked at what factors buyers of apartments in elite complexes pay attention to.
Analyzed competitors
We studied how competitors communicate with the audience on social networks. They were divided into developers, luxury residential complexes and apart-hotels:
- Apart-hotels try to maintain close communication. They constantly publish content on social networks, talk about the advantages of the format, share the stories of their residents.
- LCDs and club houses list the main advantages and benefits. But there are no descriptions, the reader does not form a picture in his head. Do not use social networks and other communication channels.
- Most developers communicate with the premium segment audience as with the economy class. For example, use the words: “benefit” and “discounts”.
Created audience portraits
For each of the complex, 4 portraits of the main buyers were compiled.
For each of these portraits, we compiled a Content-map — a chain of posts leading to a purchase:
2. Targeted advertising
In parallel with the content strategy, we launched the context and target. In three days, we launched targeted advertising for all four projects in order to quickly start receiving applications.
Emphasis on audiences
First, we used a database of audiences that already interacted with the client. Then, using the audience portraits drawn up at the strategy stage, we thought about how to find the target audience of each object:
- Elite real estate
Campaigns were set up for family people with interests in business, expensive cars and yachts. We also used geotargeting for class A business centers. - Business class
We used an audience of businessmen with interests in suburban real estate and an active lifestyle. For apartments, advertising campaigns were set up for an audience of investors and active young people with high incomes.
Create adDisable adsCryptoHow to start a beginner in crypto #2. What is ICO, IDO, INO, IGO and moreToday I will tell you what ICO, IEO, IDO, IGO and more are and how ICO, IEO, IDO, IGO differ from each other, and I will also tell you how to participate in all this, fly!
Removed non-target audience
In the real estate segment often comes across an audience not only of those who are going to buy it, but also the audience of the builders themselves. We removed them from the audiences and constantly looked for new solutions to show ads only to the target audience.
Optimized advertising campaigns
Updated creatives and USPs every 2–3 months so that users do not get bored with them, and the click-through rate of advertising offers and brand recall do not fall. We also used different ad formats: canvases, carousels, posts with buttons and stories.
We used lead forms
At the beginning of the work, they led the audience to the site. Users spent a lot of time there, but did not convert into an application. Therefore, we began to use lead forms – with them you can leave a request without leaving the social network.
Every sixth high-quality application came from lead forms.
Refused auto-complete
We tried to run auto-complete lead forms on Facebook, but there were a lot of low-quality or empty applications, so the filling was done manually.
There are no manual lead forms on Instagram, so we asked additional questions that filtered out non-targeted applications. For example: “Are you interested in apartments for yourself or for business?”.
3. Contextual advertising
With the help of contextual advertising, we closed the high demand of a potential audience, and then caught up with the people who were on the client’s website using remarketing.
Examined the current location
Collected geolocation-referenced queries for each of the four client objects in the search results. For example, apartments are located on Vasilevsky Island, so we use semantics indicating the location: “apartments on Vasilyevsky”, “apartments on Vasilyevsky”.
In addition, along with the ad on the search, we show a banner. So we occupy two advertising spaces and solve two problems: we close the hot demand and use the media format.
We tracked applications
We set up call tracking and tracking all forms on the site in order to record calls from the site and advertising campaigns. This made it possible to determine the sources of phone calls and optimize the budget.
Weeded out irrelevant queries
After analyzing the statistics and reports, we saw which search queries do not bring results for main and multi-channel conversions. Turned them off. Thanks to this, we spent our budget more efficiently by directing it to conversion channels and queries and constantly reducing the cost of going to the site.
Weeded out irrelevant sites
We paid attention not only to requests, but also to sites. Disable low-converting sites by checking them through multi-channel sequences. We spent the budget only on sites that brought results.
Worked with
responsive ads In addition to regular expanded ads on Google. Responsive search ads were added to ads. Their format allows you to specify up to 30 titles and four descriptions.
Weeded out irrelevant queries
The system analyzes user behavior and shows the most relevant heading with the USP, which, according to Google, will lead to a conversion. So we provided additional personalization of the ad.
Placed on third-party sites
Placed text and graphic blocks with ads on St. Petersburg real estate portals. For 3 weeks of work, we received 10 high-quality appeals.
Used autotargeting in search and YAN
Created separate campaigns with autotargeting. With it, ads in search and advertising networks can be shown without keywords. So we expanded low-cost low-frequency search queries. In YAN, the cost of a click was reduced by 26%, and the cost of an application by 13%.
We used look-a-like
We unloaded the contact database from call tracking, created a similar segment for advertising and launched ads for it. As a result, we received additional coverage and a new base of potential customers.
Launched ads on YouTube
Showed videos on YouTube to those who were on the site, but did not leave an application. The videos are aimed at different audiences. For example, for young people and parents – a video with a student, and for audiences 45+, a video with older people. This is how we communicated with clients from all sides, taking into account their portraits and needs.
Adjusted rates
When buying real estate, the decision-making window comes up to several months, so it is important to “warm up” the audience. From the analytics systems, we identified an audience that is interested in buying apartments and made adjustments for it – so these users saw our ads more often.
Weeded out irrelevant requests
We set up “search” remarketing and adjusted bids for those who visited the site but did not send an application – this is how “hot” users saw our ad at the top positions in extended formats.
DOOH-advertising
Made banners for digital outdoor advertising. Video screens were chosen in public places of those districts where the target audience lives or works.
4. Analytics
We improved analytics and set up detailed tracking of all advertising campaigns in order to see the sources of sales and quickly optimize the promotion strategy.
Improved the old analytics system
At the start of work, the company had poorly configured goals and analytics counters on each object. We reconfigured them and connected to the client’s call tracking so as not to lose all call statistics.
We received data from Yandex.Metrica, Google Analytics, Top.mail.ru and Calltouch call tracking and transferred them into a single report in Power BI. This is how we tracked conversion rates, dynamics, and the number of quality leads.
We tracked multi-channel conversions
The bulk of requests from contextual advertising came from branded queries. But if you leave only them, the effectiveness of advertising will decrease. After all, before entering this brand query, the user interacted with other channels. Multi-channel conversions helped track which ones.
We listened to the call center
We listened to how the client’s employees handle calls and objections from potential buyers in order to improve the quality of the call center.
We ourselves determined the quality and quantity of applications. For example, if the caller expected to buy an apartment for four million, and they cost six, then the application was considered to be of poor quality.
End-to-end analytics in manual mode
In manual mode, we tracked conversions to the application and initial requests. The client sent us closed order statistics from his CRM system and we compared them with his call tracking data. So we saw from which channel high-quality applications come from, and optimized the work with advertising tools.
We tested different promotion channels
We constantly tested new advertising channels. We left only those with which we get good results. For example, they posted a client’s article on Yandex.Zen and launched advertising campaigns in myTarget, but received a small number of conversions, so they abandoned these promotion channels.
Tips for those who promote developers
- Start promotion in parallel with the strategy. So the strategy will be supplemented with insights from the experience of promotion, and the business will not stop getting customers or start getting them faster.
- Sell lifestyle. Many of our USPs have focused on the lifestyle buyers will experience with their property.
- Work out the customer journey to purchase – determine which benefits are important to demonstrate to a particular segment, and determine the degree of importance of each benefit. Show the segment ads or banners in that order.
- Filter non-target audience in advertising. In the real estate segment, there is often an audience not only of those who are going to buy it, but also the audience of the builders themselves.
- Use lead forms, but don’t autocomplete. With us, every sixth high-quality application came from lead forms – but only on the condition that there were fields that needed to be filled in manually.
- Use automation tools in contextual advertising. Google selected for us those headlines that resonated the most with certain segments.
- If end-to-end analytics cannot be configured, do it manually. We compared statistics from the client’s CRM with leads and determined which channels bring high-quality leads.